Not entirely on subject since this is for a transceiver, but figured this would be the group most likely to generate a useful answer.
I happen to like the older Icom IC-271H and 471H, I have a pair of each where one is currently the control and one is the in-progress unit. All the normal stuff has been done, complete teardown and cleaning, all cold joints, broken traces and iffy stuff dealt with, all new electrolytics, etc. On through basic improvments like replacing the final output jumper from the filter to the rear SO239 with a RG400 jumper into a crimp type N bulkhead to reduce RF in the chassis. And of course a very careful alignment. Have been going through everything step by step replacing fets with newer lower NF units, caps and resistors in critical sections with lower TC ones. Anyway, while cleaning up the "APC/ALC" circuit to eliminate overshoot I found a great deal of noise that caused issues as the APC circuit was made faster to eliminate overshoot. Tracked it all down to the -9 supply (nominal, actual -10VDC), a very old metal can DC to DC that happens to be physically located on the CSS board but as its own seperate circuit. It feeds -10VDC to radio and that is used in both the TX APC/ALC to vary gain of the first amp (3SK74) after the SSB filter or FM modulator (as the pulldown voltage). It also feeds into several op-amps in the AGC and APC circuit. It was down < -45dbm (at xcvr output) but the APC is always running at least slightly negative causing a multi-mhz wide noise increase. In FM, it also AM modulates the carrier at 10 - 15 Khz. All well under legal limits (90 - 100db below transmitter output) but it just bothers be and any attempt to make the APC circuit more sensitive makes it worse. So first step was attacking the switcher circuit. Replaced the 100uf 16V Al caps with Panasonic OsCon types, 25V 150uf paralleled with .1uf ceramics. This took care of the wideband RF and fixed the multi-mhz wide hump, but the 10 - 15 Khz (depends on load) AM mod when in FM got better by ~ 10db but is still there at around -50dbm. I am looking at disconnecting the entire circuit on the CSS board and replacing with a modern DC/DC w/ filters on a seperate board. Initial though was a very low noise isolated output 10V DC/DC converter that could be rigged as a -10V supply. Could not find anything with really good specs so am currently thinking something like a Traco TDN 3-1213WI (200ma 15V) feeding a filter into a low noise linear reg to -10VDC followed by another filter. I was hoping someone here might have a better suggestion as to method or DC/DC converter. "Long" term stability (seconds and longer) is not really an issue, it could wander by a few hundred mv and not be an issue, but any noise is as it will end up coupled into the transmit chain and AM modulate things...At lower frequencies the circuits in question actually have a fair bit of gain too. Just to get a little closer to on-topic, the radio in question actually has crystal heaters on all XOs and my biggest issue onb that front is a repeating ~ 4.5Hz jump on a ~ 10.24 Osc that pulls the VCO by ~ 60Hz for minutes to hours before reverting. _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.