Originally Posted by
edeca
Thanks again Howard for all your excellent design in this thread. I have built the two amplifiers now, pictures attached. I get different voltages to your measurements at some points in the loop amplifier, need to investigate this further.
Two quick questions if I may, firstly what output is expected from the phase detector input amp during a phase reversal? At the moment I seem to get a "missing" pulse (in fact I seem to miss two cycles). Output from scope is attached, yellow is logic output and blue is sniffed with a scope probe.
Secondly did you add any filtering into the design? With no TX unit nearby I get spurious logic output at Khz frequencies. This does not occur when the loop is disconnected, so I am fairly certain it is other RF interference. I suspect this is because I have not yet tuned the loop to 5Mhz. I built revB but have not added the three extra jumpers at this time, only the single capacitor.
That's some very pretty prototype work!
Some variation in voltages is inevitable. If the parts values check out okay in the loop amp, then another step to take is either verify that the Vbe of the input pair (Q1 and Q2) match to within a few mV, using the diode check function on a multimeter. You can also use the optional input offset adjust pot to set the voltage at the junction of R8/R9. The voltage can vary a little, perhaps +/-0.1V. I'll update the schematic with an acceptable range when I get the chance. I've constructed several loop amplifiers now, always using transistors from the same production batch, and all have been functional with no adjustments.
Depending on the tuning of the transponder tank (and, to a lesser extent, the timing loop) you may see a missing pulse, or double pulses for a phase inversion. I believe that two missing pulses is also acceptable to the digital phase detector. In any case, please decrease the Q of the transponder tank so that it will return to its original amplitude quicker on a phase inversion.
Filtering is via the loop tuning (and for higher frequencies, the natural rolloff of the transistors). Spurious outputs are not unusual, as many noise sources contain energy in the frequency band we are using. During my experimenting, I located one particularly offensive source: the switching power supply on my laptop computer. Similar noise can result from the ESC and motor in the car. The preamble detector weeds out almost all of this, so the microprocessor can spend its time doing more important things.