From: Abner Mull Newsgroups: alt.radio.pirate Subject: BW 1W LCD PLL evaluation (long) This report details my experiences in constructing a Broadcast Warehouse 1W LCD PLL Exciter. This is BW's "old" model PLL exciter, as they have recently introduced the new and improved PLL Plus. Note that the results I report here are only valid for the device as I have constructed it - more or less careful construction techniques may result in better or worse performance. I do not have a spectrum analyzer, so the tests I performed are limited to carrier frequency drift and audio performance. Note that I have no connection with Broadcast Warehouse, and I purchased this kit with my own money, so I have nothing to gain or lose by publishing these results. I just thought they would be of interest to the community. Construction ------------ Someone recently described this exciter as "a bit fiddly" to build. I found the construction straightforward and not at all "fiddly." However, the small size of the board and lack of schematic and test points do make troubleshooting and reparing the device a tad difficult. My exciter released a puff of smoke very early in its life, and I was sure I must have blown the output transistor. Replacing it did not restore the exciter to life, however. Eventually managed to find a schematic on the web that contained the disclaimer "may have a few bugs." This schematic was enough to lead me to the real problem - inductor L3 (the .15 uH choke) had managed to burn itself up. My exciter is housed in a metal box approximately 3" x 8" x 6" in size (Radio Shack Cat No. 270-274). The unit is powered from an internal 12V power supply consisting of an 18V 25VA toroidal transformer (Digi-Key Part No. TE62064-ND), bridge rectifier, current limiting resistor, a total of 6,000 uF of filtering, feeding a 7812 voltage regulator. There is an EMI filter (Mouser stock no. 562-857-01/5) built into the IEC power connector for the mains. Test Equipment Used ------------------- The following test equipment was used for the tests: . Neutrik MR1 "Minirator" audio generator . Neutrik ML1 "Minilyzer" audio analyzer . Racal Dana 9009N modulation meter . Radio Shack frequency counter (Cat No. 22-306) . Diamond SX-200 SWR/Power meter . Vectronics DL650 Dummy Load For all tests, the output of the exciter was fed through a short length of RG-58 coax to a Diamond SX-200 SWR/Power meter. The antenna output of the SX-200 was connected to a Vectronics DL650 dummy load through another short length of RG-58 coax. The Racal Dana 9009N modulation meter was connected through a short length of RG-58 coax to an RF sample output that I provided in my exciter. The unbalanced output of the MR1 was connected directly to the Composite audio input of the exciter. Finally, the audio output of the Racal Dana modulation meter was connected to the unbalanced input of the ML1. Measurements ------------ The tests I performed were similar, though not identical, to an FCC Proof of Performance test. A true Proof of Performance test includes the whole airchain, whereas I was just testing the performance of the exciter. Also, I turned off preemphasis for all my tests, as the Racal Dana modulation meter does not include a deemphasis circuit in its audio output. All tests were performed with a carrier frequency of 99.9MHz. The carrier frequency was measured as: 99.9002MHz at PLL lock, 99.9000MHz after 10 minutes, 99.8995MHz after 1 hour, 99.8998MHz after 3 hours, and 99.9000MHz after 14 hours. Power output with the 12V supply was 700mW, and was not noted to vary at any time during the tests. With no modulation, the device draws approximately 220ma. FM Noise (or signal-to-noise ratio) was measured at 1KHz at the audio output of the 9009N, using the realtice level function of the ML1. 0dB was referenced to a 100% (+/- 75KHz) modulated signal, and the noise level was then read directly from the ML1 with no audio input. The measured noise level was -58.2dB. Looking at it another way, zero audio input produced an FM modulation level of .075KHz, which is 0.1% or -60dB. This is pushing the limits of the instrument, as residual FM noise is specified as "-46dB with respect to 10Khz peak deviation..." for the modulation meter. This corresponds to -63.5 dB with respect to 75KHz peak deviation. And that's for a new one, not a 20 year old one that could probably benefit from a new power supply filter capacitor. Asynchronous AM Noise represents the AM modulation level of the FM signal with no audio input. This modulation is caused mostly by power supply ripple. The FCC requires that it be -50dB or better, but done properly, the measurements are taken with deemphasis in the detector circuit [Mendenhall]. I measured .09% AM modulation without deeemphasis, giving a value of -60.9 dB. Synchronous AM Noise represents the AM modulation level of a modulated FM signal. The signal is modulated with a 400 Hz tone at 100% FM modulation, and the AM modulation level is measured. The FCC has no requirements for Synchronous AM Noise, though according to [Mendenhall]: "Synchronous AM of 40 dB or more below equivalent 100% AM, is considered to be acceptable." I measured AM modulation of 1.39% on positive peaks and 1.49% on negative peaks. Using the 1.49% values, we get a Synchronous AM Noise level of -36.5dB. Frequency reponse and distortion were measured at 100%, 50%, and 25% modulation levels at all frequencies supplied by the MR1 from 20Hz to 16KHz. The modulation level was set to the specified level by adjusting the input potentiometer on the exciter with a 1KHz signal from the ML1 and leaving it there for all frequencies. The ML1 output was set to 0dBu and left there for all measurements. Connecting the output of the MR1 directly to the input of the MR1 resulted in the following baseline values, with 0dB referenced to 1KHz: Freq Level (dB) Distortion(%) Freq Level (dB) Distortion(%) 20Hz -0.03 .030 630Hz 0.00 .029 25Hz -0.02 .030 800Hz -0.01 .030 30Hz -0.02 .030 1.00KHz 0.00 .029 40Hz -0.02 .030 1.25KHz 0.00 .028 50Hz -0.02 .030 1.60KHz +0.01 .028 65Hz -0.01 .030 2.00KHz 0.00 .029 80Hz -0.01 .029 2.50KHz 0.00 .029 100Hz -0.01 .029 3.15KHz -0.02 .029 125Hz -0.01 .029 4.00KHz -0.02 .030 160Hz -0.01 .029 5.00KHz -0.02 .032 200Hz +0.01 .029 6.30KHz -0.02 .035 250Hz +0.08 .029 8.00KHz -0.02 .041 315Hz +0.19 .028 10.0KHz -0.01 .055 400Hz +0.03 .028 12.5KHz -0.01 .030 500Hz +0.01 .029 16.0KHz -0.03 .030 The input impedance of the ML1 is 20K ohms, while the BW exciter has an input impedance of 1.5K ohms (accoring to the schematic). The lower input impedance could result in some increased distortion if the MR1 is not happy driving the lower impedance from its unbalanced output. Now for the numbers of interest: Values at 100% Modulation: Freq Level (dB) Distortion(%) 20Hz +0.01 .168 630Hz -0.01 .141 25Hz +0.05 .160 800Hz -0.02 .140 30Hz +0.06 .159 1.00KHz 0.00 .140 40Hz +0.06 .145 1.25KHz 0.00 .142 50Hz +0.05 .152 1.60KHz +0.01 .143 65Hz +0.04 .150 2.00KHz 0.00 .147 80Hz +0.02 .140 2.50KHz 0.00 .150 100Hz +0.01 .142 3.15KHz -0.01 .155 125Hz +0.01 .108 4.00KHz -0.01 .166 160Hz -0.01 .142 5.00KHz -0.01 .181 200Hz 0.00 .142 6.30KHz -0.03 .204 250Hz +0.06 .135 8.00KHz -0.06 .228 315Hz +0.14 .135 10.0KHz -0.12 .229 400Hz +0.01 .140 12.5KHz -0.27 .148 500Hz 0.00 .140 16.0KHz -0.59 .133 Values at 50% Modulation: Freq Level (dB) Distortion(%) 20Hz +0.02 .262 630Hz -0.01 .241 25Hz +0.06 .252 800Hz -0.02 .241 30Hz +0.07 .252 1.00KHz 0.00 .240 40Hz +0.06 .247 1.25KHz 0.00 .240 50Hz +0.06 .247 1.60KHz 0.00 .242 65Hz +0.04 .243 2.00KHz 0.00 .242 80Hz +0.03 .247 2.50KHz 0.00 .241 100Hz +0.01 .244 3.15KHz -0.01 .245 125Hz +0.01 .162 4.00KHz -0.01 .245 160Hz 0.00 .245 5.00KHz -0.01 .249 200Hz +0.01 .240 6.30KHz -0.03 .257 250Hz +0.09 .229 8.00KHz -0.07 .269 315Hz +0.21 .237 10.0KHz -0.13 .265 400Hz +0.02 .241 12.5KHz -0.28 .248 500Hz 0.00 .241 16.0KHz -0.61 .253 Values at 25% Modulation: Freq Level (dB) Distortion(%) 20Hz +0.03 .502 630Hz 0.00 .482 25Hz +0.06 .495 800Hz -0.01 .484 30Hz +0.07 .483 1.00KHz 0.00 .482 40Hz +0.07 .486 1.25KHz 0.00 .485 50Hz +0.06 .485 1.60KHz +0.01 .481 65Hz +0.05 .490 2.00KHz 0.00 .480 80Hz +0.03 .489 2.50KHz +0.01 .478 100Hz +0.02 .487 3.15KHz 0.00 .483 125Hz +0.03 .314 4.00KHz -0.01 .481 160Hz 0.00 .487 5.00KHz -0.01 .481 200Hz +0.01 .490 6.30KHz -0.03 .488 250Hz +0.07 .459 8.00KHz -0.06 .490 315Hz +0.17 .478 10.0KHz -0.12 .492 400Hz +0.02 .489 12.5KHz -0.27 .497 500Hz 0.00 .483 16.0KHz -0.60 .511 Note: Most of the apparent falloff at high frequencies is a result of the limited bandwidth of the audio output of the 9009N. A graph in the 9009 maintenance manual shows typical response off approx .2dB at 10KHz, and approx .6dB at 16KHz. Conclusion ---------- Impressive numbers for such an inexpensive kit. References ---------- [Mendenhall] Geoffrey N. Mendenhall, "Techniques for Measuring Synchronous AM Noise in FM Transmitters", available from Broadcast Electronics