Morgan-Smith Electronics Ltd., Page 1 of 3 126, Great North Road, Hatfield, Herts,. AL9 5JN TEST REPORT

Transcription

1 EMC HIRE Ltd Ivel Road Shefford Bedfordshire. SG17 5JU Tel: Fax: OUR REF : 119T/04 YOUR REF : 04PO181 DATE OF TEST : 31 st Aug, 04 REPORT DATE : 1 st Sept, 04 Morgan-Smith Electronics Ltd., Page 1 of 3 126, Great North Road, Hatfield, Herts,. AL9 5JN att: Mr. Peter Morgan-Smith TEST REPORT Subject Free Field NSA in a F.A.R (Fully Anechoic Chamber) at Morgan-Smith Electronics Ltd., Hatfield The Contract A contract was raised by Morgan-Smith Electronics Ltd. to perform a Free Field Normalised Site Attenuation ` in the fully anechoic chamber belonging to the EMC facility at Morgan-Smith s in Hatfield. This was a standard NSTL test to verify if emission testing can be performed with a fixed height antenna and produce results that would reliably equate within prescribed limits to results on an open area test site. The Test Site The chamber measured approximately 6 x 4.8 x 2.4 metres high. Ferrite tiles lined all four sidewalls, the floor and the ceiling, but no hybrid absorber had been fitted at this stage. A single access door was provided adjacent to the EUT end. Coaxial transmission I/O adaptors were provided at mid level, between the receive antenna position & the EUT end. All normally exposed metal reflecting surfaces had been diligently boxed in, with ferrite tiling custom clad to the boxed surface. The receive antenna site was positioned roughly diagonally across the chamber facing the turntable in the opposite corner, but equally distanced from the corners to facilitate a 3 metre range between it and the turntable. The special transmitter bicon cage had approx 2 metres between itself and the end wall on average, but would be transmitting from an effective cylinder 0.7m tall & 0.7m dia. The cylinder had its centre position located to match the intended EUT test point over the turntable centre. The actual test range was laid out diagonally to obtain the most optimum use of the space within the chamber & minimise coupling to wall surfaces.

2 Page 2 of 3 Modus Operandi The EN document provides the architecture for a method which is derived from work done at the National physical laboratories in the UK & elsewhere in Europe. Essentially a F.A.R (Fully Anechoic Room) should emulate a free space environment with an antenna measuring point set at 3 metres to the front face of the EUT cylinder. The EUT set-up at Morgan-Smith s can be said to occupy a volume in the form of a cylinder of 0.7m diameter & 0.7m high placed 0.8 metres above the chamber floor. The size of the cylinder has to emulate typical EUT set-ups with their associated cabling whilst matching the performance of the chamber. To test the performance of the chamber in question, a small bicon antenna is used to transmit a signal from five points on the periphery of the flat plane of the cylinder and at three heights, namely the centre of the cylinder, the top & base. On any flat plane in the cylinder we are transmitting from the centre, forward, back, left & right positions as seen from the receive antenna. With both antennas calibrated (as a pair) in the free field for 3.0 metres and the AF s taken into the calculation, a comparison on a swept basis between the swept signal with cables connected together and the radiated signal will display the performance of the chamber relative to quasi free-space by showing any unwanted reflections from the chamber surfaces. So for each polarisation there will be fifteen sets of data, thirty in all with both H & V. The calculated data is presented on a linear grid with ± 4dBs limit lines shown. The graphs are grouped together with the five transmission points from each plane appearing on one grid. The swept range was made in a single sweep 30MHz MHz. The Test Equipment The measurement is essentially differential. The heart of the measurement is the R&S ESPI-3 spectrum analyser with a tracking generator output operating between 30MHz & 1GHz. The tracking generator feeds into the miniature transmitting bicon (UBAA-9115) via a 10dB pad. The received input to the analyser comes from the Schwarzbeck (VULB 9163) Bilog antenna. Height positioning of the receive antenna is critical & in this case was set at 1.2Metre, mid point between floor & ceiling. Where possible, return cabling from both antennas should be kept at balun height for 2 metres before falling to the floor. The antennas used are listed below, all of which hold an independent UKAS calibration certificate applicable to the task of measuring the FSNSA. The whole measuring system is software controlled using EMCH s NSTL/NSIL software program. Equipment Standards Verification The Rohde & Schwarz ESPI-3 Analyser. Ser. No UKAS Cert /10/03 The VULB Bilog Ant. Ser. No UKAS calibrated jointly with the UBAA 9115 below. The Schwarzbeck UBAA 9115 Ser. No.104, UKAS calibrated 28/8/03 Cert. No.2506 Two Weinschel 10dB microwave attenuators. (VSWR elimination, attached to antennas). Portable Computer AJP-D800P with EMCH s NSTL/NSIL software package. Set of Rosenberger microwave quality coaxial cabling plus special RF antenna stands supplied by EMCH. Data The data is presented graphically with all the five results from each level of the cylinder shape printed on a single grid. The horizontal axis displays the frequency linearly, whilst the vertical axis extends to plus and minus 10dBs about a centre zero line. The tolerance limits of ± 4dBs are shown in red.

3 Results Page 3 of 3 It was anticipated that there would be some deviations from the ± 4dBs especially between 30MHz and 250MHz of the swept spectrum in vertical polarisation. This will be due to rather low ceiling height. In most cases with vertical polarisation the bow-tie elements of the receive antenna tend to couple to the surfaces of the chamber. Hence, there is a recognised problem in the correct antenna factor being applied between 30MHz and about 250MHz even though the antennas were correctly calibrated in quasi free-space. For a more in-depth assessment of the performance may I suggest you refer to the contents of the Technical report SMT4-CT issued by M. Alexander of NPL in May The data supplied during this test could conceivably be used to make some corrections or allowances to the EUT test data and present results closer to the real values. From 250MHz upward the return data predominantly remains good & reasonably within the ± 4dBs. However, some excursions were also apparent between 800MHz and 1000MHz on both polarisations. This will be due to the lack of any hybrid absorber being fitted which starts to come into effect from the 800MHz region upwards. Summary Provided the test range is maintained in the diagonal configuration as measured during the NSTL test then return data will emulate the uncertainties shown on the graphs. If emissions from the EUT can be verified as to their polarisation then particular graph structures could be used to correct the received data and thus provide a more accurate assessment of the EUT performance. In any event most troublesome emissions occur from 300MHz upward & in which case straight data can be taken with margins displayed on the graph print-outs. To cut down the possibility of failure at a Test House after undergoing in-house testing it might be worth considering in the future the cost of covering some surfaces with hybrid cones. Engineer witnessing test Engineer carrying out test : Mr. Peter Morgan Smith (M-S Electronics Ltd.) : John Wombwell (EMCH) Report compiled by : John Wombwell SIGNED: Enclosed There are six graphs (grouped as three horizontal & three Vertical) with five sets of data (from each level) on each graph grid. The file naming procedure has already been explained. The 30MHz 1000MHz is shown in an expanded linear form, which will prove helpful in using as a correction. END OF REPORT

4

5

6 EMC HIRE Ltd Ivel Road Shefford Bedfordshire. SG17 5JU Tel: Fax: OUR REF : 119T/04 YOUR REF : 04PO181 DATE OF TEST : 4 th Sept,04 REPORT DATE : 6 th Sept, 04 Morgan-Smith Electronics Ltd., Page 1 of 3 126, Great North Road, Hatfield, Herts,. AL9 5JN att: Mr. Peter Morgan-Smith TEST REPORT Subject Field Uniformity EN (1997) in a F.A.R at Morgan-Smith Electronics Ltd., Hatfield The Contract A contract was raised by Morgan-Smith Electronics Ltd. to perform a Field Uniformity test as per EN (1997), in the Fully-Anechoic Chamber (F.A.R.) belonging to the EMC facility at Morgan- Smith s in Hatfield. This is a standard test to verify that a chamber can meet the requirements of EN , by looking at the possible variations of an immunity field across the area of a flat plane 1.5 metre across by 1.5 metres high (centred on the EUT position), over a 3 metre range, when transmitted with a fixed height antenna. The test is made in both horizontal & vertical polarisations. The Test Site The Test Site was the FAR (Fully-anechoic Room) at Morgan-Smith Electronics Ltd. measuring 6 metres long by 4.8 metres wide, with a height of 2.4 metres. All six inner surfaces of the room are totally covered with ferrite tiles. A single access door was provided adjacent to the EUT end. Coaxial transmission I/O adaptors were provided just at mid floor level two metres from the door. All normally exposed metal reflecting surfaces had been diligently boxed in with ferrite tiling custom clad to boxed surfaces. The receive antenna site was positioned roughly diagonally across the chamber facing the turntable in the opposite corner but equally distanced from the corners to facilitate a 3 metre range between it and the turntable. The plane grid for the sixteen probe positions was set up directly over the normal EUT position, but measured 1.0m across by 0.8m high. The grid started at 0.8metres above floor level. All 16 positions were located by hand on a telescopic mount designed for the purpose. A reduced 16 point square grid was chosen to match the architecture of the chamber and in response to the NSTL figures produced the previous day.

7 Modus Operandi Page 2 of 3 The criteria for reaching the required standard in Field Uniformity is laid down in EN (1997). A frontal plane area in the form of a square measuring 1.5m by 1.5m and standing 0.8metres above the floor is divided into 16 symmetrical points every 0.5 metres. However, in this particular case due to the size of the chamber it was decided to look at the F.U. on a 1.0m x 0.8m grid. The transmit antenna was positioned diagonally along the main length of the chamber. The Field Probe was adjusted for 3 metres between itself and the phase centre of the 6141A X Wing Bilog antenna. This particular model Bilog would be less susceptible to floor/ceiling coupling A standard immunity swept frequency calibration is carried out under computer control from 80MHz to 1GHz at a logarithmic step change of 1% for each of the sixteen positions. A field strength tolerance of Volts/metre was set up with the forward power being monitored at each frequency and all data being stored on computer. At the conclusion of all sixteen swept positions a computer programme algorithm determines, using the forward power in dbm, how many frequency points from each position collectively fall within ±3dB or 0 +6dB of each other. Up to 25% of all data points, if falling outside the tolerance, can be discarded. The uniformity criterion can then be considered met if the remaining 75% lie within. There is however a further relaxation allowing 3% of the remaining data points to be up to 10dB out. This entire operation is carried out for both horizontal & vertical polarisations. The Test Equipment The measurement is essentially relative. The pertinent parts of the measurement are a Telescopic Probe Mount which is RF transparent & can position the field probe in each of the sixteen points of the grid. A UKAS or NPL calibrated field probe to measure the field in the appropriate plane. An RF power meter to monitor the forward power to the antenna via a directional coupling unit. The power meter, RF probe & signal generator are all on the GPIB. The transmit antenna was the Chase 6141A X wing Bilog. The whole measuring system is software controlled using EMCH s standard immunity package conforming to the EN (1997) verification programme. AR.30W1000 RF power amplifier. 30 Watts. 20MHz-1GHz Supplied by EMCH. Computer with EMCH s Immunity and software package. The Manual Telescopic probe positioner. SI-200 and FP3000 field probe & controller. (EMCH) FP-3000A field probe. Ser. N Cal d 11/2/04 by NPL ref: E Chase 6141A X wing bilog antenna. Calibration not applicable in this test. Boonton 9200B Power meter. UKAS cal d 30/6/2003 Cert:003137/01/010 Rohde & Schwarz SML-01 sig gen. UKAS cal d 17/3/04 Cert:04873 Data The data is presented in two parts, one being for horizontal polarisation with 16 files beginning PMSH cal the other in vertical beginning PMSV cal. Within each polarisation there is a log file, a report file, an error file and the sixteen data field uniformity graph. Calibration Log File The file comprises a header with relevant information derived from the sixteen positioned files from which the software has determined the 25% rejected at each of the frequency points measured. These frequency points will run from 80MHz to 1GHz at 1% interval and display against each frequency the level in dbm and the file/position rejected. Note the last two characters of the rejected file name indicate the actual physical position of the probe within the square plane grid. The Errors File The file comprises a header with relevant information derived from Cal Log and shows those frequency points which lie outside the ±3dB limit. In detail, against a particular frequency, we subtract the average power from the max power to give a result (B-A), we then do the same again with the minimum power (C-A) and if (B-A ) is more than +3dB or (C-A) is less 3dB then we have an error or rejection.

8 Page 3 of 3 The report File Again the file comprises a header with relevant information derived from the sixteen positioned files along with the maximum power from the remaining points which produce on the graph entitled Field Uniformity the two reference lines of (in this case 3dB & +3dBs) with the chosen files embraced. The Field Uniformity Graph A Pass means that all the final selected data ( excluding 3% which is allowed ±10dB) lies within a ±3dB criterion. On the failure side there are two conditions: The first being that a number of data points (exceeding the 25%) lay outside the ±3dB limit. However, the standards allow a Pass on condition that the remaining points outside the limit are no more than ±10dB for 3% or less of the remaining frequency points. A total failure is indicated when there are more than 25% of the frequency points outside ±3dB or any additional points greater than ±10dB. Our method in displaying the graph in the format shown has two advantages, one being that the End User can see how hard the amplifier is being worked over a specific range of frequencies, and the other showing the absence of the rejected points by gaps in the trace. Results A combination of the reduced grid size has permitted a pass in both Horizontal & Vertical polarisation. This particular test has been performed primarily to check the Chamber performance although the calibration files were downloaded onto the Morgan-Smith Electronics Immunity software program for future evaluation and use. Summary As Morgan-Smith Electronics Ltd. is testing devices which lie safely in size within the 1m x 0.8m grid, then a fairly high degree of field strength uniformity can be expected thus providing the necessary confidence in product evaluation. It would be stressed of course that reliable data also depends upon Morgan-Smiths calibration data concerning their own voltage field probes and power meters. Engineer witnessing test Engineer carrying out test : Mr. Peter Morgan Smith (Morgan-Smith Electronics Ltd) : John Wombwell (EMCH) Report compiled by : John Wombwell SIGNED: Attached Two complete packs of data printouts. One for vertical & one for horizontal. These include a Calibration Log File, a Calibration Report File & the Errors file (if appropriate) plus a graphical presentation of the selected files. The titles are given for each group & the figures are self explanatory. END OF REPORT

9