MM350P Woes, blowing IGBT's.

[greenbuggy]

Bought this as a nonworking unit w/o gun. Serial LF054480 Stock 907300. 200/230/460 volt unit configured with moveable board and wire connector to 230V, connected to a single phase circuit providing 245ish volts. Bought a nice Abicor gun for it too. Originally throwing error code 3, not on startup (that would indicate shorted trigger switch) but when trigger was pulled and held for more than 2-3 seconds. Voltage output of 0.00 on the display when trigger pulled. Called Miller, had a tech tell me to measure resistance between white/black wire pairs going to IGBT’s, had one pair at about 10 ohms which Millers tech told me was a shorted gate on an IGBT, replace the pair.

Go to LWS, spend a chunk of change to get 2 new IGBT’s, Miller PN 212936. Install in welder using ground wrist strap and everything. Power back up, now when I pull trigger I get 40-50 volts OCV on the display. Go to try and lay a bead (.035” wire on steel plate, 75/25 ar/co2 mix shielding gas), as soon as wire touches the surface I hear a loud “POP”, no arc and welder goes back to 0 OCV on display (and measured with meter) across output terminals. Hold trigger, get error 3.

Call Miller again, different tech this time tells me “oh yeah anytime you replace the IGBTs you also replace interconnect board” great, that would have been real useful information before I blew out a $330 pair of IGBT modules.

So I bought a reconditioned interconnect board, before I buy another pair of IGBTs I want to make **** sure this isn’t going to happen again. So viewing another thread on the this board I saw some information to check regarding other components.

Capacitors read infinite ohms one way, dead short the other. Connecting red to + and black to the other terminal and putting meter in cap test mode I get ~2450 uF when the label says 2700 uF @ 450 VDC on the capacitor. Cap vents appear fine and caps don’t look swelled. Is this value within tolerance?

Checking SR1 with diode tester on my DMM I get .43-.44V across all 6 diodes. Whats the spec on this? Nothing from any terminal to backing plate.

Anything else that can go wrong, before I make another IGBT go pop and then curse a lot?

Both times the rear IGBT was the one that tested bad (when first got dead machine and after blowing out an IGBT) if that means anything.

Can anybody link or email me the correct tech manual for this machine?

[cruizer]

Cause if you replace a fizzled IGBT, the protection diode to the left (on the board) and next to the screw on the bottom of the IGBT is also lit. Pretty easy to order and replace as apposed to getting a $$$ output board

[greenbuggy]

Cause if you replace a fizzled IGBT, the protection diode to the left (on the board) and next to the screw on the bottom of the IGBT is also lit. Pretty easy to order and replace as apposed to getting a $$$ output board

Well I did order a reconditioned board, but my concern is mostly what if the problem is with something outside of the bad IGBT and IC board components thats causing the IGBT to fry?

[cruizer]

Will be fine after you replace the output board and a new set of igbt's

[cruizer]

Before all this happened, did you change the contactor....

[greenbuggy]

Before all this happened, did you change the contactor....

I bought it as a nonworking unit, though contactor does not appear to have been changed and I am running it on single phase 240.

[greenbuggy]

Will be fine after you replace the output board and a new set of igbt's

Who is going to pay for another set of IGBT's if things aren't fine after replacing those two?

The eBay seller who is selling reconditioned IC boards tells me that if the control board goes bad and tried to switch the IGBT gates too fast that the IGBT's will blow again regardless of condition of the IC board. The tech manual doesn't say anything about how to test the control board outputs to the IGBT's to prevent that from happening & I'd rather not replace every PCB in this machine just to be "safe" I want to get it running again without spending a dime more than I have to.

[H80N]

Who is going to pay for another set of IGBT's if things aren't fine after replacing those two?

Why you are of course.....

You are the one making the decisions and attempting the repair

It was your decision to "Save Money" and do it yourself....

any advice you accept or reject is totally on you

with that possible savings you also assumed the risk of failure

Had you engaged an authorized service center with trained experienced techs and the proper tools to support them the story would be different...

You cannot have it both ways

[greenbuggy]

Why you are of course.....

You are the one making the decisions and attempting the repair

It was your decision to "Save Money" and do it yourself....

any advice you accept or reject is totally on you

with that possible savings you also assumed the risk of failure

Had you engaged an authorized service center with trained experienced techs and the proper tools to support them the story would be different...

You cannot have it both ways

That was precisely my point, I don't expect quasi-anonymous members on an internet forum to foot the bill for bad or incomplete advise regarding this repair, though it would be nice if Miller would throw me a bone since I just blew $330 as a direct result of following one of their techs advise. My hope is that I can get the attention of Miller regarding an (IMHO) incomplete troubleshooting procedure in the tech manual.

[cruizer]

Thing is , YOU ARE NOT A TECH, we techs pay quite a sum of money to fly in somewhere and take these factory courses. We have to know how things work, we also get tested on this stuff , True, we make mistakes too, however, IGBT's on that board or any board do not blow all by themselves, and always take out other things. There are prepower check lists. and you need really good meters to check stuff. Take it to a Miller service shop, then blame them if they screw up. In this case there is no one to blame but yourself....

[greenbuggy]

Thing is , YOU ARE NOT A TECH, we techs pay quite a sum of money to fly in somewhere and take these factory courses. We have to know how things work, we also get tested on this stuff , True, we make mistakes too, however, IGBT's on that board or any board do not blow all by themselves, and always take out other things. There are prepower check lists. and you need really good meters to check stuff. Take it to a Miller service shop, then blame them if they screw up. In this case there is no one to blame but yourself....

I'm an industrial technology specialist and I work with VFD's and PLCs quite a bit, this is not the first welder I've repaired and I've got myself a good DMM, nice oscilloscope, IGBT tester and a host of other tools for diagnosing problem and repairing them with. I've got myself a copy of the Miller tech manual for this machine T1327Q, I've done the prepower checklist. I cannot stress this enough, the pre-power checklist DOES NOT LIST ACCEPTABLE TOLERANCES for the capacitor pair hooked to the IC board, it does not list the acceptable specs for the rectifier module, on power up it does not give me waveform or acceptable frequencies for the output from the control board to drive the IGBT's, thats why I came here to try and get answers on these items because their documentation is lacking.

You're right, Miller techs pay good money to fly in and take factory courses, but I expect that their memories are not perfect and they are not clairvoyant.

[cruizer]

my actual trade is in Avionics, never the less had you checked the protection diodes, for the IGBT's, the likelyhood of a igbt blowing again would have been significantly reduced to zero.

[greenbuggy]

my actual trade is in Avionics, never the less had you checked the protection diodes, for the IGBT's, the likelyhood of a igbt blowing again would have been significantly reduced to zero.

Did you even read what I wrote? I am moving forward with a reconditioned IC board and new IGBT's. I *REALLY* do not want to blow up another $500+ worth of hardware. Replacing just those three components does not guarantee I won't smoke some more expensive hardware, thats why I have concerns with cap values, rectifier module testing and the output from the control board.

[H80N]

Thing is , YOU ARE NOT A TECH, we techs pay quite a sum of money to fly in somewhere and take these factory courses. We have to know how things work, we also get tested on this stuff , True, we make mistakes too, however, IGBT's on that board or any board do not blow all by themselves, and always take out other things. There are prepower check lists. and you need really good meters to check stuff. Take it to a Miller service shop, then blame them if they screw up. In this case there is no one to blame but yourself....

EXCELLENT ADVICE............
.

Did you even read what I wrote? I am moving forward with a reconditioned IC board and new IGBT's. I *REALLY* do not want to blow up another $500+ worth of hardware. Replacing just those three components does not guarantee I won't smoke some more expensive hardware, thats why I have concerns with cap values, rectifier module testing and the output from the control board.

If you let the smoke out of even more parts.... You are the only one culpable

Why not Bite the Bullet & Take it to a Miller Expert.........??

[greenbuggy]

If you let the smoke out of even more parts.... You are the only one culpable

Why not Bite the Bullet & Take it to a Miller Expert.........??

Because the cost of smoking more parts is built into a service tech's price for work. You really think if they misdiagnose or fail to diagnose the same items I'm having concerns about, and blow out an IC board and another set of IGBT's they aren't going to pass the cost right on to me? I called the Miller certified guy I've previously done business with (he also repairs Hypertherm machines and I had him work on a plasma of mine) and he doesn't have another MM350P that could swap parts to aide with diagnosis on. And as I said above, I don't particularly want to replace the control board if I don't absolutely have to, looks like control board is going to cost me at least another $500 in addition to other parts that need replacing. And thats assuming I don't need to match my control board with a different pulse board which would be another $200 on top of that.

I've got the tech book, I've fixed more than a few welders before, I've got access to a lot of the same tools, and the documentation is lacking. What exactly is going to make a Miller tech know what values are OK if Miller doesn't publish the specs on them?

My other question is this: I pulled the control board out part number on backside is 239873. Looking thru the parts book I see a couple of different boards listed for this machine depending on serial - for my serial & voltage it appears that the correct board is 223542. Does the number in the parts book cross or supercede to 239873, or did someone put the wrong board in this machine? Could that be the cause of my IGBT problem?

Edit: did some searching, board 233542 supercedes to 233597. Board 233597 appears to still be current. Google doesn't seem to find much of anything for board 239873