capturing voltage spike with fluke 289?

[coppercone2]

I think this has something to do with the control regulator on a daughter card not working, maybe something with fold back. That card was severely damaged and I thought I fixed it, but i guess not. That was too hard to trouble shoot. please don't bother with this thread because I don't even want to look at that card anymore, maybe next year. I am putting this away into the attic so I don't waste my summer, it has like 50000 HV mosfets on it and can get fucked

[BeBuLamar]

I didn't know that the diodes were Zeners. If so what's their breakdown voltage?

[BILLPOD]

Coppercone2, This board that has the diodes that are burning up,...can you purchase a new board from Miller Welding?  At the very least you can use the new board to troubleshoot the bad one.  The welder is of no use to anyone in it's present condition
and just putting it out of sight is just going to prolong the inevitable.   I've been down similar roads before, so I speak from experience.  Best of luck.

[coppercone2]

the problem is the boards are like 1.5k each lol

if I took time to replace every single part in this it might be cheaper. But the other problem is I don't have the tig torch or foot pedal or anything, so its not even appealing. My budget on this thing is basically tapped out, and I don't think I like flyback supplies or stuff powered directly from mains, and I already had a year a while back that I ruined trying to fix some expensive equipment, do not want a repeat of that summer. Tig welding is not what I want to get into with the economy in this state.. huge investment.

[BILLPOD]

WOW, I didn't know this was a big bucks unit.   My only suggestion now is to look around for a welding forum, and pick some brains there.  I wish you success.

[coppercone2]

I am gonna try one more time because I don't know if I ever really washed the PCB good because I just assumed the varnish was good. I left them all in ultrasonic for 30 minutes and some of the conformal coating did peel off, that had a slight mustard color near the main voltage bus. I wonder if there was leakage into the control circuit for a duty cycle voltage adjustment that was messing with it. But I realllly want to put it into the attic because its draining my money patience and time, but I can't resist.

One thing I forgot to check is if there was soot under the connector plastic housing... i would kick myself if its something dumb like that, but they use headers for 800VDC!

TBH if I buy a isolated high voltage differential probe, my tune might change. I could solder in DIP socket connectors to the diodes, buy a cheap string of diodes to replace and take oscilloscope captures at the gate of the different power transistors, but I got a really strong feeling that I was wasting my time and learning nothing

And this might have to do with a complicated power factor correction circuit, I think it might be failing to do something like active spike filtering or whatever (where it disables a transistor when they know there will be a spike on the circuit)

[coppercone2]

oh boy I think I found a bad 100nF capacitor connected to the resistor from the V+ bus. When I was removing it, the end came off, with one end stuck into the PCB.

I stuck one probe on the other lead and one probe into the hole that was left behind, it read 60Kohm. I was desoldering the other side and managed to break the capacitor off and it fell into a bunch of rubbish on the floor, and I found a too much garbage while sweeping so I just vacuumed the area and threw it away.


 I was not sure if this was from damage, but the capacitor came apart awfully easily. I got a very similar 100nF capcitor from my components storage box, and it took about 3x more force to remove the lead, and the lead broke off leaving the capacitor intact, without ripping the outer hull of the capacitor (i.e. it turns into a SMD part rather then getting structural damage with a big crater). I think that capacitor was damaged, I don't take the multimeter reading to heart because I could have damaged it poking it in there, but what I do notice is that it was too fragile.

So maybe there is hope yet. I can't get away from this thing lol . I notice when parts come apart too easily that might mean fracture damage.

[coppercone2]

I managed to get it stable after replacing the suspicious 100nF capacitor. I also started it from a 20A variac set to 110V, when I raised the voltage to 119V it was still OK. Did not run it directly from mains yet.


I need to do a thermograph to see if its not over heating, the diodes are at 18.5V now, and the inductor is making a slowly up-down ramping in intensity buzz. Display code -5, which is HLP-5 (missing letters), which is something with thermistor.

Once I verify nothing is burning with thermal camera, I will look at the thermistor circuit and check the thermistor for damage.

now that its not smouldering or making the bad noise I can work on it, check supply rails, etc. 

• Put out fires and wait for the 6th ☑

[coppercone2]

First order of buisness will be to replace the 20V zeners with 22V Zeners as I originally thought they were. Since I measure 18V, it must be that the designer used 22V for whatever reason to get 20V output. Part of the reason I thought something was going on with the part, since the glass fragment I recovered was a 22v zener designation. So 3W 22V zeners should be correct.