Placid HD BP.

Same pattern, less magnitude.

Ground loop it can't be. The reg wasn't connected to anything else than the scope while measuring.

Transformer EMP, hardly - I tested with both the tranny (TP supplied) mounted in the DAC and an outboard tranny, just to make sure.

Anyone else, with access to a scope - willing to make a trace of the output ?

Sure, 100% agree.

But is that supposed to be present at the output of the Placid. I read elsewhere, you once had measured the output of a Placid (HD?) and had measured 0mv or close to.

Edited by user Sunday, April 14, 2013 3:03:38 PM(UTC)  | Reason: Not specified

Yeah a couple of things at play here I think.

First the Placid HD line regulation is a configurable thing. It is a discrete circuit with a simple LTP for the error amp.

There are a couple of circuit parameters that can be changed to give better line regulation especially at higher current.

First and simplest is to simply reduce the input ripple by using a larger trafo and/or reservoir caps.

second you can change some circuit values/parts to best suite your situation.

If there seems to be some ripple at TP_VREF you could increase C3 to say 22uf(just add a cap in parallel) - you may want to experiment here.

You could also experiment with increasing R10 and R11 up to around 10K should work, but I would try 1K and then 2.2K first.

Finally it could be helpful to adjust R9 down to say 75R. This give the LTP more current to work with.

if also could be beneficial to adjust R6, but I don't think that will be necessary.

there are other changes that could be made but I would try those first as they are simplest.

Cheers!
Russ

You might actually try reducing R6 instead of increasing it. We don't want to starve the reference part of the LTP. You could even try simply shorting it - or use an very small value, say 21R. keep the cap.

Cool, next thing it to do try reducing R9 - even by half is fine. Say 47R.

Also it would be good to know how much ripple is present there as that will let us know about sizing of trafo and res caps.

10.6v

Try adding a large cap across C2 as well (same as C3). You can revert R9 was just curious if the shunt transistor was starved for current.

I was able to reproduce your results given that large input ripple.

The good news is I also was able to fix it. :)

There are two ways to fix it, the first is obviously to reduce the input ripple.

The second is to isolate the error amp from the ripple.

This is not quite as simple but very doable. It requires a little trace cutting so it's up to you if you think its worth it.

On the bottom of the board cut the trace from V+ to QP7/QP8 emitters (see attachment)

Then add say 221R from V+ to either emitter pad (they are connected)

Then add a suitably rated (say 25V) 220-1000uf capacitor from either emitter pad to GND observing correct polarity.

When I tried this even with a very high input ripple I had < 50uV ripple on the output.

Worth a shot. :) Its pretty easy.

I will probably add this feature to the next revision. Thanks for making me think a little more deeply about it.

Edited by user Sunday, April 14, 2013 7:09:02 PM(UTC)  | Reason: Not specified