Loudspeaker Active Spectral Divider Design Project

Transcription

1 Loudspeaker Active Spectral Divider Design Project paste in team picture here

2 Project Report Evaluation Team ID: 6 Team Member App Tech* TOTAL Jing Zhao Jose Manuel Conejo Matt Cavenaile Maximum Points Possible * technical content, writing style, professionalism, clarity, completeness Instructor comments:

3 TABLE OF CONTENTS Abstract Introduction Design Narrative Results References 7 Appendix A: Activity Logs 8 Appendix B: Schematic (Analog) or Tuning Parameters (DSP) 12 Appendix C: Frequency Response Measurement 16 Appendix D: Subwoofer Design Documentation 19 Abstract Your job is to design an active 3-way spectral divider (crossover network) for a stock 3-way (tri-amplified) labyrinth loudspeaker system. Both the loudspeaker system and class D amplifier modules (4 x 100W) with power supply will be provided. Your spectral divider can either be analog (op amp based) or digital (DSP based, implemented using a stock development board). Level controls should be provided for each loudspeaker output drive. Primary goals include optimizing frequency response (target ±3 db over the range of 40 Hz to 16,000 Hz) and minimizing THD (total harmonic distortion). For bonus credit an (add-on) subwoofer can be designed and constructed, which will require a 4-way spectral divider. For analog designs, a PCB for the filter circuit can be created for additional bonus credit

4 1.0 Introduction Matt Cavenaile I am a junior studying Sound for the Performing Arts. My motivation for this class mostly stems from my longtime interest for audio electronics and, specifically, speakers. Throughout the course of this project I ve been able to learn the basics of how to program a DSP crossover and how to properly test loudspeakers frequency response. In the far future I hope to work in the film sound design/editing industry, hopefully for a well-known production company such as Warner Bros., Fox or Skywalker Sound. Short term I landed an internship at Legacy Audio located in Springfield, Illinois. Throughout the course of the internship I hope to expand my knowledge on what makes a loudspeaker high quality. José Manuel Conejo My motivation for taking this class is to be a better sound designer, artistically speaking, by knowing how to place speakers and deliver an even frequency response and sound pressure level to the audience. Also, I wanted to increase my level of knowledge on speakers, from the engineering standpoint of view. It is my wish to analyze the information that vendors and inhouse production house engineers can give me in the future regarding their speaker system, and judge it with an objective ear instead of just the brand name on the speaker. I learned that makers, vendors and people in love about a brand can tell you marvelous things about sound equipment, and of course, I want the best sound I deliver out of system because I know what I am doing, not because the brand or maker is magically doing my job. Second, I realized that sound design is a way more complex, yet passionate, area of study. I love to be in touch with physics and math that explains why sound behaves the way it does, but I haven t mastered (at least now) engineering knowledge. I definitely will come back to re-read McCarthy s book with your notes because the book makes wayyyyyyyyyyyyyyyyyyyyyyyyyyy more sense now. Finally, I will be creating a sound design career at my university in Costa Rica: Universidad de Costa Rica, in August 2017, so long journey, don t you think? Jing Zhao I am a theatre sound designer. In my line of work, I deal with theatre sound systems all the time. I have drafted speaker plots for shows that I have worked on without fully understanding the physics behind how loudspeakers work together in a system. Taking this course has helped me a great deal in getting to the bottom of why certain things are done certain ways. The speaker crossover network design project specifically has helped me better appreciate the power of modern DSP s for this application. I was able to hear how equalization introduces phase problems which, sadly, a lot of people ignore. I will work at the Alliance Theatre in Atlanta, GA this summer for my internship, and assist in building two touring sound systems there. My long term goal is to compose music for drama

5 DSP Design Narrative We decided to choose a DSP-based spectral crossover for three reasons: One is that, in comparison to an analog crossover, it s incredibly easy to implement. Choosing crossover frequencies is as simple as typing in your desired value into the minidsp user interface. This leads to our second reason for choosing a DSP crossover, which is the preciseness of tuning and crossovers that can be achieved by implementing one. With an analog crossover, 48dB/Oct is much more difficult to achieve. The digital interface gives users the chance to choose exactly what frequency they desire. Our third reason for choosing a DSP was the chance to fine tune the sound of our system through the use of graphic equalizers. After tuning our crossover frequencies we noticed some anomalies in the sound, which we were able to kink out through the use of the EQ. Also, in theatre sound our duty is to spend time making what is already in place sound great, instead of trying to make it work, which we would have been doing had we chosen an analog crossover network. In order to choose the correct crossover frequencies we utilized a reference microphone and a db/frequency meter. We wanted a relatively flat frequency response spectrum from 40 Hz to 16 khz, at 80 db SPLA with an accuracy of up to +/- 3 db, which we successfully achieved. We also calculated the size of the drivers, the mid-range is 4 and the woofer is 10. Based on class lectures, we knew that the highest frequencies they could produce without producing side lobes was 4.13 khz for the 4 mid-range and for the 10 woofer. We have yet to make any tweaks to the system for the spark challenge, but we are planning on switching out possibly every single speaker for more superior drivers. We will also build a custom subwoofer with much more power and sensitivity than the test sub that we used. The greatest difficulty in tuning our system was the presence of some anomalies across the - 3 -

6 spectrum. This took some time and patience and testing, but eventually we got the flat response we were looking for. Subwoofer Design Narrative We have yet to build a subwoofer enclosure, but we have all the plans for one prepared. We utilized ajdesigner.com s sealed subwoofer enclosure calculator to get all of our design values (Subwoofer). We chose a sealed enclosure based on the sound and quality of a sealed enclosure subwoofer we are very fond of in the sound studio in PAO Hall. The sub is a Dynaudio BM10s, it has a 10 inch driver and 200 watts of power (BM10s). To tune our system, we used the test sub provided by Professor Meyer. We considered the type of materials that would make the enclosure, and the subwoofer itself. For our subwoofer we chose a kicker CVR 12 because it was the best option we could find. It s much better than the subs offered to us by the stock of engineering department. Just a few of our subwoofer s relevant small signal parameters are: Vb (Box Volume) = Fl (+/- 3dB frequency) = Box Frequency = The rest of the box parameters can be seen in Appendix D. The value of Qts for our box was chosen because of the specified Qts on the subwoofers specs page. Again, we have yet to design our personal subwoofer, so we have not been able to see how similar our constrained design is to the unconstrained dimensions

7 3.0 Results Based on our measured frequency, we successfully achieved our target constraint of +/- 3 db over the range of 40 Hz to 16 khz. Photos of our test setup and meter results can be seen below in Appendix C. Every member of the team chose two of the songs we listened to that put different drives through more work. Some of the songs we played were bass filled, some midrangy and some with crystal clear highs. Gasoline Audioslave This song puts the bass and mid-range of the system on display. The kick drum that gives the track its hasty tempo is one of the most bass-y kick drums we ve ever heard in a rock recording. Even the guitar seems to have been transposed to be deeper than the sound it actually produces. Like a G6 Far East Movement Although our group is not particularly fond of this song, it was a great taste for the lower end of our system. There was loud, thunderous bass and a mid-range with high definition. The song is mostly artificial synth-y sounds that have been mixed and produced expertly. The vocals also reach fairly high, giving us a chance to hear some brilliance from the mid-range and tweeter. Don t Stop till You Get Enough by Michael Jackson This song is used to test how well the speaker responds to fast and percussive sounds. We are pleased with the result. The shaker sounds intimate. The clarity of the percussion instruments in general is above expectation. The texture of the guitar is well reproduced and quite delightful. There is not much thunder, but it is due to the way the song is mixed because the bass of the speaker is actually quite impressive based on tests with other songs. Come Away with Me by Norah Jones This song is used to test how well the speaker performs with a clean and clear vocal. The performance of the speaker is acceptable, but not great. The intimacy and warmth in the recording are both well reproduced, but the texture of the voice is not. The vocal sounds a little dry and at times even distorted

8 The Package: I know the sound has a very tight bass guitar and punchy kick drum. Overall speaking, the configuration we made totally accomplished my expectations in the low end in terms of punchiness and boom in this song. The mid/high end area sounded fine. I could listen to details in the vocal interpretation and it was clear and intimate. Listened to some nice acoustics when the drummer hits snare drum ring, giving us a good sense of clarity. Overall speaking, this song translated very well in our speaker from the ear memory I have about this mix. Bomb Track Rage Against the Machine This song is mixed by the same mixing engineer of The Package, Andy Wallace, and as I expected, the bass guitar and kick drum translated very well: powerful, tight and with body. Surprisingly, this song had less crispy sound than the previous one. Also, the guitars and the vocals sounded a little dimmed. Listening carefully, I could tell the mix sounded warmer than I expected, and I asked if the problem might be tweeter, since I was missing detail and brightness in the high end. Compared to the rest of the songs, this one did not sound mid-rangy to me, but definitely, I know we were missing a little of clarity a the top part of the frequency spectrum. Our design met our sonic expectations, which were to achieve an equal level response with +/- 3dB accuracy from 40 Hz to 16 khz. With this flat sound, we decided we could listen to mixes the way they were meant to be heard, with no bass or treble boost. Our desire was for the artists intention to be obvious. Some tweaks we intend to make in the future are replacing the tweeter, mid-range and woofer with more efficient and less resistant drivers. We will also be building a subwoofer enclosure featuring a more powerful driver than the test sub

9 4.0 References BM10S subwoofer - Complementing your BM series monitors. (n.d.). Retrieved April 06, 2017, from Subwoofer Box Enclosure Design Calculator - Sealed Ported Bandpass Closed Vented. (n.d.). Retrieved April 06, 2017, from

10 Appendix A: Activity Logs - 8 -

11 Activity Log for: Jing Zhao Role: Engineer Activity Date Start Time End Time Time Spent Troubleshooting + choosing crossovers 3/26 1:30pm 3:45pm 2.25 Fine Tuning Frequency Response 4/5 6:30pm 8:00pm 1.5 Subwoofer Enclosure Designing 4/5 8:00pm 8:30pm

12 Activity Log for: Jose Manuel Conejo Role: Engineer Activity Date Start Time End Time Time Spent Troubleshooting + Choosing crossovers 3/26 1:30pm 3:45pm 2.25 Fine Tuning Frequency Response 4/5 6:30pm 8:00pm 2 Subwoofer Enclosure Designing 4/5 8:00pm 8:30pm

13 Activity Log for: Matt Cavenaile Role: Engineer Activity Date Start Time End Time Time Spent Troubleshooting + Choosing crossovers 3/26 1:30pm 3:45pm 2.25 Fine Tuning Frequency Response 4/5 6:30pm 8:00pm 2 Subwoofer Enclosure Designing 4/5 8:00pm 8:30pm

14 Appendix B: System Tuning Parameters (DSP)

15 Figure 1. Input Gain at -7.0dB Figure 2. Four Way Crossover

16 Figure 3. Parametric Equalizer Figure 4. Output Channel s 1 and 2 - Gain and Delay Values

17 Figure 5. Output Channel s 3 and 4 Gain and Delay Values

18 Appendix C: Frequency Response Measurement

19 Figure 6. Measurement Setup Note - Reference Mic is approximately 1 meter away and the test sub is being used

20 Figure 7. Tuning at a level of 80 db SPLA Figure 8. Zoom of Figure 7 Settings Figure 9. Successful +/- 3dB SPLA Accuracy across the spectrum Figure 10. Zoom of Figure 9 Settings

21 Appendix D: Subwoofer Design Documentation

22 Figure 11. Subwoofer Enclosure Parameters Figure 12. Frequency Response of Enclosure Types