Nixie DMM updateWeather is crap today, no way I am working outside on the garage construction. So I am spending all day on that DMM...
Just looked at the forecast, decent weather coming back only on Saturday. So, today and the next couple days, I will stay inside, working on the DMM... hopefully I will make some progress !
So I am starting with the basics of course. Power rails and filter caps...
The power rails and many signals are readily available at the back of the instrument, on the track side of the board that holds all the sockets for the card edge connectors, as you noooo doubt saw in my presentation pics. No doubt...
PSU schematic below.
So we have 200V for the Nixie tubes and the Neon bulb indicators.
Then +/- 6V, and +/- 12V.
Full wave rectification, and Zener diodes. Two in series, so that if you tap one only, you get 6V, and if you tap the two of them, you get 12V.
The 200V has lots of ripple, 12Vpp or so, so the cap could do with a replacement but... the Nixie tubes couldn't care less, they work fine regardless. So since it can't cause my problems, I will leave that cap alone for now.
Then the two positive rails... they are good, zero ripple.
However the two negative rails had a lot of ripple. 400mVpp for the -6V, and one volt or so for the -12V rail.
I removed all the Nixie boards so I can get to the PSU board which is hiding in the middle of the instrument, sitting atop of the transformer.
It has 17 terminals and none of them are identified. Fun. They are numbered on the schematic but what is the pinout on the board ? Good luck. The filter caps are vertical cans bolted to the chassis. They are not identified either....
So I spent a bit of time probing things and drawing notes... then I could start troubleshooting the thing.
Tested all diodes, looking good. Tested all the beefy 82R 1W series resistors upstream of the Zener, good too. Good enough at least.
Then scoped the filter caps (3 of them, in different places) of the negative supply. See the screen capture of the main cap, right at the output of the bridge rectifier.
It does not filter anything whatsoever, ie, it's open circuit...
Power off, disconnect one of its wires, test it with the Chinese component tester : "No, unknown or damaged part ". OK.
So I replaced it. It's a 1000uF 50V one. Looked in my stock of crusty old salvaged caps. Found a 1000uF 63V "Jamicon", and it tested good on the Chinese meter. ESR at 10R looked a bit high I guess, but still good enough for a quick test. Better than open circuit anyway...
I soldered it quick an dirty underneath the chassis, hence the DMM laying on its side no choice.
Power up. A few seconds later that cap goes POP !!!!!
Pulled the power plug quick. Noticed the cap was BURNING HOT !
I thought oh, maybe the ESR was too high then....
So I searched my stock / left over of brand new Nichicon low ESR caps. No direct replacement sadly. Best I could come up with was a couple 470uF 63V in //. Good enough.
I tested them first.. and their ESR is 5R ?! Not much better than the one that popped !
So I thought well, the old cap must have been defective, who knows.
So I power up. I kept my fingers on the caps to feel if they would heat up like the previous cap.
THEY DID !! It was getting freaking hot freaking fast !
So pulled the plug again, didn't want my nice Nichicon to go pop, for sure.
Then started scratching my head.. what the fuck is going on here....
After a few seconds, light bulb moment : it's the NEGATIVE rail, that black wire is NOT supposed to go to the negative terminals of the caps !!!
Oh dear.....
I let the caps cool down, soldered them the right way around and scoped the result. Looking much better for sure !
Put all the Nixie boards back in place so that the power rails are loaded as they should and I can do a realistic ripple measurement.
Ripple now ? Couple volts on the first cap I just replaced, and zero at the final output.
Voltage levels are closer to nominal as well. -12V used to be -11,4V, now -11.86V
-6V was -5.8V, now -6.0V.
So.. making progress, hopes are high... but no, bloody instrument is not grateful at all, symptoms still 100% the same.
OK maybe it's because of the missing resistor hence missing Vref, on the Vref board.
So, let me cobble some external Vref quick, see if that improves things.
Anyway, testing and fixing the power rails was a pre-requisite, it needed to be done. So I still call that progress, and I can now rule the PSU out.
Stay tuned...