Nixie supply safety and noise reduction questions

Started by Max999, November 22, 2018, 06:37:44 PM

Previous topic - Next topic

rankot

#20
Spikes are on the output of a power supply. I am still too busy with my job, so I'm just messing with SMPS, don't have time to try it with some kind of preamp. I will try LC filter, actually dual LC - ltspice simulation shows that I need that for super-clean power output. Something like this:

  • SUPPORTER
60 pedals and counting!

Rob Strand

QuoteOn the output of a power supply.
The *cause* is probably just the MOSFET and/or the diode and the reason you see it it probably the cap ESR.

When you measure ripple you have to make sure you connect the CRO ground to the correct point otherwise the CRO won't see what you expect.  Normally that is across the output cap on of the PSU.   However if you want to measure what the tube sees you would measure across +HV and 0V at the tube.  You don't always get the same result.

You can do this experiment:  Add a low pass filter (say R + C) to the output of the PSU and see if you can remove the glitches.  In theory the filter should remove them.  If they don't get removed then the glitches are artefacts that are somehow getting into the CRO; maybe via the ground.   If they do get removed then you are fairly confident they are real and are actually present at the PSU output.
Send:     . .- .-. - .... / - --- / --. --- .-. -
According to the water analogy of electricity, transistor leakage is caused by holes.

anotherjim

When scoping for high frequency stuff, I've always found it helpful to use a separate ground reference connection instead of a probe clip. Actually, a ground clip on the probe you are probing with is a PITA anyway...
So use another channels probe ground or a wire to a dedicated ground post on the scope to make the ground contact.
Thing is, with high frequencies, the probe ground clips are electrically different from the probe tip, the self-inductance in particular, enough to unbalance what you see on the scope. Also, you might have a good old ground loop.
When the thing you are testing has an AC ground connection and the scope does too, it can be better to not use a probe ground at all.
My bench amplifier has AC ground and so does my scope, but my bench psu is isolated. I have to remember to ground scope probe when the amp isn't connected.