Can anyone describe what a Waveform DC shift is?

Started by krawitjr, October 30, 2020, 01:15:04 AM

Previous topic - Next topic

admin

#15
James, what you are showing is probably more common than what krawitjr is referring to but not the same thing. He is talking about a horizontal shift of the entire signature not just the breakdown point.
Here is a simulated DC shift using the Pulse Generator on my Tracker.
The most important troubleshooting tool is your eyes.

jvthorsley

#16
Ah I think as has been said in this discussion, this refers to the  horizontal shift you get when there is a CMOS backup battery or high value storage capacitor in the circuit.

But you can read all the books / manuals until the cows come home, but the only way is to keep using a Tracker to fault find circuits, and in the process learn different ways of getting the best out of it.

krawitjr

Yeah,
Huntronian is completely correct, the difference being I've seen both a purely capacitive signature and a zener-like transistor signature (MOSFET) shifted to the extreme left instead of the right of the origin, on two different boards, months apart.  But otherwise, yes.

I've given up on the answer, as I assumed this would be a known and observed phenomenon and it turned into a UFO debate over if I saw anything at all.  So I'll not pursue this here any further. 

But I'll just leave the info in case anyone else ever sees it - The last board to exhibit this was a control board for a large APC UPS, but the part that was scanned was a MOSFET (had IR part number and data sheet confirmed part was a MOSFET, all I recall).  It was very close to the main logic/CPU on the control board, it was a through-hole component that stood straight upwards, among SMD parts.  The UPS used to work but now would not longer even start. Failure was likely age-related. 

I touched from the large top-most drain tab to body/chassis ground and it wasn't shorted according to the signature, but was suspected of failure due to getting hot when the UPS was trying to operate (but having no-boot up of logic board).  However, the "good looking" signature was extremely shifted to the left on a 5v range. Rescanned 4 times over 30 minutes to prove I wasn't seeing things, the shift was the same every time.  No shift was observed on any other part on that board.

No batteries were connected, no coin cell batteries observed on the board, there were caps but I had drained them and my fluke multimeter showed transient voltage in the thousandths of a volt or less on them before we began testing. 

So that's the end. Still a mystery, maybe one day I'll know what that means.