Woofer Tester Pro Speaker Tester User Guide

Transcription

1 Woofer Tester Pro Speaker Tester User Guide

2 Smith & Larson Audio PO Box 229 Savage, MD USA Phone

3 Table of Contents 1. INSTALLATION & SETUP GETTING TO KNOW YOUR WOOFER TESTER PRO INSTALLING THE SOFTWARE WTPRO.LIC LICENSE FILE SETTING UP THE HARDWARE SOFTWARE AND HARDWARE INSTALLATION NOTES HARDWARE CONNECTIONS NAVIGATION MENU BAR TESTS, SIGNAL CONTROL AND REAL-TIME METER WINDOW WT CONTROL WINDOW SETUP CONTROL WINDOW OVERLAY WINDOW T/S ENTRY WINDOW FFT AND OSCILLOSCOPE WINDOW IMPULSE AND WATERFALL CONTROL WINDOWS TIME ALIGN - AVERAGING MICROPHONES X-Y PLOT WINDOW VIEW MENU OPTIONS WOO FILES & SCREEN CONFIGURATION VIEWING, FORMATTING & CONVERTING RESULTS OPTIONS MENU TOOLS MENU PRINTING FILES/GRAPHS EXPORTING RESULTS HELP MENU CALIBRATION CALIBRATION, COMPONENT & LINEAR TESTING LOW POWER PORT CALIBRATION HIGH POWER PORT CALIBRATION THIELE SMALL DRIVER MEASUREMENT REVEALING THE THIELE-SMALL MODEL USING ELECTRICAL TESTS FINDING MASS AND SPRING CONSTANTS THAT SET FREE AIR RESONANCE FS DRIVER Q DRIVER FREQUENCY DEPENDENT INDUCTANCE (LE IS NOT SIMPLE INDUCTANCE) NON LINEAR SUSPENSION AND MOTOR CHARACTERISTICS THE TS TESTS (WHAT IS MEASURED) Q AND FS TEST (LAUNCH FROM Q,FS TEST BUTTON) VAS TEST (LAUNCH FROM VAS TEST BUTTON) Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 3

4 4.9. BOX TEST (LAUNCH FROM BOX TEST BUTTON) MAKING A SUITABLE TEST BENCH DRIVER BREAK-IN ARBITRARY IMPEDANCE PLOTS (ARB1 AND ARB2 BUTTON) DRIVER SIMULATION AND BOX DESIGN SIMULATION AND PHYSICAL BOX DESIGN FIRST PASS SIMULATION IN-BOX ELECTRICAL TESTING BOX ELECTRICAL TEST AUTO-ALIGN VBOX (ALIGN SIMULATED AND IN-BOX IMPEDANCE) SIMULATED VERSUS MEASURED IN-AIR RESPONSE ADVANCED THIELE SMALL TESTING: AC & DC COMPRESSION AC COMPRESSION TESTING DC COMPRESSION TESTING INTERACTIVE CROSSOVER DESIGN (ICD) A PHYSICAL VERSUS SIMULATED CROSSOVER DESIGN EXAMPLE CREATING AN ELECTRONIC CIRCUIT FILE CIRCUIT FILE SYNTAX ICD COMPONENTS AND DEVICES ICD AND XVR EDITOR AND SIMULATION CONTROL ICD DRIVER PROTECTION CIRCUIT ARBITRARY CROSSOVER DESIGN (XVR FILES) HARMONIC & SINAD DISTORTION MEASUREMENT TEST TOOLS AND CALCULATORS ALIGNING THE TESTER FOR MEASURING CAPACITORS ZOBEL CALCULATOR DEMONSTRATION FILES CONDUCTING A SIMPLE ELECTRICAL TEST THIELE-SMALL SIMULATOR & BOX DESIGN DEMONSTRATION DEMONSTRATION OF AN IN-AIR TEST DIGITAL SIGNAL PROCESSING TERMINOLOGY AND CONCEPTS NYQUIST SAMPLING THEOREM RECONSTRUCTION FILTERS UP-SAMPLING FILTERS Smith & Larson Audio Woofer Tester Pro User Guide v1.1

5 INSTALLATION & SETUP 1. INSTALLATION & SETUP 1.1. Getting to know your Tester Tester Front Panel The front of the test measurement box contains the line-in and line-out ports as well as the phantom powered microphone inputs as shown below: Line-in port Line-out port Power LED Phantom powered microphone inputs Tester Back Panel The back of the measurement box contains an auxiliary 5V power port (for when there is only one low power port), USB port, a high-power test port and a low-power test port: USB port Auxiliary 5V port for low power USB hosts (not required in most cases) Low-power test port High-power test port Note: The WT Pro uses about 160 from the USB port. Most USB ports will supply this directly. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 5

6 INSTALLATION & SETUP 1.2. Installing the Software 1. Insert the CD-ROM into your computer. If the installation process does not start automatically, run setup.exe from the CD-ROM. This will start the software install wizard: 2. Click Next. The next dialog box shows the license agreement. Please read it and select I agree if you concur and then click Continue to proceed with the installation. 3. Next, select an existing directory using Browse or create a new one by typing a name in the text box. By default, a WTPRO directory gets created on your C: Drive and you ll be asked if you want to create a new directory. Click OK to continue. 6 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

7 INSTALLATION & SETUP 4. The last step in the software installation shows you the latest readme.txt file and asks if you want to add a desktop icon and an entry in the Windows Start Program. Select the appropriate options and click Finish. The software installation is now complete WTPRO.LIC License File WTPRO.LIC (a software license and hardware key-file) must be copied to the installation directory to enable the WT-Pro hardware and software. This file will be important if you download software updates from the web Setting up the Hardware 1. To setup the hardware, connect the USB cable between the Tester and computer s USB port. Windows will indicate new hardware has been detected and the following dialog box pops open: 2. Select the Install from a list of specific location (Advanced) option and click Next Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 7

8 INSTALLATION & SETUP The following dialog box appears: 3. Click the Browse button In the file dialog box that opens, select the WTPRO directory that you created earlier and click OK 4. Click Next If you are using Window XP, a warning may appear about Windows Logo testing. This driver is compatible with all versions of Windows. Click Continue Anyway to continue the installation 8 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

9 INSTALLATION & SETUP 5. When file copying is complete, click Finish to complete the installation In Win XP You will see a system tray message that your new hardware is installed and the hardware setup is now complete. If you manually installed the drivers and did not already plug the Tester into your computers USB port, do so now. Windows will now detect the Tester Software and Hardware Installation Notes Software Installation The software can be installed by directly copying files from the CD, or by using the SETUP.EXE installation program. There are no Windows registry entries made and the only drivers required should already be available if you meet the following requirements. Windows Generic USB Driver Installation When the tester into your system you may find that the drivers are already installed and there is nothing to do. If not, Windows will identify that a 'USB Audio Codec' has been installed and the new hardware wizard will start. Following the wizard's instructions will eventually configure the Windows default driver for USB Audio Devices. This may require you to install drivers from the Windows installation CD. Note: Windows always assumes that any newly installed device will become the new default device. This will redirect your PC's default sound output to the tester. You will want to disable this by going into the control panel and changing the default sound device. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 9

10 INSTALLATION & SETUP Additional Steps for Windows Vista Users When Windows Vista installs the USB Audio Device driver, the initial default will be set for 1 channel 22 khz. This must be changed for the Tester to operate properly. The changes made here may revert back if the operating system is updated, or a new audio device is installed. 1. From the Start menu, open the Control Panel and select Hardware and Sound. The Tester will be identified as using an USB Audio Codec. 2. Click on the Recording tab and look for the Microphone that uses the USB Audio Codec. 3. Click on the microphone properties, and then select the Levels tab. 4. Next, click on the Advanced tab and change the settings to 2 channels, Hz sampling rate. Check the two Exclusive Mode check boxes giving the tester application control of the device. Click OK. 10 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

11 INSTALLATION & SETUP 5. Repeat Step 4 for the Playback tab (2 channel Hz sampling). 6. Go to the Enhancements Tab and make sure that the Disable all enhancements checkbox is selected. 7. Make sure that the Playback, Record and Tester sampling rates all match Hz is the recommended sampling rate for Vista and is the default setting on the Tester. Sampling rates are changed in the Options pull-down menu or from the Comp & Size tab in the Setup Control window. Software Installation Notes The software can be installed by directly copying files from the CD, or by using the SETUP.EXE installation program. There are no Windows registry entries made and the only drivers required should already be available. Windows Generic USB Driver Installation Notes When the tester is plugged into your system you may find that the drivers are already instaled and there is nothing to do. If not, Windows will identify that a 'USB Audio Codec' has been installed and the new hardware wizard will start. Following the wizard's instructions will configure the Windows default driver for USB Audio Devices. This may require you to install drivers from the Windows installation CD. Windows always assumes that any newly installed device will become the new default device. This will redirect your PC's default sound output to the tester. You will want to disable this by going into the Windows control panel and changing the default sound device. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 11

12 INSTALLATION & SETUP 1.6. Hardware Connections! WARNING! DO NOT CONNECT ANYTHING EXCEPT FOR A DRIVER TO THE LOW POWER TEST PORT Basic Thiele Small testing and parameter extraction does not require more than a few milli-amps of current, which is easily powered by the USB port. The high power port is a pass through port that senses the voltage and current at the load. The high power port is amplifier safe (ground referenced, non-bridged), but do not let the amplifier touch the low power port! Connecting a Driver The provided test leads with a banana jack at one end and alligator clips at the other will be used to connect drivers to the Woofer Tester. Calibration resistors (and shorts) are also used during the calibration routine. These will be connected directly to the Woofer Tester, or to the end of the test leads depending on the parameter being measured. Connecting lineout to an amplifier for breaking in a driver Before testing a driver, it is often desirable to break the driver in for an hour or two. Basically 'fresh' driver parameters are not the same as when the driver has been exercised for some time. This is accomplished, and with far less voice coil heating, by using a test tone frequency equal to the driver s free air resonance Fs. The line level RCA jack outputs can be connected to an amplifier to provide this signal. Frequency is controlled from the WT Control Window. Note: In normal operation, the Y-jacks are used to provide the line out-in signal loop back. Removing the Y-jack opens the signal chain allowing line-level devices like preamps to be tested. Note: Since the Woofer Tester must share PC resources, dropouts may occur if the operating system becomes overloaded. Using only software to control signal levels is therefore not advised. A pre-amplifier or an integrated amplifier with a volume control allows you to set the software level to maximum and then cut back the level manually. 12 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

13 NAVIGATION 2. NAVIGATION This section of the user guide gives an overview of the software features, explains how to view, print and export results and configure system settings. Now that you have connected up your Woofer Tester Pro and started the software, you will see the main window shown below. The program consists of an outer Frame window that then contains multiple child windows. Figure 2-I - Main Window Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 13

14 NAVIGATION 2.1. Menu Bar On the top of the screen is the menu bar with eight pull-down menus from which you can perform the following functions: Menu File View Options Tests Tools Results Window Help Functions Open and create WOO Files (described in the next section), export results to various file formats and capture graphics Open test windows, set graphing options and customize screen Configure sweep parameters, measurement frequency, drive level, signal mode, sample rate, buffer size, set microphone gains, button modes and select operational modes Calibrate and conduct measurements Set atmospheric conditions for tests, calculate effective size of non-circular speaker, design an air core inductor Export test results to the editor, format output, convert electrical results into effective box parameters, design a tank circuit Arrange windows on screen Help on using the Woofer Tester Pro 2.2. Tests, Signal Control and Real-time Meter Window Figure 2-II Tests, Signal Control and Real-time Meter Window Test and Data Selection Test Frequency and Signal Level The left hand side of the screen contains the Thiele Small Test buttons, the signal frequency and level controls and the real-time meter window as shown in the picture on the left. This window contains buttons for various tests and shows the current real-time frequency, measurements, levels and other parameters. Real-Time Meter 14 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

15 NAVIGATION Buttons with Multiple Click Regions Buttons can consist of multiple click regions. The first five buttons, Q/FS Test, Vas Test, Measure Box, Arb1 sweep and Arb2 sweep have two regions. Clicking the button on the left changes the data view in the main window. Clicking the button in the middle or on the right conducts the test: Click on the left side of the button to switch the data view X indicates current data view Click the button on the right or middle to conduct this test Clicking the Frequency Hz and Drive Level buttons on the left decreases the frequency and drive level while clicking on the right increases them. Clicking the buttons in the middle brings up a dialog box to enter a specific value. Decrease levels by clicking on the left Increase levels by clicking on the right Click button in the middle to bring up a dialog box Enter desired value and click OK STOP-RUN-RUNn/Stop Sweep/Start Sweep button The text of this button changes depending on whether the signal mode is SINE wave sweep, or real-time processing. If the mode is real-time, the button is split into three active STOP, RUN and RUNn regions. The stop mode action (what halts) is set in the Options pull down menu or Size&Comp tab in the Setup Control Window. When not testing impedance and phase, the last two buttons in the window allow you to quickly set the line sources. The first button is reference A and the second button is reference B. Each has the following options: LL = Line-in Left Channel LR = Line-in Right Channel ML = Microphone Left MR = Microphone Right Click on the appropriate label to select the line source. Real-time Meter Output The measurement window below the buttons shows the main real-time outputs of the ongoing test. It also indicates the current sweep point and sweep ratio, which are variable settings that can be changed in the Options menu. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 15

16 NAVIGATION 2.3. WT Control Window The WT control window consists of a results area on top and a graphing area on the bottom. The results area shows the output of the latest Q/FS, Vas and Box tests. The results area can be minimized or maximized using the [-] and [+] buttons. Minimize Maximize Text Changing the Title in the Output Window Clicking on the first row of the output area pops up a dialog in which the graph title can be changed. This title will also be entered into the Setup Control Window Overlay tab that controls overlay and active data destination buffers. Buttons in the Graphing Area The graphing area displays sweep and real-time signals that are generated. There are two groups of buttons at the top of the graph; two in the left corner and three in the right corner labeled Setup, Auto and Data. Buttons Auto Single clicking this button turns autoscaling on and off. With autoscaling turned on, the range of the phase degree or response axis automatically adjusts to fit the data. This is useful in the initial stages of a new driver test, where you do not yet know what levels of response you will get from the driver. Turning autoscaling off fixes the range of the x or y axis at the current level. Data The Data button on the left turns the phase response line off while the button on the right turns the response line off. Clicking the button again, turns the lines on again. Setup This button opens the Setup Control window, which is described in section Smith & Larson Audio Woofer Tester Pro User Guide v1.1

17 NAVIGATION Setting the Measurement Frequency The two Xs in the graph mark the current data measurements at the selected test frequency. This data is also displayed in the real-time meter window. When enabled, clicking on the graph will change the current measurement frequency. As the mouse hovers over a data line, an additional value is interpolated from the known data points (the drawn line) and displayed. The data label disappears as the mouse moves away: Interpolated Data Hover Current measurement frequency X s show current measurement frequency and values Adding/Removing Datapoints from the Graph After you have measured a driver, depending on the number of data points used in the measurement, you might find pieces of a curve that are not smooth. The missing curve segments can then be smoothed out by adding additional measurement points: 1. Go to the Options menu Left Mouse Button Action Set Frequency Control in WT Control Window and make sure this option is checked 2. Left-click on the segment of the curve that you want to smooth out. The frequency will change and the measurement Xs will move to that location. 3. Wait for data processing to complete. Processing is complete when the impedance reading in the real-time measurement window displays an Ω sign. New measurement is processing (spinning / on Z reading) New measurement is complete (indicated by Ω on Z reading) 4. Now click the right mouse button to add the new data point to the curve. It is also possible to remove a bad data point from the graph: 1. Go to the Options menu Left Mouse Button Action Set Frequency Control in WT Control Window and make sure that it is checked 2. Hold down the Shift key and move the mouse to the data point you want to remove. Notice how the Xs change color. 3. While holding down the Shift key, right click the mouse button to delete the data point. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 17

18 NAVIGATION 2.4. Setup Control Window Calculation of Response and Impedance - Frequency Domain Division Tester response and impedance are mathematically defined as the ratios of 'measured' and 'reference' signals. Frequency domain division yields direct measurements of amplitude and phase and eliminates a post-processing step where pre-measured reference signals are subtracted from the measured signal. Setting the Measurement and Reference Signals The 'Setup & Control' window is opened from the View menu or by clicking the Setup button in the WT Control window. The Setup tab then shows the measured 'A' and 'B' reference sources as well as the test and signal types. Other controls include microphone gain, compensation and response smoothing. The left and right ADC channels each select one of four signals. The left side includes Line Left, Mike Left, LoZP volts, HiZP volts and the right includes Line Right, Mike Right, LoZP amps and HiZP amps. Some combinations like using the LoZP channel as a microphone response reference will not make sense. A B Signal Anal Mode Comment MikeL LineR Sine db(a/b) Mike(L) sine response (ARBx sweep) MikeL LineR Impulse db(a/b) Mike(L) Real-time response ->Buffer MikeL LineR MLS db(a/b) Mike(L) Real-time response ->Buffer MikeR LineL Impulse db(a/b) Mike(R) Real-time response ->Buffer MikeR HiZP V Impulse db(a/b) Mike R response - AFTER amplifier MikeR MikeL Impulse db(a/b) Measure room gain LineR LineL Any db(a/b) Response of a line level device LoZP V LoZP A Sine LoZP Low Power Swept impedance and phase LoZP V LoZP A Impulse LoZP Low Power Fast impedance and phase HiZP V HiZP A Sine HiZP High Power impedance and phase HiZP V HiZP A Impulse HiZP High Power Fast impedance and phase Drive Signals Each signal type has a different frequency distribution characteristic. Pure tone sine waves are single tone and therefore have the highest possible signal density. Broad band or real-time signals contain all frequencies at the same time. The speed this provides is convenient, but this also spreads out the energy and it opens the possbility that non-linear effects will cause computation errors. In theory, all real-time signals will give the same result, but this is not always the case, especially at higher power levels. The Impulse mode for example has more low frequency content than the MLS or NOISE modes making it more suitable for measuring a woofer. Conversely, MLS and Noise are good for measuring tweeters. Care should be taken with the MLS and Chirp sine modes as these modes have zero crest factor meaning the low and high frequency energy can be quite high. 18 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

19 NAVIGATION Line and Microphone Compensation Both microphone and line input sources can be compensated using FRD files, or two-stage user defined filters. Additionally, compensation can be applied to either the measurement or reference channels, and it can be enabled or disabled by signal source type (line or mike). When an option is not valid, it will be grayed out. The filter method provides up to third order high pass compensation by cascading a first (upper adjuster) and second order (lower adjuster) high pass filter. When properly adjusted, this will often yield reasonably flat response for the supplied Behringer ECM8000 test microphones down to ~10 Hz. The two corner frequencies can be adjusted by comparing the measured response to a system with a known roll off. For example, a sealed box system exhibits a predictable 20db/decade roll-off rate below cut off. Just be sure to close-mike the driver (the microphone is less than a few cm away) and do not forget that many amplifiers limit low frequency output to prevent speaker damage. The low power test port s constant current source can also be used to measure the pressure inside a sealed box. In this case, flat response occurs since the constant current will produce a constant force and therefore a constant pressure, all the way down to 0 Hz. Like the sealed box, be sure to measure well below the resonance point and make sure that the box is totally free of leakage and loss. This also happens to be the same pressurization effect seen in automotive and small room installs. FRD file compensation loads an FRD formatted microphone compensation file in either a 2 or 3 column format of Frequency db Phase or Frequency db with no commas. Laboratory grade calibrated microphones can be quite expensive, but this method does provide the best possible high frequency compensation. Show Impedance at Tester Terminals or End of Cable If a full low-level calibration is performed, the resistance and inductance of the test leads will be known. This allows the tester to recalculate the measured impedance for either end of the cable. Measuring at the tester terminals is often desired when measuring resistors, capacitors and inductors. It is also sometimes helpful to measure driver Thiele-Small and in-box parameters at the tester leads,as this is what the amplifier sees as its load. See the section on Calibration for more information. Size & Comp: Buffer Size and Creating a Zero Reference The current buffer size is set using the radio buttons on the left hand side. As the current buffer size (used for FFT and continuous sine) size is increased, the computation window becomes longer in time, and frequencies are defined with greater resolution. You can trade buffer size for speed. The buttons along the right hand side are then enabled or disabled depending on the availability of a zero reference compensation file. Zero referencing is the subtraction of the compensation value from the current measurement. This can, for example, be used to measure the change in room response for different microphone locations (room modes). You would do this by measuring the response in front of your speaker (the reference) and then zero reference that response. The next step would be to move the microphone around the room. Options include creating, deleting, loading or enabling a compensation file. The number of Windows buffers used and the number of frames used to create averaged results is set at the bottom of this window. Sampling Rate - Set the tester s sampling rate here. This rate should match the default device settings for Windows Vista. Number Audio Buffers - Audio buffering requires at least two buffers, one for computing and one to be passed to the operating system for input or output. This is known as double buffering. The idea is that if the OS becomes busy the buffer will be large enough not to have a data interruption. Since the Tester is averaging data and the frame sizes can vary, adding an additional audio buffer can sometimes help. Buffers Averaged - Box-car averaging computes a return value from the average of the last N frames. This reduces noise, but it also slows down response. Larger FFT sizes, more Windows audio buffers and longer box car averaging all slow down response time. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 19

20 NAVIGATION RUN/STOP all IO and Processing - When a real-time processing signal mode is stopped using the STOP-RUN-RUNn button, analog input is halted, output is zeroed and data processing halted. This freezes the display. RUN/STOP IO only (allow re-process) - When a real-time processing signal mode is stopped using the STOP-RUN-RUNn button, analog input is halted and the output is zeroed. However, data processing continues allowing adjustment of the impulse time gate, octave binning and similar data processing. RUN for N - Clicking the right hand side of the STOP-RUN-RUNn button will cause the analog IO to operate for N frames and then halt. To operate properly, N must be set higher than the number of audio buffers+2. This count is needed to flush old or incomplete frames through the hardware and software. Axis Control The Axis control tab is identical to the Axis dialog that opens when the left hand side of graph window is clicked except this is a child window. A child window allows you to select and work in other windows. The controls in this window include X- display Extents: Low and high frequency limits Y- Extents: Y axis min and max limits (one for each half-buffer). Line Smoothing: Drawn line positions are effected (not data averaging). Real-Time Octave Binning: Complex real-imaginary data is averaged before becoming display data. Line width and color schemes: Set black or white background Sweep Points and Sweep Ratio: SweepRatio^N = SweepHi/SweepLo Test Bailout Ratio: This is the final frequency step that will be taken as the Q,Fs and Vas tests zero in on their final values. Smaller values will take longer time to complete, but will result in more accurate frequency readings. Sine mode test points sweep ratio^n = SweepHi/Lo Real-Time Octave Binning Test Bailout Ratio 20 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

21 NAVIGATION Overlay Control The Overlay tab is used to enable one data buffer for data entry in the WT Control Window and up to 32 data overlays in the Overlay Window. The first column is used to enable multiple overlays, and the second to enable the current destination buffer. Buffers Displayed in Overlay Window Box, ARB1, SimZP and SimRx ARB1 is Data Destination Displayed in WT Control Window Title Column Each buffer can be given a unique name by clicking in the list box. Some buffers will have pre-assigned names that go with a particular usage. For example, the first buffer is assigned to the Q/Fs test procedure and is therefore traditionally assigned that name, as are the first seven buffers. Info Column The info column contains information about how a buffer was configured when it was activated. This process is automatic. If you type something in this column it may be overwritten. +db, +Ph, Dly and Pol Columns The data entered in these columns is used to shift response curves (impedance is not affected) up and down. Additional phase polarity (enter + or -) and wrapping control is provided, whereby a time shift (in data samples) can be applied to the raw data before it is processed. Left-Button Sets If the Frequency option is selected, the current test frequency can be set using a left mouse click in the graph area of the WT Control window (or Overlay window, if the same buffer is enabled). If the Phase Wrap option is selected, a time delay (in data samples) will be calculated that will cause the phase to unwrap until it equals the selected mouse point. This is an easy way to set values in the Delay column. Buf->File and File->Buf Buttons These buttons are used to save and restore the currently selected file to an external file. This option allows you to virtually extend the number of buffers beyond 32. STOP-RUN-RUNn button The button text changes depending on whether the signal mode is SINE wave sweep, or real-time processing. If the mode is realtime, the button is split into three active STOP, RUN and RUNn regions. The stop mode action (what halts) is set in the Options pull down menu or Size&Comp tab. This button is also found in the WT Control window. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 21

22 NAVIGATION Calc +db Offset, Calc +Ph Offset These buttons open the following dialog box. A frequency entered into the Measure at field will fill the Measurement field with the value at frequency, in this case If the desired value at frequency is then 10.0, the required offset difference would be or Clicking on the Copy Calculated Offset from Above button enters the offset into the Overlay Control. Mike Position and Gating Control Microphone and reflector distances entered in this window are converted to a time gate that is applied to the burst sine signal. This time gate is not the same as the impulse time gate window for real-time response measurements. In this case, the gate width is automatically adjusted to the timing and width of the burst. Timing is derived using a secondary MLSsignal that is added to the burst. This improves frequency response, but also creates a variable time delay that corrupts phase. Disable MLS to measure phase. 22 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

23 NAVIGATION THD/IMD (Distortion) Setup & Control Harmonic content, Inter-modulation and Noise and Distortion tests are controlled from this window. Harmonic (i.e. THD) is relative to a single test frequency, while inter-modulation results from the non-linear combination of two frequencies. Signal-versus-Noise and Distortion is what is left over from a THD test after the fundamental is removed. THD and IMD levels and frequencies can also be viewed and measured from the FFT child window. A SINAD view similar to an oscilloscope is available in the XY plot window. H1&H2 goto Buffer 13 H3&H4 goto Buffer 14 H5&H6 goto Buffer 15 Set Test Point with Mouse click 1 khz Fundamental 3 khz, 3 rd harmonic 5.5 khz Primary signal 200 Hz secondary signal 5.1, , 5.9 khz, IMD Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 23

24 NAVIGATION SnapTS Fast Testing Real-Time Thiele-Small Testing can be useful for rapidly testing a driver or when tracking parameters during a test. The constant current sine test method should be used when accuracy is important. This includes the Q,FS and Vas tests. Sine testing is capable of far wider dynamic ranges, much improved frequency accuracy, a true constant current mode, and is able to reject noise. It therefore offers much higher precision. Portions of this test are embedded into the DC offset test. 1. To use Snap TS, first select a real-time impedance mode and then connect the driver. 2. Next enable the Acquire QFs Data radio button. If the data is clean and stable, and the sweep lo and sweep high points are sufficiently far apart to measure Re and Le, the upper data values will be filled in. 3. Click on the Freeze button to freeze the Q,Fs information 4. The next step requires entering the effective piston diameter and selecting either the Delta Mass or Test Box method for calculating Vas. Enter the appropriate values. 5. Click on the Aquire Vas Data button. If the Vas results are valid the remaining TS parameters will fill in. 6. Freeze the data and optionally copy the test results to the TS Control window s test data or simulator. 24 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

25 NAVIGATION 2.5. Overlay Window The Overlay window compares multiple data sets including response and impedance data from the Thiele-Small Simulator. The graph display is similar to the WT Control window, except that it includes a row of data buffer selection buttons along the top. These buttons correspond to the first seven buffers that are assigned to particular tests and functions. The remaining 32 buffers are multiple use arbitrary buffers and are selected in the Setup & Control window s Overlay tab. The image on the left compares the measured Fs/Q data from a 4 mid-bass driver to the simulated impedance and phase using a large box to simulate open air. The extremely close fit is a good indication that the model is accurate. Another useful overlay window feature is that real-time response data can be overlaid with a second buffer. The right hand image shows the real-time response overlay of a physical soldered together crossover and the ICD simulated equivalent circuit. The extremely close fit shows that the real world and modeled world are in agreement. Advanced Tip: Viewing Driver Compression The tester configuration db (Mike R/HiZP V) is interesting since this calculates the right channel microphone response relative to the amplifier s terminal signal. Response will be unaffected by amplifier gain but not by driver compression. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 25

26 NAVIGATION 2.6. T/S Entry Window The Thiele-Small Entry window is used to both design and verify box designs. Vented, sealed, passive radiator and band pass box models are supported. Other advanced features include the automatic alignment of the simulator to measured box values, simulation of frequency dependent inductance (Le, Rem, Xem), pipe resonance effects, various Q loss factors, small room and car boost effects, selection of TS models from compression tables and cable loss. This example shows how well the measured and then simulated vented box model has been matched to the actual in air response. Simulated and measured Response and Phase are an accurate match Buttons select an Overlay of simulated and measured response and phase Select Box Type Copy Test Data Into SImulator Auto align simulator to measured box impedance The overlay window is used to view simulated response or impedance curves. As shown in the example above, this can be very useful when comparing simulated and measured impedance or response. T/S data is normally copied from the test environment, but T/S parameters from a manufacturer s data sheet can also be entered manually. Once in the simulator, parameters become adjustable. Parameter adjustments in the simulator only affect the simulator. If necessary, simply recopy from the test environment. Tip: An overlay of the simulated and measured impedance will show how well the auto-align feature has guessed the effective box size, tuning and Q. Manually adjusting the simulation parameters will sometimes achieve an even tighter fit. 26 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

27 NAVIGATION 2.7. FFT and Oscilloscope Window The FFT and Oscilloscope window displays the measurement and reference signals in either the frequency or time domain. Pull down and check menu options control the display and measurement options. FFT (OSC off) When the FFT response mode is chosen, the left mouse button can be used to select a measurement point. There are four measurement modes that find the largest magnitude over a particular range. The frequency response can also be displayed on a linear or logarithmic scale. FMAX - DC to F-nyquist (1/2 sampling rate) range is searched MOUSE - A left mouse click sets the measurement point FOUT LOCK - The measurement point is set to the tester s frequency output - The measurement point is locked and does not change OSC (FFT off) An oscilloscope view is displayed, where the time base is set to the largest magnitude signal. The displayed units are in raw ADC values, voltage or amperage, depending on the display mode and signal. The signal names appear below the buttons. Example A point FFT was selected and an input frequency of Hertz was applied. The 80 db hertz third harmonic was then measured at Hz showing a Hz accuracy. Note: Fourier theory defines the frequency domain resolution of an FFT (bin resolution) to be: Hz/bin = SampleRate / PointsTaken FFT resolution is often on the order of several hertz or tens of hertz and may not be sufficient for some measurements. An additional algorithm significantly improves this resolution at the selected frequency. Resolution is still dependent on frame size (bigger is better) and sampling rate, but resolution is improved to <0.01 Hz when the peak is sufficiently large and is uninvolved with nearby signals (and the crystal time base is corrected). Keep in mind that mathematical accuracy is not the same as absolute accuracy and that the testers frequency accuracy is determined by its time base. The tester s time base tolerance is +/-30 PPM or 0.03 Hz per 1 khz of measured signal. If you require higher absolute resolution, please contact us. Similarly accurate 'math' routines are used when generating the sine wave outputs. In this example the output frequency was set to Hz, and the loop back confirms this. If however two tester s are used with the output of one feeding the input of the other, the potential measurement error will be +/-60 PPM or 0.06 Hz per 1 khz of measured signal. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 27

28 NAVIGATION 2.8. Impulse and Waterfall Control Windows Real-time response is calculated by dividing the measurement and reference. A second algorithm then finds the signal time displacement that is used to unwrap phase. This response is passed back through an inverse FFT revealing the system time domain impulse response. When applied to the in-air frequency response of a speaker, multiple impulses are seen. If a time gating window is applied, the primary speaker impulse can be separated from the secondary floor, ceiling and wall reflections. The forward FFTprocess is called again, revealing the speaker response, minus the secondary room reflections. Tip: Shorter time gates decrease low frequency content and reduce resolution. Impulse time gating is most useful when measuring mid-range and tweeter responses. If the Impulse window is not open, the time gate parameters are unknown and time gating is disabled. The Impulse Control window controls both the response and 3-D Waterfall views. Time Span=37 ms Longer CSD Time Span shows more low frequency content Extended low frequency resolution shows Rippling from reflections Time Span=2.25 ms Less rippling, but low frequency is less defined 28 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

29 NAVIGATION Impulse View Modes and Edge Control (Windowing) Unwanted frequency domain energy can be introduced if a sharp transition occurs at either the start or stop tick. The process of multiplying the signal by a function that softens these edges is known as windowing. In this case, a leading and lagging modified raised cosine function (Blackman window) is applied. If the edge transition is minimal, the window is more rectangular. An adjustable edge allows the window to be modified from rectangular to Blackman. Impulse, Step and Log(Impulse) View Modes Theoretically, the mathematical integration of the impulse response reveals the step or square wave response of the system. This is, however, highly dependent on a number of factors like the low frequency performance of the microphone, preamp and A-to-D converter. It is nevertheless often useful, since it shows low frequency decay as a time signal. STEP response the integral of the impulse response Another useful time domain view is the logarithm of the impulse magnitude. This view shows the impulse and room reflections, where the magnitude is shown on a logarithmic scale revealing reflections that would otherwise not be visible. This is known as room decay and is similar to RT60. RT60 is the time that it takes acoustic energy in a room to decay by 60 db after being passed through a band pass filter. Here, absolute magnitude is shown on a logarithmic scale versus time. The room energy has decayed by nearly 60 db before becoming limited by the noise floor Looking at the same data in the frequency domain shows that the noise floor is mostly low frequency energy. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 29

30 NAVIGATION Cumulative Spectral Decay, Growing and Sliding Fourier Transform Views Different frequency domain viewing methods reveal different types of information. The setup for each depends on what you want to see. Cumulative Spectral Decay (CSD Button) CSD (left) shows FFT slices as the time window shrinks from the left hand side. The display shows the total spectral energy within the time window as energy is removed, hence the name, Cumulative Spectral Decay. Since the initial higher energy pulse is removed first, this method shows the decay characteristics of a driver or speaker. Growing Waterfall 'Forward Time' FFT (Growing Button) The growing 'Forward Time' waterfall (middle) shows response slices as a function of time, as the time window grows from the right (the left is stationary). This method shows energy building with time. Again, tick marks '1' and '2' can be moved to change the gating window trading off low frequency resolution for rejecting room reflections. Sliding Fourier Transform (SFT Button) The SFT plot (right) maintains a constant width time window by simultaneously sliding the left and right markers. A constant width is analogous to perceived loudness making this display useful for analyzing room decay and echo. This is similar to RT60 where the filter Q is set by the SFT gate width. The SFT decay mode uses the first slice as a zero reference with each successive slice showing the energy difference with respect to time. Waterfall Controls Zoom (Zoom button) Clicking on the left zooms out, clicking on the right zooms in. Controlling the Modified Raised Cosine Transition Region (RCOS) The lead-in and lead-out transition region can be 'smoothed' using a modified raised cosine window function. However, since the impulse response has been located in time using a frequency domain technique, windowing in a traditional sense is not required. This adjustment may help in some cases. Flip front To Back (FLIP button) The 3-D display is drawn either front to back or back to front. Slice, Sheet, Color and Wire Frame View (SLICE and SHEET buttons) Clicking the left side creates a wire frame display, and the right side fills in with color. Frequency Domain Smoothing (Binning) FFT response is combined into bins on a per-octave resolution basis. Refresh Data (Refresh button) The 3-D data set is refreshed. Auto-Scale Data (AUTO button) The current 3-D data is auto-scaled and displayed. Export (Export button) This button exports the impulse view data in ASCII format. Sizing and Rotating the Display The 3-D plot can be rotated and sized from the impulse control window using the arrow, page up, page down, insert, delete, home and end keys. 30 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

31 NAVIGATION 2.9. Time Align - Averaging Microphones The tester s ability to accurately measure acoustic distance allows it to sum and add the left and right microphone time signals. The effect is that the incident impulse is preserved and secondary impulses are averaged. This real-time method requires the simultaneous measurement of both microphones when measured against the internal MATH reference for the B-side divisor. Note: Time of arrival is relative to the internal reference. Absolute distance is no longer meaningful. No time gating or line smoothing was used. Figure 2-III - Left Microphone Figure 2-IV - Right Microphone Figure 2-V - Left + Right Microphones (Averaged) Reduced time domain reflections Reduced frequency domain reflections Setup: Open Setup Control Window, select A side measure MikeR (or MikeL), B side reference MATH, Analysis db(a/b), signal type impulse, MLS, noise or chirp Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 31

32 NAVIGATION X-Y Plot Window The X-Y plot window is used to view and compare similar data sets. In particular, this window is used to view changes in TS parameters and distortion as AC or DC levels are changed, or it can be used to view any two time-domain inputs in either a Lissajoux or oscilloscope display. The AC compression test creates a tabular list of driver TS parameters with increasing drive level (you can get this report from the Results pull down menu). A simple examination reveals parameter variation with drive level that can be called up in the Simulator to show the AC response at a particular level. On the other hand, the X-Y plot window allows you to visually graph each parameter versus drive level, or any other parameter. Compliance and BL versus AC drive level The small discontinuity between the low and high power port data is the result of suspension settling (see the section on TS testing regarding Cms variation with drive level). A discontinuity will also occur if the two ports are not exactly calibrated identical. Use the high power calibration resistor to align both ports. 32 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

33 NAVIGATION DC bias testing also produces a table of TS parameters, but these parameters are derived from a small signal analysis at a particular DC bias point. AC simulation of the TS model does not make sense in this case, but these parameters do reveal useful information about driver construction such as mechanical offsets, suspension and motor linearity with excursion. These views show an offset in the suspension as well as the voice coil. Note on Suspension Compliance: A careful examination of the AC and DC results will often show a wide range of input levels, usually up to the point where suspension limiting begins, where the suspension mechanical spring constant (inverse of compliance) significantly decreases as the AC test signal level is increased. This effect is much larger than the static changes seen in the DC offset test. Cms (suspension compliance) is significantly dependent on break-in, drive-level and temperature. The underlying mechanism(s) are a function of how the molecular bonds respond to these conditions: The suspension mechanically loosens with usage. This is commonly referred to as "break-in" and is dependent on the excursion incurred (itself a function of frequency and drive level) and duration. The suspension compliance is affected by changes in ambient temperature. The suspension will self-heat due to mechanical energy loss. This is mostly a function of Rms and drive level, though at high drive levels, misbehavior of the suspension can come into play. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 33

34 NAVIGATION As the DC bias test is run, a sine wave signal can be specified that either tracks Fs or remains fixed at the frequency when the test begins. A harmonic analysis can then be used that reveals additional mechanical problems. However, keep in mind the raw data provided in this case is a cumulative effect of all DC and small signal AC parameters. In the absence of an infinite baffle this test can be performed in an open-air test jig using a closely spaced microphone. The absolute level of Fs and its harmonics are therefore typically low due to back-side cancellation. A trend at a particular frequency will however still be valid. The XY window can also be configured to view SINAD, or Signal versus Noise and Distortion. Normally SINAD measurements are given in decibels as the amplitude ratio of the fundamental signal to all remaining noise and distortion. The XY plot version of this algorithm uses the measured amplitude to create a sine wave at the fundamental frequency that is then subtracted. If the remaining signal is then plotted against the fundamental, data views of NADvs Magnitude, Lissajoux and oscilloscope are possible. Note: The Xn / int(xn) button numerically integrates the data. Microphones produce signals that are equivalent to pressure, that in turn is proportional to mass-air velocity. Integration shows mass-air displacement. 34 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

35 NAVIGATION View Menu Options There are various ways to view the data and test results in the Woofer Tester Pro. Selecting View from the menu bar gives the following list of viewing options, each of which is its own separate window: View View Name Function WT Control Graphs impedance and response of the current test signal, shows realtime Thiele Small parameters and current test frequency Overlay Overlays multiple real-time or static data series T/S Entry Enter Thiele-Small and measured box electrical parameters for use in simulator Results Editor Measurement results are copied into this editor to save in a text file and for further processing ICD Editor Enter the circuit for the crossover design you want to simulate FFT The upper FFT window shows the left channel and the lower window shows the right channel of the digitized signal. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 35

36 NAVIGATION View View Name Function Waterfall Displays a waterfall graph of the response in multiple time windows Impulse & Waterfall Control Controls the time window of the response data sample. Zoom in on the data. Setup Control Set test parameters here: input/output channel, drive signal, analysis mode, microphone controls, compensation and offsets. Graphing parameters can also be adjusted here. XY Plot XY plot the following data: - AC compression Thiele Small parameters - DC compression Thiele Small parameters - Harmonic content versus DC bias. - Noise and Distortion versus fundamental drive signal Graph Axis Control Modifies the graphing options by changing the X and Y axis range, turns autoscaling on and off, sets frequency ranges and number of sweep points. In addition to selecting data and test windows, the View menu contains a number of options to change attributes in these windows, such as applying Bezier Line Smoothing, setting the FFT Resolution, changing the line width in graphs, adjusting the color schemes and changing the legend. These options are described next. View Bezier Line Smoothing This option applies to the WT Control and Overlay View only and is turned off by default. When turned on, Bezier line smoothing applies a polynomial algorithm to interpolate between points in the graph resulting in a smoother looking response curve. However, as the curve may not even pass through the original data points caution has to be taken when using this feature so as not to misrepresent the measurement. 36 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

37 NAVIGATION View FFT Resolution (Octave Granularity) This WT Control and Overlay graphs show data points that are converted from linear spaced FFT data using a ratio metric or octave scale. The display granularity is set here. By default, the resolution is set to 256 points providing reasonable resolution. High Resolution, 47 points/octave Low resolution, 3 points/octave View Line Width This option adjusts the width of the lines in the graphs. Adjust to value upward to increase the thickness of the line. View White Color Scheme By default, the background color in graphs is set to black. For publishing, however, a white background might be preferred. The View White Color Scheme menu option changes the background color. View Pick Colors The Pick Colors option is used to change the color scheme of graphed data. The 16 Custom colors are arranged in 8 vertical pairs, with each pair corresponding to a pair of colors that can be used in the tester s graphing windows. The first pair corresponds to the first of 32 data buffer pairs. That buffer is commonly called the Q,Fs buffer since it is associated with that test. The second vertical pair corresponds to the Vas buffer (the second of 32), the third is Box, followed ARB1, ARB2, SimRX and SimZP. Since there are only 8 custom colors, the colors then repeat. As an example, the default curve color settings for the Q,Fs test buffer is Turquoise for Phase, and Red for Impedance, the 1st and left-most Custom color pair. To change the color of the Q,Fs phase plot, first left-click on the bottom box (Turquoise) of the left-most vertical pair. The box highlights, and the curve color appears above Color/Solid. Next, select a new color from the basic color palette, the color spectrum, or enter values for Red, Green and Blue. Picking Bright Green from Basic colors causes the new color to be shown in Color/Solid. The highlight will move away from the Custom color Turquoise box, but this is remembered. Now, left-click the Add to Custom Colors button: The bottom left-most Custom colors box changes to Bright Green. Finally click OK. The Phase Curve and the left margin scale, etc., are now displayed in Bright Green. View Reset Colors This option resets any changes to the color scheme back to the default settings. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 37

38 NAVIGATION View Change Legend If you want to use the frequency response graph in a test report, for example, you can add a custom legend to the graph with the View Change Legend option. The following example shows how to add a legend for a driver called TEST DRIVER 1 to the frequency response graph along with the current date and time. Add Legend and pointers to Frequency and Impedance 1. Select View Change Legend from the menu bar to open the dialog box below. You will see some lines of text consisting of codes shown in [ ] and their assigned values. The codes determine the size and position of the legend text and whether pointers are drawn or not. 2. To add a legend, change the first line in the dialog box from [OFF] to [ON] to show the legend. 3. The TEXTPOS(column,row) code sets the location of the legend. 4. In the third line, set the font size by changing the size code from [SIZE(20)] to [SIZE(16)]. 5. Replace Verdana with Arial next to the FONT tag to change the font. 6. The WIDTH attribute sets the line width of the arrow pointers. Let s change it to To change the color of the font or a line, use the [COLOR(RGB())] statement. On line 6, we are using it to set the color of a border we are going to draw around the text in line 7. Change the RGB values to RGB(0,0,0) for black. 8. Line 7 draws a border around our legend text. The values (6,0,28,5) specify the columns and rows of the upper left and lower right corners. 9. On line 9 is the first line of text in the legend 10. To automatically include a current date and time, use the %DATE% code as in line The COLOR(ZBUFx) codes on lines 11 and 13 set the colors for the frequency and impedance values. The value (ZBUF3) sets the color to match the color scheme in the existing graph. 12. %FREQ%, %PVAL%, and %ZVAL% represent the current frequency, phase and impedance values. 13. To draw an arrow pointer to the current impedance and phase values in the graph (represented by the two Xs), use the [ARROWTO(x,y)] statement. In our case, the x value is derived with the GETX(FRQ) statement, which reads the location of the current frequency value on the screen. PVAL(FRQ) reads the phase value associated with the current frequency. The number 15 after the PVAL(FRQ) code represents the size of the arrowhead. Change it to Lastly, add an [OFF] statement to signify the end of the legend definitions. Any text after an [OFF] statement is considered a comment. The text in your window should now look like this (line numbers not included): Line # Code 1 [ON()] 2 [TEXTPOS(8,2)] 3 [TEXTWT(700)][SIZE(14)] 4 [FONT("Arial")] 5 [WIDTH(2)] 6 [COLOR(RGB(0,0,0))] 7 [BORDERTXT(6,0,28,5)] 8 [COLOR(RGB(100,255,150))] 9 TEST DRIVER 1, WT-Pro 10 %DATE% 11 [COLOR(ZBUF3)]Frequency=%FREQ% 12 [COLOR(PBUF4)]Phase=%PVAL% [ARROWTO(GETX(FRQ),PVAL(FRQ),12)] 13 [COLOR(ZBUF3)]Impedance=%ZVAL% [ARROWTO(GETX(FRQ),ZVAL(FRQ),12)] 14 [OFF()] 15. Click OK to add the legend. 38 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

39 NAVIGATION The following table lists all the codes available to create a legend: Function Syntax Description ARROWTO [ARROWTO(x,y,size)] Draws an arrow from current position of the legend text to a specified [x,y] position. The x,y position can be determined by a function, ie GETX or PVAL. Size is arrow point size BORDERPIX [BORDERPIX(50,10,200,50)] Draws a box. The location/size of the box is specified in pixel pairs [x1,y1]&[x2,y2] for the upper left and lower right hand corner respectively. BORDERTXT [BORDERTXT(6,0,28,5)] Draws a box. The location/size of the box is specified in columns/rows pairs [x1,y1]&[x2,y2] for the upper left and lower right hand corner respectively. COL [COL(c)] Sets text to start at column c. COLOR [COLOR(RGB(r,g,b))] Sets font/line color using an rgb() value where r=red, g=green and b=blue. R,g,b values range from 0 to 255. DRAWX DRAWX[(100,50)] Draws an 'X' at pixel position [x,y]. FONT [FONT("Arial")] Sets the font. GETX [GETX(f)] Gets horizontal pixel position x at freqeuncy f. LINETO [LINETO(200,50)] Draws line from current legend position to [x,y] specified in pixels. MOVETO [MOVETO(200,50)] Moves current line position to [x,y]. OFF [OFF()] Turns legend off, adds a comment. ON [ON()] Turns legend on. PVAL PVAL(f) Gets P value (Phase) for frequency specified in f. RGB [RGB(r,g,b)] Returns rgb() color. ROW [ROW(r)] Sets text row to r. SIZE [SIZE(s)] Sets text size to s. STARTCOL [STARTCOL(c)] Starts all text at column c. STARTROW [STARTCOL(r)] Starts all text at row r. TEXTPOS [TEXTPOS(c,r)] Sets upper left corner of text to column, row c,r. TEXTWT [TEXTWT(w)] Sets the thickness of the font. WIDTH [WIDTH(w)] Sets the line width of the arrows to w. ZVAL [ZVAL(f)] Gets Z value (Impedance) for frequency specified in f. FRQ FRQ Gets current frequency. ZBUF ZBUF Sets color of Z value (Impedance). PBUF PBUF Sets color of P value (Phase). %DATE% %DATE% Displays current date in mm/dd/yyyy form. %FREQ% %FREQ% Displays current frequency as a string. %NAME% %NAME% Displays the driver name from the real-time meter window. %PVAL% %PVAL% Displays current P value (Phase) as a string. %ZVAL% %ZVAL% Displays current Z value (Impedance) as a string. Note: If the X value in the ARROWTO function is replaced with GETX(FRQ), the arrow will dynamically move as you select a new test frequency. The following example draws a line and arrow from point(20,30) to the current frequency and phase value. [moveto(20,30)][arrowto(getx(frq),pval(frq),15)] Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 39

40 NAVIGATION Arranging the Test Windows When you have multiple windows open, a quick way to arrange them is via the Window menu option. The windows can be arranged vertically, horizontally or in cascading order. For example, with the Tile Vertical option, four open windows will arrange like this: WOO Files & Screen Configuration A.woo file is a configuration file for Woofer Tester desktop and measurement settings. It allows you to save a specific window arrangement of your desktop along with the parameters of your current test setup. The software also loads and saves a.cfg file (actually a WOO file) each time it starts and stops. File Woo File Save Woo File To save your current desktop configuration, chose File Woo File Save Woo File from the menu bar, select a filename and click Save. File Woo File Load Woo File The next time you start up the Woofer Tester application, you can simply load a saved.woo file to bring you back to where you left off. Go to File Woo File Load Woo File, select the desired configuration file and click Open. This will restore you desktop view and measurement options. Note: Some parameters that make no sense are not saved. For example, the SINAD oscilloscope view and the Bezier line smoothing options are not kept. File export [Export Options] The tester exports data to a number of different file formats: ASCII Exports data from all buffers and result windows to a text file. BassBox Exports Thiele Small parameter measurements to a.log file for import into BassBox LMS Exports frequency, impedance and phase data points for import into LMS software. ZMA - Exports frequency, impedance and phase data points for import into WTPro crossover designer (refer to chapter 8 on Crossover Design) or third party software. FRD Exports frequency, db amplitude and phase for third party crossover design software. 40 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

41 NAVIGATION File Capture Graphics [Graphics and Print Options] The current graphical image can be saved to a BMP file or directed to your printer. Depending on the selected options, this can be the entire desktop or just the child window you are working with. Keep in mind that print conversion drivers, such as those that convert to PDF, can be used for the target printer. If the window is not graphical or contains text fields these are sent to the printer as text. LPI is the acronym for Lines Per Inch. More lines per inch (bigger numbers) produce smaller printed text Viewing, Formatting & Converting Results Results from the latest Thiele-Small parameter measurements, box tests, calibrations and Arb sweeps are automatically saved and can be viewed via the Results menu option. For a shortcut to opening these windows, use the function keys F4 F9. All result windows have a similar format, which is a text editor that lets you copy, save and edit the results, as shown below: On the left-hand side of the results windows are buttons for the editing functions: Load loads a previously saved results file Save saves the current results to a text file Copy copies selected text to the clipboard Paste inserts text from the clipboard Cut deletes selected text SelAll selects all text Results New Speaker Title With the Results New Speaker Title menu option, you can modify the title in the WT Control window output. Selecting this option, opens a text dialog box. Enter the new title here and click OK. This will change the title as shown below: Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 41

42 NAVIGATION Results Results Data Formatting The second formatting option in the Results menu, Results Data Formatting, determines in what format the results are written to the results window. By default, results are saved in scientific number format and columns are separated with a pipe character [ ]. The pipe-delimiter is useful when results are later imported into Excel, for example. Results formatted in scientific number format and pipe delimited Results formatted in decimal delimiter format without pipe To change the number format, select Results Data Formatting from the Results menu and uncheck the Scientific number format option: Similarly, to remove the pipe-delimiter, uncheck the first option and click OK. Results Create Driver Label This option takes the real-time measurement results from the WT Control window and saves them in a format useful for creating a label that can then be affixed to the driver being tested. Results Copy between buffers For a simulation, it can be useful to be able to temporarily save a data set that is later to be used in a data overlay. Selecting the Copy between buffers option opens the dialog box shown below. Choose the result set to be temporarily saved from the left-hand column. From the right-hand column, select the target buffer to hold the data and click OK. You can also save and load buffers to files. This is a good way to move data between woo files without copying the entire woo file. 42 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

43 NAVIGATION Results Clear Data Buffer(s) The Clear Data Buffer(s) function will reset results to zero based on one of the tree options selected. Clear all data on start resets all result sets each time the application starts up Clear all data now instantly resets all result areas Clear current data buffer resets the buffer currently selected as indicated by an X Current buffer selection Options Menu In the Options menu, various system and test configurations can be set. Options Frequency Dependent Le Test Method This option allows you to choose between three different Thiele Small measurement models: 1) the classic Bare Tank test, 2) a two point frequency dependent test method and 3) a three point frequency dependent model. The effects of frequency dependent inductance (Le is not constant) can be 'backed out' during the Q, Vas and Box tests resulting in models that lead to higher accuracy box simulations. The two point curve fit method uses the impedance and phase at SweepHi and Revc (Z at DC) to calculate Rem/Xem coefficients. This simpler method always produces results but may be slightly less accurate. The three point method finds two high frequency points near SweepHi that have a positive phase slope. This usually produces a better upper frequency fit but it may not converge. High frequency negative phase slopes are created when the test lead inductance is significant compared to driver Le and the test leads are not calibrated or moved. Options Sweep The Sweep start and Sweep end menu options allow you to set a test frequency range. This feature is useful for zooming in on a particular range. These options automatically adjust the x-axis scale in the graphs of the WT Control and Overlay windows. The number of Sweep Points is configurable. A higher number of sweep points yields more accurate results, but also increases the measurement time. Enter the desired number of sweep points in the dialog box and type OK. The maximum number of points allowed is 512. The SweepLo, SweepHI, Sweep Points and Sweep Ratio are related to each other by the following equation. Basically this says that as a test progresses from the lowest frequency (SweepLo) the test frequency increases by SweepRatio until SweepHi is met: Sweep_High = Sweep_Low * Sweep_Ratio^Sweep_Points Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 43

44 NAVIGATION Search Ratio Min determines the minimum sweep ratio, which is the stopping point for the search algorithm as it zeroes in on the final frequency value. A minimum search ratio of means that the final value will be determined by a sweep ratio of 0.001%. Options Measurement Frequency This submenu option is used to set the measurement frequency. It can be set either to a predetermined frequency ranging from 20.0 khz to 1Hz or a user defined frequency, which is entered via a dialog box. Options Drive Level Drive levels can be set to predetermined value, user-defined values or full scale (100%) with this menu option. User defined values are input into a dialog box. The range of predetermined drive levels is from 0.1% to 50 %. Small drive levels like 0.1% are used for high resistance values or T/S testing during an AC compression test. Tip: The measurement frequency and drive level can also be set in the WT Control window using the signal control buttons. Options Mute button includes AGC AGC stands for Automatic Gain Control and applies to drive level control. With AGC turned on, drivel level will increase to just below the clipping point. This feature is helpful for setting levels in high impedance circuits. When AGC is turned on, the mute button has three modes: MUTE, UNMUTED and UNMUTED AGC. Options Show drive level in Amps Changes the drivel level on the WT Control window signal button from % to amperes. Options Left Mouse Button Action This option controls the behavior of the left mouse button when used in the WT Control window graph. Set Frequency Control in WT Control Window - when checked, enables the left mouse button to select the current measurement frequency in the WT Control window. This feature is used to add or remove data points from a curve as described in section 2.3 WT Control Window. Adjust Phase in WT Control Window - when turned on, enables the left mouse button to adjust phase in the graph of the WT control window. Options Signal Mode Select the type of signal mode to use in a test with this option. There are 14 modes to choose from: Real-time modes include Impulse, MLS, Noise, Chirp and are used for fast impedance testing Sine mode is used for producing high precision results 44 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

45 NAVIGATION Note: Each signal type will have a different drive level and frequency content characteristic. The Impulse mode for example has more low frequency content than the MLS or NOISE modes making it more suitable for measuring a woofer. Conversely, MLS and Noise are good for measuring tweeters. Care should be taken with the MLS and Chirp sine modes, as these modes have zero crest factor meaning the low and high frequency energy can be quite high. In all cases, excessive drive levels can damage delicate drivers. Please see the ICD tweeter protection circuit, described in Chapter 8, for a possible solution. Options Smoothing (all real-time modes) Select this option to turn FFT averaging on and off. When enabled, it smoothes response curves from noisy signals. Smoothing applies to Impulse, MLS, Noise, Chirp signals. Tip: Signal mode and Smoothing can also be set in the Setup Control Window Options Buffer size, frames and averaging This menu option allows you to the change buffer size, the number of buffers and the number of frames averaged in signal processing.! Changing these settings can impact your system performance. Buffer size - determines the size of the data sample processed at one time. A larger buffer size will require more memory from your system. Depending on the performance of your CPU, you might see a delay in signal updates at larger buffer sizes. Number of buffers is the amount of data chunks processed at one time. The recommended setting is 2, but the system handles up to 8 buffers. Increasing the number of buffers can enhance performance of lower CPU systems. To change the number of buffers, type a number from 1 to 8 into the dialog box and click OK. Frames averaged sets the number of frames averaged per data point. For higher precision, enter a higher value for this setting. However, keep in mind that a higher value will increase processing time. Therefore, for fast testing, this value would be set to a lower number. The system will average up to a maximum of 8 frames. Options Sample Rate The system has four sampling rates to choose from: 48kHz, 44kHz, 22kHz and 11kHz. The default is set to 44kHz, which is the sample rate used in CDs. A low sample rate is used to measure low signals and a high sample rate will increase the quality of the data. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 45

46 NAVIGATION Options Measurement Channel & Reference Channel The measurement channel option sets the signal source, while the reference channel sets the base signal for the test. These settings can also be selected in the Setup Control window and are covered in the following chapters. Line L left channel Line R right channel Mike L left microphone Mike R right microphone LoZp V low power voltage channel LoZp A low power amperage channel HiZp V - high power voltage channel HiZp A high power amperage channel Ref reference channel Options Mike Gain The tester has a Mike Gain option that controls the input level of both microphones. In room response measurements, for example, the gain on the left microphone can be set to a different level than the right microphone to compensate for the distance between the two. Options Analysis Mode This option selects the analysis mode for a particular test and will be covered in later chapters under various topics. Analysis mode can also be set in the Setup Control window Tools Menu Tools Set Atmospheric conditions Atmospheric conditions that affect the calculation of Thiele Small parameters and the simulator response are set here. Inputs include temperature, humidity and atmospheric pressure. Outputs are density, speed of sound and acoustic impedance. 46 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

47 NAVIGATION Tools Ellipse/Oval Calculator This calculator finds the effective diameter and area of elliptical/oval drivers, or drivers that have a phase plug or exposed pole piece. Vas is defined as the volume of enclosed air that creates an air spring of equal stiffness to that of the driver. Therefore, stationary mechanical components like phase plugs and pole pieces with unsealed dust caps do not contribute. Furthermore, the outer suspension is measured at its 50% points since the outer edge does not move while the inner edge moves with the cone. To calculate diameter and area, enter the following data in the dialog box: Major dimension (mm or inches) enter the length of the long side from center of suspension to center of suspension Minor dimension (mm or inches) enter the length of the short side from center of suspension to center of suspension Phase Plug Diameter Enter the diameter of the central phase plug or exposed pole piece. Enter zero if the dustcap is sealed. The calculator then calculates area in m 2 /inches 2 and effective diameter in mm and inches. entry for phase plugs, but does not include an oval calculator. The Vas test includes a similar Tools Design Air Core Inductor The air core inductor tool will help you design air core inductors for your crossover designs. 1. Enter the desired inductance value in the first field and click enter. 2. Input the inner diameter (D) - larger diameters produce lower inductance. Hit enter. 3. Next, enter the height in inches (H). Click enter. 4. In the fourth field, enter the wire gage in AWG and hit enter. 5. Next, input the wire insulation in inches and click enter. 6. Lastly, (if important) set the temperature in degrees Celsius and click enter. 7. The results at the bottom tell you how much wire you will need to wind on to the form. Tools Master Level Mixer (Windows Vista Users only) The input and output session levels can be temporarily changed with this menu option. Note that the default settings, defined in the MIXERLVL.CFG configuration file, are not changed by the mixer. You must manually change MIXERLVL.CFG to make a permanent change to the input and output levels. Mute turns the input/output off. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 47

48 NAVIGATION Lock 100%/25% - sets the master output level to 100% and the microphone-in level to 25% (Tester default settings). Manual Entry select this option to manually enter a level setting between 0 and 1. Enable All enables all audio devices in the list. Adjust the volume with the UP and DN buttons Adjust the volume by selecting the Manual Entry checkbox and entering a value Tools Auto-Set (Windows Vista Users only) This menu option resets the current mixer levels to the values specified in the MIXERLVL.CFG configuration file. The message on the right appears on startup if the mixer settings do not match the specified levels. The message asks you to confirm changes in mixer settings. Windows Vista allows users to set different master volume levels for the various audio devices connected to the computer. By default, Vista requires the (microphone) input level to be set to 25% and the output to 100%. MIXERLVL.CFG is located in your Tester s installation directory and can be configured by editing the file (you should not need to change this). Contents of MIXERLVL.CFG: Edit volume settings in MIXERLVL.CFG file. Values are in % and range from 0 to Smith & Larson Audio Woofer Tester Pro User Guide v1.1

49 NAVIGATION Printing Files/Graphs Depending on the type of window open, you have the option to print or save to a file, either as a BMP graphics image, or as an ASCII text file. Note: Some print drivers that perform format conversions are compatible. For example, printing a graphics image to a PDF converter will produce a PDF Exporting Results Test data can be exported in a number of formats. The export type will depend on what you want to do with the data. A common export is a ZMA files that is used in the Interactive Crossover Designer Help Menu Online help, as well as calling up your web browser and connecting to the Smith & Larson Audio Website can be done here. The Enable Button Help option turns popup help messages on and off. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 49

50 CALIBRATION 3. CALIBRATION CAUTION: You may not want to connect an amplifier to the tester until you are familiar with the tester and using/calibrating the low power test port Calibration, Component & Linear Testing The low power port must be calibrated to use the tester. For best results, calibrate both the High and Low Test ports. Two calibration routines are run for each port. The first routine calibrates impedance and phase, and the second calibrates the voltage and current readings. Cable effects will also be mathematically removed Low Power Port Calibration Low power test port calibration begins by configuring the tester for a Sine test signal and LoZP operation. This can be done from the Options pull down menu, or from the Setup Control Window. Selecting the LoZP mode automatically sets the Measurement and Reference sources. Sine Wave Signal Mode Low Power Impedance & Phase Analysis Low Power Calibration Steps Start the calibration routine using the ZP Calibrate button, or using the Tests pull down menu: Begin ZP Calibration 50 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

51 CALIBRATION The first dialog box that opens will ask whether you want to perform a full or simple calibration. Both methods are similar in that they will use the calibration resistor and or short to set the tester s internal gains and offsets. The difference is that, in full calibration mode, the test leads inductance and resistance are measured. If a full calibration is made, test lead impedance compensation can be turned on or off. In simple calibration mode, the tester and lead parameters are combined making measurements at either end of the test cable the same. Note to Windows Vista Users: The following dialog box opens reminding users to check and/or adjust the master level settings. The master level settings are: 100% for playback and 25% for recording. The dialog flashes red if an improper configuration is detected. The Continue button brings up step 1 of 5, which asks that the calibration resistor be attached to the tester (full), or at the end of the test leads (simple). This measurement is made at 200 Hz and sets the low frequency gain. If you are using a high resistance cable, the Pass message may not appear. You can override this by clicking OK. If the resistance is stable, calibration will be valid, but a little less accurate. Full Calibration Simple Calibration Measuring at the tester terminals isolates the cable The internal tester and external test lead resistance are combined The WT2 is identical in operation to the low power port Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 51

52 CALIBRATION Note: Measurements are not instantaneous. High-level noises around the test area are rejected using tuned narrow band filters. This slows down the calibration. The 1-10 Hz band is particularly slow and takes several seconds to complete. Wait until the Check Connections message changes to Pass before hitting NEXT. Process Step Setup Info Measurement Wait for Check Connections to change to Pass The calibration resistor is re-measured in step 2 at the test lead terminals (Full) or end of the test leads (simple) using a higher test frequency. Full Calibration Simple Calibration The calibration resistor is then replaced with a short. The measurement is again either made directly at the tester terminals (full) or at the end of the test leads (simple). A shorting bar made from a banana jack and section of buswire may be helpful here. For the simple calibration method simply connect the alligator clips. Full Calibration Simple Calibration Using a shorting bar Short the test lead ends 52 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

53 CALIBRATION Step 4/5 is another shorting test. In full calibration mode, the cable is inserted into the circuit and measured using a short at the far end. This physical configuration is already setup for the simple calibration mode. Wait for Pass and press NEXT. Full Calibration Simple Calibration Step 5/5 is the final high frequency test that measures cable inductance. Full Calibration Simple Calibration Calibration is complete and the test results are sent to the results window. Using an ARB Sweep To Verify Calibration An arbitrary sweep will show calibration quality. Use the Axis Control to set the SweepLo and SweepHi points and click ARB1. As the test runs, you will notice the impedance is very flat. To see just how flat, use the Axis Control window to narrow the impedance range to and ohms. The next image shows the change in test lead inductance. The testers calibration process will take this into account as long as the cables do not change. To get the best possible results use a zero length test cable by plugging directly into the tester or use a section of speaker cable. Zero length calibration is very useful when measuring components and cables. Arbitrary sweep showing calibration flatness Scale: ohm/division degrees/division Comparison of shorted test lead inductance loop when tightly wrapped or wide-open. 900 nh tight, 2200 nh open Tip: The inductance of pair of simple test leads varies considerably. A test cable can be made from a section of speaker wire, banana jack and alligator clips. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 53

54 CALIBRATION Low Power Port Voltage and Amperage Calibration Impedance and phase calibration tunes the voltage and current ratio on an absolute scale. This test must be run from the Test pull down menu. Leaving the 10-ohm calibration resistor connected to the low power port, choose Tests Calibrate Calibrate Volt/Amp. Using an accurate DVM, measure the test load voltage and enter the value as shown. A 60 Hz frequency is used since many AC voltage tester s are intentionally band limited. Enter the value from the DVM into the Volts from DVM window in the dialog box and click OK. Enter AC RMS voltage from DVM here The low power test port is now fully calibrated. Alternate Configurations A shorting bar made from a spare banana jack can be quite helpful. The second image shows the zero length test cable configuration that is used when extreme accuracy is required. Using a Banana Jack for the Short Zero Length Calibration with Cal-R and Short 54 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

55 CALIBRATION 3.3. High Power Port Calibration High power test port calibration begins by connecting the line outputs to your test amplifier and returning one of the amplifiers outputs to the LOAD side of the high power test port. Be sure to connect the ground-side of the amplifier to the black 5-way binding post. CAUTION: A ground-referenced amplifier, with the ground connected to the black terminals, must be used for proper operation. Voltage at either black terminal (ends of the sensing resistor) should not exceed 4 Vrms. If the leads are swapped, or if the amplifier ground lead is resistive, a large common mode signal will be created at the differential sensing inputs (between the black terminals). Placing the sensing resistor in the ground side of the amplifier improves the high gain sense amplifier s Common Mode Rejection Ratio (CMRR). This increases the tester s usable dynamic range. The 0.5 ohm sense resistor is connected between the black terminals. The amplifier s ground-side terminal is connected to the left AMP side black terminal. High Power Port with amplifier leads connected to the AMP terminals (left) and 10 ohm calibration resistor on the LOAD side (right) The next step configures the tester for sine test signal mode HiZP operation. This can be done from the options pull down menu, or from the Setup Control Window. In both cases selecting the HiZP mode also automatically sets the Measurement and Reference sources. Sine Wave Signal Mode High Power Impedance & Phase Analysis Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 55

56 CALIBRATION Start the calibration routine using the Calibrate button, or using the Tests pull down menu: Begin ZP Calibration High Power Calibration Steps The initial dialog that opens will ask for you to OPEN the load side terminals. This establishes a baseline measurement of the open circuit voltage and amplifier gain. Open circuit voltage is first measured at 60 Hz. Open circuit voltage is then measured at 15 khz The third step measures internal signal isolation. In this step you disconnect or disable the amplifier output, but DO NOT ADJUST THE VOLUME. 56 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

57 CALIBRATION The amplifier is reconnected allowing the tester to measure the terminal voltage with the test load applied. Terminal voltage should be a little lower than in steps 1 or 2. Step five is the last calibration step. It finalizes measurement of the total amplifier side resistance and inductance. Calibration is now complete. You are then asked if you want to include any LOAD side cable effects in your measurements. Select NO (you can enable this later in Setup control) High Power Calibration Results The high power calibration results include a measurement of the total amplifier side resistance and inductance. This is the sum of the amplifier and AMP side cable. If the CABLE characteristics are known, they can be subtracted leaving a measurement of the amplifier output resistance and inductance. This can then be used to compute a damping factor. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 57

58 CALIBRATION High Power Voltage and Current Calibration High power voltage and current calibration is similar to that of the low power port, except that an optional additional calibration is included for DC offset testing. Select Tests Calibrate Calibrate Volt/Amp from the menu. The tester automatically performs the calibration. If you do not have a DC coupled amplifier, or do not need to do those types of tests, ignore the Cal DC option. The high power test port is now fully calibrated. Capacitor Measurement The menu option Tests Capacitor Measurement Setup is used to fine-tune the low power port for measuring small value capacitors. Crossover capacitors are typically quite large compared to the internal capacitance of the test port that is typically pf. When capacitance is large, simply placing a 10 kohm resistor across the port will create a DC path for the port s constant current source. If these values are fined tuned, sub 100 pf measurements are possible. The process is simple. Click on the Auto Complete button and wait for the alignment to complete. Then click ACCEPT. You can also step through manually to see how each alignment step works. Measuring Revc The Tests Measure Revc menu option simply calls the Re test from the Q/Fs test (and stops). Linearity Tests The Woofer Tester Pro also includes two linearity tests: The first test measures Thiele Small parameters versus changes in the AC drive level (compression). The second test measures Thiele Small parameters versus DC bias revealing voice coil and suspension alignment and linearity. Select the Tests Linearity Tests menu option to conduct these tests. More information on linearity tests can be found in the XY plotting window, and in the section on TS parameter testing. 58 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

59 THIELE SMALL DRIVER MEASUREMENT 4. THIELE SMALL DRIVER MEASUREMENT! WARNING! The LOW POWER TEST PORT differential current source output (banana jack) is not ground referenced and cannot be safely connected to grounded or externally powered equipment without risking damage to the device. This applies to Amplifiers and Oscilloscopes. Simple two terminal devices like Loudspeakers, Crossovers, RLC devices, or hand held meters (that are free floating relative to ground) are the only acceptable loads. The actual process of measuring a woofer can be fairly complex but the basic concept is that drivers are mechanically resonant transducers that are coupled to an amplifier on one side and air on the other. The physical electrical, mechanical and acoustic characteristics are then broken down and modeled. The tester s job is therefore to reveal this information with minimal effort and with maximum accuracy. Extensive literature on T/S modeling already exists on this topic, so only basic theory is covered. An overview of how the tester performs these tests and why this is important is provided at the end of the section. Baseline measurements are typically made using a modest 1-3mA drive, so you may not hear the tester working. This modest level provides a large enough signal to make accurate measurements while at the same time not driving the mechanical mechanism to the point that a suspension or motor non-linearity has appreciably changed the at rest characteristics. This is known as a small signal model. Using smaller or much larger (WTPro high power port) test signals later reveals these characteristics in detail over a wide range of drive levels. Baseline Thiele Small models are found using a search algorithm and sine wave test signals. Sine wave testing is used since this provides the best possible signal to noise ratio, is highly repeatable, provides constant current at all frequencies, does not excite other resonance points, and ensures the mechanical components of the driver are biased equally positive and negative. The real time impedance modes will show similar graphs, but these are not sufficiently accurate to extract information suitable for further analysis Revealing the Thiele-Small Model using Electrical Tests The electrical load of a speaker is a very important parameter for an amplifier, but more importantly, these characteristics reveal the Thiele Small (T/S) model. This is achieved by measuring the driver in an un-modified open-air state followed by a modified state where either the mass or the compliance is physically altered. Analyzing the height and sharpness of the electrical resonance peak also reveals the electrical and mechanical Q or ring down characteristics. Later, when the driver is placed into a sealed or vented box, these same principles are applied to reveal the system level characteristics of the completed system Finding Mass and Spring Constants that Set Free Air Resonance Fs The free air condition is achieved when a driver is attached to a solid un-moving frame of reference with no enclosed air volume behind it. In this configuration the cone, voice coil and suspension each have a contributing mass. Likewise, the suspension components are mechanical springs attaching the mass components to the driver frame. The resonant frequency of a spring and mass system is given by: Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 59

60 THIELE SMALL DRIVER MEASUREMENT (1) Fs 2 * Kms Mms where Fs = Mechanical system resonance in Hertz (also known as Fms) Kms = Mechanical system spring stiffness in Newtons/Meter Mms = Mechanical system moving mass in kg Electrically, the tester sees a mechanical resonance as a parallel circuit consisting of a capacitor and inductor with an additional resistor (the voice coil) in series. This equivalent circuit then rings at the same frequency as the mechanical resonance. The tester measures this frequency by applying a series of test signals that are designed to search for the zero phase crossing that coincides with the impedance maxima. Note: In the T/S model mechanical suspension stiffness (Kms) is electrically equivalent to a capacitor whose value is Cms=1/Kms. Therefore Cms compliance is often used when describing a T/S model. Compliance is simply the inverse of stiffness. The next step is to modify either Mms or Kms. Modifying Mms is known as the delta mass method and modifying Kms is known as the delta compliance or test box method. The new resonance equations are shown in equation 2. (2) Fsmod 2 * Fsmod 2 * Kms Mms Madd ( Kms Kbox) Mms where Madd = Additional mass added to the cone Kbox = Stiffness of air spring from a test box By knowing the before and after resonance, plus the added mass or box air stiffness, a little bit of algebra reveals the following two equations. The first is used when the delta mass method is selected and the second for delta compliance. (3) Mms Madd 2 Fs 1 Fsmod Kms Kbox 2 Fsmod 1 Wfree 4.3. Driver Q Driver ring down characteristics are determined by various mechanical and electrical losses that dissipate energy. Mechanical losses include friction losses, mainly in the suspension, and acoustic radiation. Electrical losses include the voice coil resistance, but can also include cable loss and eddy current loss in the magnet structure. Electrical and mechanical system Q s (Qes and Qms) are found in the free air test by measuring three frequency points Fs, F1 and F2. Fs is the point where Zmax occurs (also the zero phase crossing), and F1 and F2 where Z=sqrt(Zmax*Revc). Here are some equations for Q: 60 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

61 THIELE SMALL DRIVER MEASUREMENT Zmax Re Qms Fs* ( F2 F1) Qms * Re Qes ( Zmax Re) Qms * Qes Qts ( Qms Qes) 4.4. Driver Frequency Dependent Inductance (Le is not simple inductance) The next evolution in T/S modeling includes voice coil inductance Le. When Le is added to the standard T/S model, it not only affects response but it also causes Fs to shift upward in frequency. That is, if not taken into consideration during a driver measurement, the true value of Mms, Kms and Rms will be hidden. On yet even closer examination, Le is found to be a combination of frequency dependent resistance and inductive reactance. The tester measures these values first and then backs them out during each test to reveal the true values of Mms, Kms and Rms. The simulator model then uses these parameters to achieve a very high degree of simulation accuracy. Rem Krm * Xem Kxm * Lem Kxm * 2 F 2 F 2 F Erm Kxm Exm1 Resistive component of Le reactance Inductive component of Le reactance Xem Expressed as an Inductor 4.5. Non Linear Suspension and Motor Characteristics Conventional Thiele Small measurements are performed at low signal levels to avoid non-linearity and distortion. Therefore simulation and modeling would only be valid at low levels, conveniently avoiding suspension and motor non-linearity issues. The tester includes an AC analysis of the electrical versus drive level (see Advanced Topics). Note on Suspension Compliance: A careful examination of the AC and DC results will often show a wide range of input levels, usually up to the point where suspension limiting begins, where the suspension mechanical spring constant (inverse of compliance) significantly decreases as the AC test signal level is increased. This effect is much larger than the static changes seen in the DC offset test. Cms (suspension compliance) is significantly dependent on break-in, drive-level and temperature. The underlying mechanism(s) are a function of how the molecular bonds respond to these conditions: The suspension mechanically loosens with usage. This is commonly referred to as "break-in" and is dependent on the excursion incurred (itself a function of frequency and drive level) and duration. The suspension compliance is affected by changes in ambient temperature. The suspension will self-heat due to mechanical energy loss. This is mostly a function of Rms and drive level, though at high drive levels, misbehavior of the suspension can come into play. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 61

62 THIELE SMALL DRIVER MEASUREMENT The AC drive level test relies on the testers ability to measure a very accurate baseline driver model. For example, if the tester can accurately measure Fs variation with drive level, and Mms is assumed to be constant, any variation in Fs must be attributable to variations of the driver suspension compliance Cms. A simple calculation, but the value of Fs must be very accurate. Similarly, Qes variation is an indicator of motor characteristics. Monotonic sinusoid test signals are used to focus energy at a single frequency while tuned narrow band filters are used to reject out of band signals. An additional advantage is that the drive current can then be held constant at all frequencies (the high power port simulates constant current). Using this type of signal, the Low and High power test ports each cover a 100:1 range in drive level or 10,000:1 in power. When combined (e.g. WT2 and WT-Pro), the drive level variation is 10,000:1 and power is 10,000,000:1. Broadband excitation signals like MLS, noise, impulse and chirp are excellent for fast impedance testing, but should be avoided when precision T/S modeling is required. Simply put, these signal types do not provide the same degree of precision or noise immunity. An additional problem is that the drive level density with respect to frequency is not constant. True constant current (or constant voltage) methods will produce better and more reproducible results. Note: The test port directs the load voltage and current to the measurement and reference channels. Matched filters on each channel then provide a complex vector value on a sample-by-sample basis allowing a precise calculation of phase and amplitude. A vector divide then reveals a precise measurement of impedance Z and phase P The TS Tests (What Is Measured) The industry accepted standard drive level for Thiele Small testing is that a low level constant current source should be used to avoid various non-linear effects, but not so low that readings become erratic. The precision of the low power test port exceeds almost all previous hardware methodologies allowing the user to select a wide range of drive levels. The Woofer Tester compression tests have been written to automatically extract T/S models over a continuum of drive levels. As the Woofer Tester sweeps from low to high frequency, it will automatically find and measure woofer resonance and other test points and then automatically calculate the Thiele Small model parameters. As it does this, you will see the test frequency go up and down in ever decreasing step sizes as it zeros in on the point of interest. Tester settings that define accuracy, runtime and the final good enough ratio are all defined in the options pull down menu. The zeroing-in methodology is similar to measuring Fs, Qms, Qes and Vas using a frequency generator, voltmeter (and phase meter if you have one). It is also known as the constant-current method of measuring TS parameters Q and Fs Test (Launch from Q,Fs Test Button) The Q and Fs Test is launched from the control window using the Q, FS Test button on the left. Voice coil DC resistance is measured first, followed by frequency dependent inductance (Rem and Xem coefficients), then Fs and finally the F1 and F2 Q measurement points. When complete, results will be posted to the text area at the top of the window, as well as the Results window. Click Button Middle to start Test Click[-] to shrink text area Click left of the Button Edge to select and view a buffer 62 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

63 THIELE SMALL DRIVER MEASUREMENT Q,Fs Free Air Test DC resistance is found by measuring the complex impedance at two low frequency points, calculating the real component at each frequency and then interpolating to zero hertz. The low power port uses 1 and 2 Hertz as the test frequencies while the high power port uses 5 and 10 Hertz. The 5 and 10 Hz frequencies were chosen since it is unlikely most users will be using DC capable power amplifiers. There is also the option of manualy entering a value for Re. Coefficients for Rem and Xem (frequency dependent Le) are found next by making one or more high frequency measurements. Rem and Xem are then calculated for each test frequency and subtracted from the measured terminal impedance leaving only the effects ofmms, Cms, Rms and Re. The test then moves on to finding Fs by starting a sweep at the SweepLo frequency. If the tester is properly calibrated, and the test environment good, phase should be positive. The sweep then continues upward looking for the first zero phase crossing. The frequency step ratio is then decreased and the direction reversed. The tester uses this methodology to zero in on a value for Fs. Phase is used because the tester can directly measure phase at any test frequency and the phase slope at Fs is steep. Depending on setup, precision can be quite high. The step ratio, bailout and other parameters can be adjusted to trade test time for precision. F1 and F2 are found next by zeroing in on frequency points above and below Fs where: Z 0 Z max Re The test finishes by filling out the rest of the impedance curve. Note: You should periodically check calibration. The low power test port output is a constant current source. Therefore, voltage is proportional to impedance and high impedances will produce proportionally higher voltages. If the required voltage exceeds the available internal voltage the clipping indicator will let you know the tester is producing erroneous results. If this occurs during a test, simply decrease the test drive level and start over. Sometimes it is helpful to run an arbitrary sweep first to roughly determine driver characteristics. Do NOT block or restrict vented pole piece airflow. This will create a tiny sealed box air spring that will shift your measurements. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 63

64 THIELE SMALL DRIVER MEASUREMENT 4.8. Vas Test (Launch from Vas Test Button) VAS testing requires that you modify the free air set up using one of three options. As before, the test begins by clicking in the middle of the Vas Test button. The delta mass or delta compliance methods are most popular, while the third, calculating Vas from a known efficiency, is most useful when creating electrical models for mid ranges and tweeters. Click Button Middle to start Test Delta Mass Easy Nickel Measuring Piston Diameter After entering the piston diameter, added mass or box volume, each of the delta tests is quite simple. The only measurement made finds the modified Fs frequency allowing the remaining TS parameters to be calculated. Similar to the Q and Fs test, the effect of frequency dependent inductance is backed out of the impedance measurement. The known efficiency method is a mathematical reversal of the efficiency equation. Knowing efficiency, piston diameter, Fs and Qes is enough to reveal a full model. Simulated response accuracy will however be compromised if there is any error. This method is most useful when tracking parameter variation of a known driver, or when creating an electrically equivalent TS tank circuit for a crossover design tool. In that case, response is a don t care. Delta Mass The delta mass test is arguably the easiest to perform given that suitable test masses are easy to acquire and almost any solid surface can be used as a test bench. In this test, a mass is added to the driver cone causing the free air resonance to go down. Accuracy is a function of the weight measurement, suspension linearity and the tester s ability to resolve the resulting frequency change. Given that traditionally Thiele Small tests were performed with analog sine sources and voltmeters whose accuracy and repeatability might be questionable, the accepted practice has been that a 25% change in resonance should be targeted. The Woofer Tester is easily more accurate so this can be cut back considerably. The software suggested mass however still honors the traditional 25% guideline. 64 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

65 THIELE SMALL DRIVER MEASUREMENT The Easy Nickel Test Mass method is particularly helpful for acquiring suitable test masses. Newly minted US Nickels have weighed 5.0 grams since 1938! If a lighter mass is needed, the copper coated zinc pennies (after 1983) are 2.5 grams. The following link contains information for US, European Euro, Australian and other coins. Delta Compliance The delta compliance method uses a known test box volume to increase the total stiffness, which then also increases Fs. The quality of the test results, in this case, depend on the accuracy of the box volume calculation (excludes driver displacement, bracing, etc.) and new resonance frequency. The test box should not be stuffed with damping material. Tips: The tester has enough excess precision that a 25% decrease in resonance is no longer necessary. A 10% change is usually sufficient. This is particularly useful when drivers are simply set on a firm surface with gravity acting on the cone and test mass. This deflects the suspension from its zero point. That is, a less than traditionally optimal test mass can be used. Decrease the drive level if the coins begin to jiggle or bounce. Use caution when applying sticky substances to a cone. Stains are possible so you may want to consider using the back side of the cone. Excessive force can also weaken or damage a cone Box Test (Launch from Box Test Button) In-box testing reveals information about the actual alignment of the driver after it has been mounted into a suitable enclosure. When this data is used in conjunction with Thiele Small Simulator the effective box size, tuning and losses are revealed. The simulator will also auto-align itself using this data revealing the actual in-box acoustic response. The file demo2_simulator.woo contains previously acquired data for a 165 mm (6.5 inch) computer sub-woofer. In-air data was also taken showing how the simulated and actual response will eventually align. See the demo section for additional information on this demo. This test finds the impedance peaks and minima where the zero phase crossings occur. The simulator then uses this information when an auto-align is requested. If the alignment is successful, the resulting simulated response will closely match the actual response. See the simulator section for information on using the auto-align feature Making A Suitable Test Bench Electromechanical devices such as loudspeakers exhibit impedance and phase variation whenever a resonance occurs. Driver resonance is created when the kinetic energy (moving mass) and potential energy (suspension spring) are in balance. Electrical and mechanical loss (resistance) then limits the magnitude of the peak by bleeding off some of the stored energy. The position, height and width of the resonance peak are then converted to parameters suitable for box simulation. To accurately measure these parameters you will need to create a suitable test bench where outside influences do not disturb each test. There are two basic setup configurations that you can use, each with pro's and cons. Vertical Orientation (Cone is facing upward) Simply laying a driver on the floor and performing tests is definitely easy to do, but other suitable stiff and heavy surfaces may be preferred. The tester can help you identify suitable test surfaces by examining the phase of an arbitrary sweep. An unsuitable surface will show a in the phase plot, and if very bad, a kink in the impedance plot. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 65

66 THIELE SMALL DRIVER MEASUREMENT Some floors (even concrete) will resonate strongly. This has to do with the mass to stiffness ratio and loss factors for the particular test point. The center and corners of a table would be an example where the leg at the corner substantially increases stiffness. The corner of a large and heavy speaker cabinet will often work quite well. An alternative method is to suspend the driver using a bungee cord. In this case the resonance is set by bungee cords spring stiffness and the mass of the driver resulting in a resonance of about 1 Hz. Driver Fs is ultimately affected because this forms a two mass resonant system. That is, since the driver is no longer firmly attached to a solid reference, Fs will shift slightly upward. If you want to be more precise, increase the overall hanging mass by adding weights to the driver. Suitable test masses are also easy to find. In the United States, Nickels have been manufactured with the same composition since 1938 and weigh 5.0 grams. Note: A Blocked Vented Pole Piece will form a small sealed enclosure An 18 Driver suspended from a Telescope Tripod Additional information can be found at this link Horizontal Orientation (Cone is facing sideways) In practice the suspension rest position depends on all of the forces acting on it, and this comes down to orientation and gravitational forces. Suspension non-linearity and mechanical offsets also affect the measured suspension stiffness. Mounting a driver sideways solves this problem (no suspension sag), but a different kind of test jig and some kind of sticky test mass will be required. Mounting the driver between two old (massive) speaker boxes is one solution, or a sling can be improvised that mounts the driver in free air, similar to the example shown above. You will also need to use an accurately measured mass of clay or soft caulking. Keep in mind that placing clay on the front side of a paper or composite cone can stain or damage the cone Driver Break-In Driver break-in is often necessary to mechanically loosen newly manufactured suspension components. For example, the chemical and mechanical bonds that are formed in a resin impregnated cloth spider can change dramatically as they become stretched. It therefore makes sense that to design a box you will want to get measurements that are based on how the driver will be ultimately used. Break-In can be a fast or slow process depending on the driver. Break-in is most easily accomplished, and with far less voice coil heating, when the test frequency is set to the driver free air resonance, Fs. At Fs mechanical energy is alternately stored as inertial moving mass or as potential energy in the suspension spring. High excursions are also possible at lower frequencies but with higher power requirements. As the driver breaks in, Fs will decrease and or excursion will increase. Frequency and amplitude can be controlled using the control window but it is strongly advised that you control volume at the amplifier instead to avoid accidental overdrive. Tip: You may experience dropouts if the operating system becomes overloaded. Adding a pre-amplifier or integrated amplifier with a volume control may be desired since this will allow you to set the tester output level to maximum. You can then set the maximum level manually. 66 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

67 THIELE SMALL DRIVER MEASUREMENT! WARNING! MECHANICAL OVERDRIVE IS PARTICULARLY EASY TO DO WHEN FREQUENCY IS SET BELOW FS. FS ALSO TENDS TO DROP WITH BREAK-IN. Begin with a moderately low level for the first 15 minutes and then slowly increase level. Mechanical resonance should be visible to the naked eye. If not, occasionally go back and retest the driver. UNDER NO CIRCUMSTANCES SHOULD THE LOW POWER TEST LEADS BE ALLOWED TO TOUCH OR CONNECT TO THE AMPLIFIER Arbitrary Impedance Plots (ARB1 and ARB2 Button) Arbitrary sweeps do not perform any kind of test. These simple impedance plots over a frequency range are useful when comparing drivers, crossovers inductors and more. The real time impedance modes and data buffer copy options are particularly useful when working with these buffers. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 67

68 DRIVER SIMULATION AND BOX DESIGN 5. DRIVER SIMULATION AND BOX DESIGN 5.1. Simulation and Physical Box Design A simulation is initially performed using the driver s measured Q/Fs and Vas test data to find a suitable box alignment. In theory, the effective box volume and tuning will align with the physical but this is rarely exact. In the final box alignment step, the in-box electrical impedance curve will be used to auto-align the simulator revealing the true response. The simulator is controlled from the TS Entry Window and its output is viewed from the Overlay window allowing the simulated output to overlay measured impedance and response data either passively or in real time. The copy buffers option can also be used to save and compare a number of response runs. View Simulated Response (SimRX) View Simulated Impedance (SimZP) Overlay Window TS Entry Window Tip: If you do not see an output, the drive level may be set to zero. Check 1W reference level to bring the display back. 68 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

69 DRIVER SIMULATION AND BOX DESIGN 5.2. First Pass Simulation Box design begins by copying the test environment driver parameters (Q/Fs and Vas test data) into the simulator using the Copy- >Simulator button. A dialog box will then open asking you if you want to copy the Q/Fs, Vas and Box data. Check the Q/Fs and Vas options leaving the Box data option unchecked. Box data is valid only after testing and converting electrical box data. There is no box to test yet, so this is irrelevant (used later). An initial good starting point is to start with a sealed box by clicking on the Sealed radio button and setting or adjusting the box volume equal to Vas. Next, set or adjust all of the Q loss parameters to a value higher than At this point, the box type, volume and tuning are iterated until a satisfactory alignment is found. An initial port tuning for a vented box will be the driver Fs. Note, that in a vented box a low port tuning frequency is essentially the same as a sealed box. At this point, you can evaluate acoustic response and phase, cone displacement, port air velocity and expected electrical characteristics. You can also enable and simulate pipe resonance effects and room/car pressurization effects. What to copy Want to View Simulated Impedance Box Adjustments What to view Copy Test Data to the Simulator Auto-Align (use later!) Tip: A very large box volume is essentially free air and can be used to compare the simulated and measured impedance plot from the Q/Fs test. Using the frequency dependent inductance RemXem model will yield the best possible results. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 69

70 IN-BOX ELECTRICAL TESTING 6. IN-BOX ELECTRICAL TESTING 6.1. Box Electrical Test A very useful simulator feature is the ability to compare simulated and measured in-box impedance curves. When the two are aligned, the simulators box volume and tuning are said to be the effective box volume and tuning, but more importantly the simulated response will accurately reflect the measured response. In-box electrical impedance is measured by connecting the speaker box (without the crossover) to the low power port and clicking on the Measure Box button. The Box test finds three points of interest, Flo, Fmin and Fhi and from these computes Fm, Fsb, Ha and Alpha. Fm is the port tuning frequency. Fsb, Ha and Alpha are parameters that are used with charts that are found in various other simulators and publications. Points of Interest Flo, Fmin, Fhi Begin Box Test 6.2. Auto-Align Vbox (Align Simulated and In-Box Impedance) The Auto-Align Vbox button will align the simulator to the measured data. By aligning the simulator, not only will the electrical data be similar, but also the response. Initially you will be asked to choose between two alignment methods that constrain the estimated box volume, or the original TS model. Auto-Align method Constrain Cms This method finds a best fit by finding a box volume and tuning while holding the TS model constant. This method is generally quite accurate when the original TS model is accurate and the box volume is comparable or smaller than Vas. Under these conditions the position of the upper resonance peak is dominated by the box air compliance. 70 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

71 IN-BOX ELECTRICAL TESTING Constrain Vbox When box volume is large compared to Vas, the driver compliance dominates the box air spring compliance. In this case small variations in Cms would cause significant errors in the estimated box volume method. This method is similar to the Vas from a vented box method, except that by pre-measuring the TS parameters the motor constants and driver mass are well known. This method is also preferred when the drive level is substantially higher than when the TS model was originally made, or if a new un-broken in driver is used. This is because driver suspension Cms is highly dependent on break-in, drive level, aging or heating. Auto Alignment Complete A status message is then shown letting you know the parameters found before committing them to the simulator. If necessary, adjust the box volume and tuning, followed by the Q loss parameters. Auto-Align status Peaks and Zero crossings line up 6.3. Simulated versus Measured In-Air Response If the simulator electrical alignment is successful, the final step is to view the response. This example compares the actual in air response using a Woofer Tester Pro. The woofer and port in this case are on opposite sides of the enclosure allowing the driver response to be shown in isolation. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 71

72 ADVANCED THIELE SMALL TESTING: AC & DC COMPRESSION 7. ADVANCED THIELE SMALL TESTING: AC & DC COMPRESSION 7.1. AC Compression Testing To measure suspension non linearity versus AC level select the following menu option: Tests Linearity Tests Measure TS versus AC Level This test requires a precision low power TS baseline before the test begins (this is not automated and must be done first). This baseline establishes the driver s moving mass, Re and motor constants. The driver is then configured for open air Q/Fs testing. As the test progresses, drive levels are ratio-metrically decreased from 100% to 1% resulting in changes in Fs, Qms, Qes and Zmax. By then using the established Mms as a constant, the only parameter that can effect Fs is compliance Cms. Similarly, variations in Qms can only be attributed to changes in Rms and Qes by changes in BL. The process is continued resulting in a graph like the one shown below. The X-Y plot window is then used to view and compare the Thiele Small Parameters. Example Fs for this driver dropped from a high of 84 Hz at 29 A to 58hz at 100 ma (usable Xmax). Fs then climbed steeply with increasing drive level as the suspension bottomed out. This example clearly shows why different test levels produce different results for the same driver. The WT2 and WTPro test discontinuity was caused by the suspension 'settling' as the initial test level was quite high and was sufficient to begin bottoming the suspension causing considerable mechanical noise. Note On Suspension Compliance: Suspension non-linearity is sometimes intentional. An example would be controlling displacement at maximum excursion. This test shows a continuum of TS parameters dependent on drive level. This is known as stiction. What is not shown is how the suspension changes with time, break-in and temperature. 72 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

73 ADVANCED THIELE SMALL TESTING: AC & DC COMPRESSION A careful examination of the AC and DC results will often show a wide range of input levels, usually up to the point where suspension limiting begins, where the suspension mechanical spring constant (inverse of compliance) significantly decreases as the AC test signal level is increased. This effect is much larger than the static changes seen in the DC offset test. Cms (suspension compliance) is significantly dependent on break-in, drive-level and temperature. The underlying mechanism(s) are a function of how the molecular bonds respond to these conditions: The suspension mechanically loosens with usage. This is commonly referred to as "break-in" and is dependent on the excursion incurred (itself a function of frequency and drive level) and duration. The suspension compliance is affected by changes in ambient temperature. The suspension will self-heat due to mechanical energy loss. This is mostly a function of Rms and drive level, though at high drive levels, misbehavior of the suspension can come into play DC Compression Testing To measure suspension non linearity versus DC level select the following menu option: Tests Linearity Tests Measure TS/THD versus DC Bias DC offset testing also requires a precision low power TS baseline to establish the driver s moving mass, Re and motor constants (this is not automated and must be done first). The driver is then configured for high power open air Q/Fs testing. This dialog box is used to set the test options. Pick Real-Time Impedance Signal Type A real-time impedance mode is preferred in this test. However, different signal types may produce differing results. The options here provide both low and high frequency energy. It should be noted here that real time signal types are actually not symmetric. Choose if Le and THD are to be measured Le and THD can optionally be measured with higher precision using a sine wave. Le is measured at 4x the Fmin zero phase crossing. Choose AC and DC drive levels AC and DC levels are entered as a percentage of amplifier drive when the test reaches 100%. The indicated levels are found by measuring the amplifier gain using an AC signal. As the test progresses, DC levels are increased from 0% to 100% of the selected maximum level in increasing positive and negative steps. This allows the test to be stopped or the amplifier shut down. As the levels increase, changes in Fs, Qms, Qes and Zmax are detected using a technique similar to that in the AC compression test. The X-Y plot window is then used to view and compare the THD and Thiele Small Parameters. DC bias testing also produces a table of TS parameters, but these parameters are derived from a small signal analysis at a particular DC bias point. These parameter variations reveal useful information about driver construction such as mechanical offsets, suspension and motor linearity with excursion. This example shows an offset in the suspension as well as the voice coil. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 73

74 ADVANCED THIELE SMALL TESTING: AC & DC COMPRESSION As the DC bias test is run, a sine wave signal can be specified that either tracks Fs or remains fixed at the frequency when the test begins. In this case, harmonic analysis also reveals mechanical offsets. However, keep in mind that the raw data provided in this case is a cumulative effect of all DC and small signal AC parameters. In the absence of an infinite baffle, this test can be performed in an open-air test jig using a closely spaced microphone. The absolute level of Fs and its harmonics are therefore typically low due to backside cancellation. A trend at a particular frequency will however still be valid. The XY window can also be configured to view SINAD, or Signal versus Noise and Distortion. Normally, SINAD measurements are given in decibels as the amplitude ratio of the fundamental signal to all remaining noise and distortion. The XY plot version of this algorithm uses the measured amplitude to create a sine wave at the fundamental frequency that is then subtracted. If the remaining signal is then plotted against the fundamental, data views of NAD vs Magnitude, Lissajoux and oscilloscope are possible. 74 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

75 ADVANCED THIELE SMALL TESTING: AC & DC COMPRESSION Note: the Xn / int(xn) button numerically integrates the data. Microphones produce signals that are equivalent to pressure, that in turn is proportional to mass-air velocity. Integration shows mass-air displacement. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 75

76 INTERACTIVE CROSSOVER DESIGN (ICD) 8. INTERACTIVE CROSSOVER DESIGN (ICD) CAUTION: Failure to use the ICD Tool Properly can result in damaged drivers or amplifiers. Refer to Section 8.6 Tweeter Protection Circuit for more information. With the Interactive Crossover Designer (ICD), you can design both, physical and arbitrary crossovers. ICD simulates soldered together physical crossovers using a SPICE circuit file (*.CIR extension) that defines the connections, RLC (resistor, capacitor, inductor) values and driver impedances (as ZMA files). The circuit file itself can be input by hand, or captured using a number of third party schematic capture programs. Additional features include a driver protection circuit that is virtually removed in software as well as a time delay option that simulates changing the driver to driver front to back spacing. Arbitrary filters are defined solely on a mathematical basis using *.XVR extension files. These files have uses ranging from simulating DSP and other electronic crossovers to defining hypothetical conventional crossovers that are based on driver measurements. The format itself is similar to the FRD export format except that two channels are defined and in this case the data is an input that defines the tester s output signals. In this case the simulated outputs represent adirect connection from the amplifier to the driver. The ICD simulator allows you to test crossover designs on your desktop before having to physically build them. Simulated and actual crossovers can then be tested in real-time by moving around the room with the microphones and seeing the responses on the screen. Notes: Signal I/O is only active when the WT Control window is open. The Math Reference is always used when calculating ICD response 8.1. A Physical Versus Simulated Crossover Design Example We will use an example to show the process of designing a physical crossover. The process consists of the following steps: 1. Create an electronic circuit file. There are free, third party applications available for designing electronic circuits. Linear Technologies LT SPICE (or SWITCHER CAD) is used in this example but other schematic capture tools will also work. This step is described in detail in the next section. 2. Convert the electronic circuit schema into a Berkeley Spice net-list file. Conversion as well as language syntax are described in section 8.3 Circuit File Syntax. 3. Add the speaker s characteristics via a ZMA file. This is described in section Simulate the crossover. The sample crossover we build in this section was constructed from a 6.5 woofer and 20 mm polycarbonate tweeter. Terminal block connections for each driver were then positioned on the back of the enclosure coinciding with connectors in the removable physical crossover. Each component was carefully measured and then input into an ICD circuit file. 76 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

77 INTERACTIVE CROSSOVER DESIGN (ICD) Figure 8-I - Front, Back and Back with Crossover Installed views of the Test System Note: This tester will measure resistors, capacitors and inductors very accurately. There is also an air core inductor design wizard in the Tools pull down menu The following ICD circuit file was created to model the physical circuit. Figure 8-II File Contents of 'DEMO.CIR' ******************************************************* * 2nd order high pass, 1st order low pass * * (1) 615uH (2) * o----l * * C1 * === 6uF +-+/ Z1 * (3) ) DEMO_WO.ZMA * +-+\ * R1 4 * (0) * o * * (1) 3uF (4) 5ohm (5) * o R * C2 * +-+/ Z2 * L2 R3 ) DEMO_TW.ZMA * 274uH 4ohm +-+\ * (0) * o ******************************************************* Vsig 1 0 AC 10 ************************** L uH RL C uF R Z1 2 0 "demo_wo.zma" ************************** C uF L uH RL R R Z2 5 0 "demo_tw.zma" 5 *************************** Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 77

78 INTERACTIVE CROSSOVER DESIGN (ICD) Notes: To ensure accuracy, each component was individually measured using the low power test port. This also provided additional information like the DC resistance of the inductors. Channel balance should be adjusted for maximum accuracy. Driver compression effects will cause the impedance ZMA file data to be drive level dependent. For maximum accuracy use a test level and test method that is similar to the ICD simulation. Additional modeling elements might include cable effects Tip: You can use the balance control of your amplifier to simulate level padding Figure 8-III shows the final result; a near perfect amplitude and phase match between the physical and simulated crossover was achieved. Figure 8-III - Comparison of Simulated & Physical Crossover Response (2.5 db/division scale) No frequency or time domain smoothing has been used Creating an Electronic Circuit File This section describes how to use third party schematic entry tools to generate an electronic circuit file and compatible Spice net-list. Examples are given for a professional tool, OrCad by Cadence, as well as a free tool, Linear Technologies LT-Spice/Switcher-Cad that can be downloaded from To be compatible with as many schematic capture tools as possible the ICD compiler follows a the original Berkeley Spice component format. That is, components are considered simple and do not support loss, parasitic or other additional parameters. Enter these as extra circuit components. 78 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

79 INTERACTIVE CROSSOVER DESIGN (ICD) Linear Technology: LT-Spice/Switcher-Cad (A PSpice Derivative) LT-Spice/Switcher-Cad is primarily intended for simulating circuits made from devices made by Linear Technology Corporation. In normal operation, net-lists are internally generated and compiled using the PSpice compiler. An ICD compatible net-list can, however, be exported using a copy and paste after viewing the SPICE netlist. A driver schematic symbol does not exist, so you will need to create one, or download the.asy from Notes: The LT-Spice net-list view box is not an editor, but you can select and copy the text using ctrl-a to select all text and ctrl-c to copy it to the windows copy buffer. Use the paste command or ctrl-v to copy the net-list into the ICD editor. Zn Driver variables are passed in by right clicking and modifying the model parameters. They can be entered on one line, or multiple lines. Use the Visible check mark to make them visible. Spice Net-List (copied from view window) Listing for SPEAKER.ASY file * D:\wtpro\ltspice_demo_xo.asc L1 N µH C1 N005 N µF R3 N002 N003.1 V1 N001 0 Z1 N003 0 demo_wo.zma 0.1 Z2 N006 0 demo_tw.zma 5.1 L2 N001 N µH C2 N003 N µF R1 N R2 N R4 N006 N backanno.end Version 4 SymbolType CELL LINE Normal LINE Normal LINE Normal LINE Normal LINE Normal LINE Normal RECTANGLE Normal RECTANGLE Normal RECTANGLE Normal RECTANGLE Normal WINDOW Left 0 WINDOW Left 0 SYMATTR Value 8.ZMA SYMATTR Value2 0 SYMATTR SpiceLine 0 SYMATTR Prefix Z SYMATTR Description Driver (Left=0 Right=1) PIN 0-48 NONE 0 PINATTR PinName + PINATTR SpiceOrder 1 PIN 0 64 NONE 0 PINATTR PinName - PINATTR SpiceOrder 2 Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 79

80 INTERACTIVE CROSSOVER DESIGN (ICD) OrCad Capture For Windows: Schematic Entry Tool OrCad Capture is widely used in the EDA industry for generating schematics and circuit boards. A SPICE simulation option is also available. The spice net list is generated using the generate Net-List option. OrCad allows a great deal of latitude when specifying component designators, allowing a wide range of symbols. Check the netlist output for compatibility. Note: To facilitate mixed digital, analog and RF designs, OrCad uses multiple ground symbols. These symbols need to be connected to the SPICE ground by the user. Click on the ground symbol and rename the name to GND to create a compatible netlist. Use the first circuit you build this was as a template. * Revised: Sunday, October 21, 2007 * Revision: NA * R R L uH C uF V Z demo_wo.zma 0 Z demo_tw.zma 5.END 80 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

81 INTERACTIVE CROSSOVER DESIGN (ICD) 8.3. Circuit File Syntax ICD Virtual crossovers are defined using Berkeley Spice formatted ASCII files containing a net-list. Each line in the net-list contains a device statement followed by node connections and a component value. Each node represents a wire soldered together in the physical crossover. The ICD compiler converts this list into left and right test signals that then simulate the effect of a physical crossover. With care, the resulting crossover will often be close enough that little or no additional tweaking of the physical circuit will be needed. ICD eliminates the need to physically build a crossover greatly reducing time, effort and money spent. Example The following example shows an input signal connected between node 1 and ground (node 0). A capacitor C1 then connects the input signal to the tweeter that is connected between nodes 2 and 0. Likewise, inductor L1 passes a signal to the woofer connected between nodes 3 and 0. Since tweeter and woofer loads are not simple resistors, highly accurate simulations are only possible when these loadsare defined using ZMA impedance files. ZMA files are created from either the low or high power port using the FileExport option. If you do not have a ZMA file, you can replace the FILE.ZMA argument with a numeric value. A driver protection circuit consisting of a series resistor and diode clamps can also be included with each load (this is highly recommended). Simply add the resistance value after the ZMA statement and the signal will be modified resulting in an accurate simulation, but with a well protected driver. ****************************************************************** * (1) C1 4uF (2) * protection circuit * L1 2mH * /\/\ (3) * Z=tweeter.zma * Vsig + Z=woofer.zma * * * /// /// /// ****************************************************************** Vsig 1 0 AC 1 * Input signal C uF * High pass capacitor L mH * Low p ass inductor Z1 2 0 "tweeter.zma" 5.0 * Five_ohm+diode clamp protection Z2 3 0 "woofer.zma" * No protection (direct connect) ******************************************************************! Note: A ground point must always be defined and it must always have the value zero 0 Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 81

82 INTERACTIVE CROSSOVER DESIGN (ICD) 8.4. ICD Components and Devices ZMA Impedance Files The left woofer and right tweeter loads interact with the ICD defined circuit using ZMA files to define the load impedance and phase. ZMA files can be captured and saved using any of the available test signal types and impedance test modes. Capturing ZMA data using a test signal type and drive level similar to the one used during the ICD simulation produces the best results. This is because driver and box parameters often shift with drive level. ZMA files are exported from the File pull down menu: File Export Export buffer to ZMA (impedance) Z1 and Z2: DRIVER LOADS The left and right Zn driver load statement connects the ZMA or constant value load to the simulated crossover. On the physical side, a driver protection circuit in the form of a series resistor and diode-clamping network can be specified. Usage: Z1 Z2 ZL ZR N0 N1 value impedance.zma [Rseries] Examples: Z * Right channel, constant 6 ohm load, Rs=5 Z2 5 6 "tweeter.zma" 4 * Left channel, ZMA defined load, Rs=4 Notes: Zn: ZL and ZR connect to the left and right respectively. However, some schematic entry tools will insist on a numeric value here. In that case, even values of n (a number) will connect to the left channel and odd values to the right A front to back time adjustment can be set in the ICD control window Amplifier Signal Source A single signal source must be defined representing the crossover input from the amplifier. The signal source is then mathematically swept from DC to F=SampleRate/2, and the network evaluated filling in the Fourier spectrum. One of the node numbers for the signal source must be '0' (ground). Usage: Vn N0 N1 [value] Example: Vsig 0 1 * signal is connected to nodes 0 (ground) and 1 Resistors, Capacitors and Inductors: RLC components are specified as simple devices. Parasitic and loss effects such as inductor DC resistance are entered as separate components. Usage: Rn N0 N1 value Cn N0 N1 value Ln N0 N1 value Examples: R * 6.1 ohm resistor connected nodes 0 (ground) and 3 C uF * 4.2uF capacitor connected nodes 3 and 4 L E-6 * 820uH inductor connected to nodes 4 and 0 (ground) 82 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

83 INTERACTIVE CROSSOVER DESIGN (ICD) 8.5. ICD and XVR Editor and Simulation Control The ICD and XVR Editor and Simulation control window is similar to the results editor except for the addition of simulation controls. Simulations are allowed for *.CIR and *.XVR extensions only. ICD circuit files can be created manually, or using third part schematic entry tools. Use the View Interactive Crossover Designer menu option to open the ICD edit window. Standard Edit Controls Start/Stop Simulation Negate or Delay Left Channel View Real-Time or simulated XVR file loaded (ungrayed text) Offset (m or ms) Adjusts the effective acoustic distance from the left to right channel either by meter distance or millisecond value. Negate - Negates the left channel output and simulates swapping of the driver connection ShowZP Calculates and displays system level impedance using the SimZP buffer. The SimZP buffer is used, since this buffer is updated with TS simulated data. If you want to keep this data, copy it to another buffer or file. View Measured - Shows db response of line or microphone signal View Modeled - Shows db response of mathematical model (no noise) 5 and 3 Column XVR Format Sets the format of your XVR file Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 83

84 INTERACTIVE CROSSOVER DESIGN (ICD) 8.6. ICD Driver Protection Circuit Circuit Analysis Diodes D1 through D8 clip excessive voltage and limit the power that can be applied to the load. In this example a high wattage 2.5-ohm resistor (and F1, which is a light bulb) becomes the amplifier load should diode clamping be required. Fuse F1 is chosen to match the amplifier or resistor s maximum safe current and power. A similar circuit appears on the left side of the clamp resulting in a nominal 5.0 ohms series resistance. Light bulbs also make suitable fuses. In this case the positive temperature coefficient of the filament resistance keeps resistance low until a substantial current begins to flow. Tweeter test signals usually have a low average energy so the bulb has little time to heat. A 12W (12V, 1Amp) automotive bulb usually works nicely. Simply account for the filaments cold resistance by measuring the total resistance. Diode clamping will not occur under normal test conditions and the current in F1 and F2 is low. Many variations of this circuit are possible. You could for example easily add LED clipping indicators Arbitrary Crossover Design (XVR files) XVR response files can be used to pick the initial crossover frequency point, help find an initial filter topology, or to create a totally arbitrary response. In particular, XVR files set the amplitude and phase at a particular frequency. The response of a DSP or line-level electronic crossover can therefore be simulated using an XVR file. The Interactive Crossover Designer uses an XVR file format, which is similar to the FRD format, except that both, left and right channels, are defined. In addition, the data defines both outputs and not the measured response of an input. That is, XVR files are the reverse of FRD files. As an example, consider that when a crossover topology is initially chosen it is often desirable to know the front to back acoustic distances between driver pairs. This is found using an arbitrary response filter by setting up a sharp transition and looking at the phase difference. Furthermore, an amplitude step would also indicate the need for some kind of padding circuit. XVR File Formats The ICD handles two XVR file formats shown in Figure 8-IV below. The first format, shown on the left, contains 5-columns with left and right channel data on the same line with a common frequency. This format is suitable for automatic export and read-in to a spreadsheet for example. The second format contains only 3-columns of data and is similar to two concatenated FRD files with LEFT and RIGHT directing the response definition to each output. In this case, the left and right frequencies do not need to match. Response for undefined frequencies points are interpolated from the data points that are known. 84 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

85 INTERACTIVE CROSSOVER DESIGN (ICD) The two examples below are identical with both abruptly transitioning the woofer and tweeter test signals at 5 khz. Figure 8-IV - '5KSHARP.XVR' 5 column and 3 column formats ************************************ * High + low pass filters * Freq db(l) ph(l) db(r) ph(r) ************************************ *************************** * High + low pass filters * Freq db phase *************************** RIGHT END *************************** LEFT END Notes: Frequencies must be >= 1Hz and < SweepHi Frequencies must increase The outputs are rescaled (same scale both sides) XVR Overlay of Woofer & Tweeter Response The first picture below overlays the ARB1 and ARB2 buffers by using the amplifier balance control to collect and capture data for each channel. The second image then shows the acoustic summed response captured to the BOX buffer by setting the balance at center. The abrupt phase change of about 100 indicates an acoustic distance of D=(Angle/360)*(C/Frq) = 32mm between the drivers. Note: The delta effective distance between drivers is a function of the phase angle at a particular frequency. In this sense, both drivers are variables also making the effective delta distance a function of frequency. However this information is still useful input for crossover design. Figure 8-V - Frequency Defined Response filter (XVR) Overlay of Woofer and Tweeter Response Woofer Balance-LEFT Tweeter Balance-RIGHT Abrupt 3 khz transition with two responses within this sharp edge Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 85

86 INTERACTIVE CROSSOVER DESIGN (ICD) Acustic sum woofer+tweeter Balance-CENTER Phase step (acoustic distance) The last image shows the final overlay of the woofer, tweeter and acoustic summation using an ICD designed crossover. The crossover point was shifted up to 5 khz to get a smoother transition. 86 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

87 HARMONIC & SINAD DISTORTION MEASUREMENT 9. HARMONIC & SINAD DISTORTION MEASUREMENT Harmonic Distortion (HD) is a measure of the harmonic content produced when a single frequency is applied. Therefore, this test is performed using a sine wave sweep. HD analysis is enabled in the Setup Control window Distortion tab. Six harmonics are measured with results being plotted to any three user selectable buffers (picking buffers >7 is suggested since the lower 7 are for primary tests). You can either view HD directly from this window, or if a sweep is used, from the Overlay window. Primary and secondary (IMD) frequencies Select Distortion Test And data destination buffers Select Buffers for display 2 nd through 6 th Harmonics Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 87

88 TEST TOOLS AND CALCULATORS 10. TEST TOOLS AND CALCULATORS The tools and calculators mentioned in other chapters are described in greater detail in this chapter Aligning the Tester For Measuring Capacitors The low power tester measures a very wide range of resistors, inductors and capacitors when properly configured. Since the tester s constant current output produces an output voltage proportional to load impedance, some care must be taken to prevent impedance from going to infinity, or the drive level must be decreased. Decreasing the drive level is a suitable solution for resistors and inductors, but capacitors require an extra shunting resistor. Start by switching to the LoZP mode using a sine test signal. Then add a 10k shunt resistor using a spare banana jack, which can be easily plugged into or removed from the tester. Next, click on Tools Capacitor Measurement Setup to open the alignment dialog box. Finally, click on the Auto Complete button. The alignment algorithm then automatically measures the shunt resistor, followed by the internal tester capacitance. Larger capacitors can be also be measured but they must be non-polarized. In these cases, a smaller shunt resistor (like kohms) will work. The goal is to have a shunt capacitor that is sufficiently large in value that the full dynamic range of the tester is used for the capacitor being measured. Large capacitors have lower reactance and therefore can be measured with lower shunt resistance values. Since capacitive reactance is inversely proportional to frequency, capacitors are typically not tested at low frequency. Testing large capacitors at high frequency can also lead to a problem where the capacitors reactance begins to be small compared to the test lead reactance that is inductive. These effects are easily seen when a swept test over a suitable frequency range is used. 88 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

89 TEST TOOLS AND CALCULATORS This table shows the results of measuring a 100pF+/-2%, 1200pF+/-2%, and 2.0uF+/-2% capacitors. The 2uF capacitor was measured plugged directly into the tester as well using the supplied test leads. Note: Wiring can have a measurable effect. Frequency 100pF+/-2% 1200pF+/-2% 2uF+/-1% 2uF+/-1% short leads Long leads E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E-06 Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 89

90 TEST TOOLS AND CALCULATORS Zobel Calculator Zobel compensation networks negate the rising driver impedance caused by series inductance Le. Le is, however, a function of frequency and even includes frequency dependent resistance. In addition, the target impedance might not be Re, but an arbitrary value. The Zobel calculator will do this and is launched from a Simulator Entry window button. Begin by selecting the frequency you want to compute a solution at. The simulator is then called to compute the uncompensated impedance and phase. If you then enter a desired impedance and phase, the tool will compute the required RC Zobel. You can also optionally have the simulator calculate the resulting impedance plot. Note: SimZP is recomputed each time the simulator is called, so the data will not persist. If you want the resulting impedance plot to persist, use the ARB1 buffer destination. Overlay Original and Zobel compensated Impedance plots Select Frequency and Target Impedance Start Zobel Calculator Note: Zobel compensation affects impedance, not driver terminal voltage. Changes in response are a function of the crossover. Refer to the section on Interactive Crossover Design for designing and building crossovers. 90 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

91 DEMONSTRATION FILES 11. DEMONSTRATION FILES The demonstration files in this section are found on the installation disk or from the web. Demo files showing the WT2 are compatible with the Speaker Tester and Woofer Tester Pro Conducting a Simple Electrical Test 1. Connect the test leads to the Woofer Tester 2 and to the driver as shown below. 2. Double-click the SL icon on your desktop to launch the Tester application 3. Next, open the demo file by selecting File Woofile Load Woofile from the menu bar. In the file open dialog box that pops up, select the file called demo1_simple_zp.woo and click Open Your screen now displays the WT Control window and you can hear the Chirp-Sine signal the tester is using to conduct the electrical test: Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 91

92 DEMONSTRATION FILES Thiele-Small Simulator & Box Design Demonstration To use the Thiele-Small Simulator, leave the tester connected. The WOO file used in this demo contains previously acquired data for a 165 mm (6.5 inch) computer sub-woofer. Additional in-air data was also taken to show how the simulated and actual response will eventually align. 1. Open the demo file by selecting File Woofile Load Woofile from the menu bar. In the file open dialog box that pops up, select the file called demo2_simulator.woo and click Open 2. Your screen now displays the T/S Entry window on top and the Overlay window on the bottom. (You might have to re-tile them to fit your screen via the Window Tile Horizontal menu option). The left half of the T/S Entry window displays the Thiele Small parameters of a speaker in this case a 6 name brand computer speaker. The right half of the window contains the box values. 92 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

93 DEMONSTRATION FILES Driver T/S parameters Box values Buffer select Box types Simulation models Acoustic Simulation Output What-if Simulated Response What-if Simulated Acoustic Phase 3. At this point, the TS parameters are preloaded into the simulator environment and you can begin modifying the box volume, tuning and port dimensions to achieve the desired response. This is the initial what-if stage of designing a box. As you make these adjustments and turn on and off various display options, you will see the response and other parameters change. For example, you can change the blue line to display various acoustic simulation outputs such as simulated acoustic phase, Xmm or velocity of the driver, port or combined speaker response. 4. Next, disable the SimRX curve and enable SimZP by clicking on those buttons above the graph in the Overlay window. The display changes to show the expected impedance curve of your what-if design. 5. The next steps are to physically build the box based on the values from your simulated box design and measure the in-box electrical data. To measure the in-box electrical data, reconnect the speaker (box with driver installed) to the Woofer Tester and click on the Measure Box button in the WT Control window (see also Chapter 6). The tester then generates data that will be used in the Simulator Auto-Align feature described next. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 93

94 DEMONSTRATION FILES 6. After building the box and measuring the in-box electrical characteristics, you don t actually know if your physical box design has measured up to your what-if design. Reasons for this include potential box dimension or port size errors, forgetting to include the volume displacement of the driver/bracing, or an interaction of the port or driver to the cabinet walls. To find out how closely your new box matches your previously simulated box design, click the Auto-Align button in the T/S Entry window. This finds the effective box volume, tuning and loss. A dialog box then tells you what the effective volume and tuning need to be in order to make the curves match. By clicking OK, the effective box volume and tuning are loaded into the simulator and the new simulation reflects the actual response of the speaker. If this is not what you had originally intended in your what-if design, you would now need to make modifications to the physical box. Results This first image shows the What if (yellow) and Effective (pink) response from aligning electrical measurements. The next picture shows the simulated and actual driver response (using a microphone) with a deep notch at the port tuning frequency. In this case, the port was acoustically isolated by being physically located on the back side of the cabinet. 94 Smith & Larson Audio Woofer Tester Pro User Guide v1.1

95 DEMONSTRATION FILES Demonstration of an In-Air Test To conduct an In-Air test, leave the current driver connected. The low power port acts as a low power amplifier and this setup is sufficient to see the response. 1. Open the demo file by selecting File Woofile Load Woofile from the menu bar. In the file open dialog box that pops up, select the file called demo2_simple_in_air_response.woo and click Open 2. Your screen displays the WT Control window in the front and the Setup Control window in the back. The graph displays the response of the speaker. Woofer Tester Pro User Guide v1.1 Smith & Larson Audio 95