Need help reviving a dead Miller Dynasty 200DX

[adatesman]

Took a chance on a Craigslist Miller Dynasty 200DX yesterday, and "I promise it works" ended up being "it doesn't even turn on". No need to tell me what a bad idea that was; I already know. In my defense, it's kind of hard to test a welder in the parking lot 3.5 hours away from home without a power plug on the machine or torch/stinger/ground/pedal/gas/regulator (supposedly all of that got sold with the other welder he had, leaving this just a bare machine, although the lack of a power plug is rather strange).

Anyway, I got home and hooked up a new 120v plug. When I flip the power switch I hear the contactor in the click over, then there's just a hum from the transformer at the front of the machine. No lights on the display and no fan. Same goes for hooking it up to 220v.

I pulled the cover, and it's fairly clean inside (no mouse nests, minimal grime), no bulging caps, and no scorch marks. Pulling and reseating the two cards on top (and all of the cables that go to them) didn't make a difference (cards are LD11 and LD12). On the off-chance it was simply the display not coming on I checked for voltage between the connectors for both AC or DC, but nothing.

Any suggestions where to start troubleshooting it would be *greatly* appreciated.

thx.

EDIT: Just noticed a date sticker on the inside... The machine is from 2003.

[7A749]

Yea, unfortunately, the "Ill just meet you somewhere" and end up with a dead machine scams are quite common on CL. I've learned the hard way taking someone at their word too. It happens.

PM me your email address and I'll email you a tech manual for it. These machines are very complex, and parts, incredibly expensive. You almost might be better just selling it for parts on eBay. Hopefully you didn't pay too much for it.

The tech manual will guide you thru a quite lengthy pre power check that will involve taking the machine apart to get to the board on the underside of it. You'll need a good multi meter to perform the checks. I'll also email you a testing booklet Miller electric provides to their techs that will show you how to test individual components, and how to identify them. The checks are going to take some time. Doing only part of them is as good as doing none. I've only had one Dynasty 200 in the shop and after finding a couple pretty serious things wrong, I stopped and dealt it to a friend for doing me a favor. I believe he ended up selling it for parts and did very well on it.

Good luck with it.

[adatesman]

Thank you *SO MUCH* 7A749! Greatly appreciated.

BTW, my background is in engineering and I know my way around electronics fairly well, so the good multimeter is already out and I've been slowly working my way through tracing power from the switch onwards looking to see if it's something simple like a bad connection. I don't particularly care for working blind like this though, hence asking for help.

[adatesman]

On a side note, I'm very relieved to see that the boards are through-hole rather than surface-mount, as it means replacing any failed components is easy. If I can find them, that is...

[7A749]

Check your email. Start at the PP checks. They're lengthy for that machine, but you might get lucky.

ccawgc here is a great resource for tech info here. He might be able to offer you some assistance if you ask him. He's one of the sharpest techs I've ever met.

Post up how you make out.

[adatesman]

Thanks again, 7A749!

Well, I've worked though all of the pre-power checks in the Tech Manual and everything looks fine as far as they go (found a copy online before getting your email). Moving on to section 7-10 for checking voltages and resistances, all of the resistances are withing spec except for R1 (the fan), which is 0.944meg rather than 1.5meg and R11 (a thermistor) which is 36.6k instead of 30k. Neither of those things strike me as being so far out to keep the machine from powering on, so I went ahead and applied power to check voltages.

Nothing, nada, zip. No readings on V1, V2, V3, V4, V6, or V7. (since the display doesn't come up, I can't set the output amps to check V5)

Seems to me that's probably an indication that either the power interconnect board or the inverter control board is dead.

Sigh.

Off to poke at them some.

[adatesman]

Hm. This isn't making much sense... All of the diodes in the pre-power check of the Power Interconnect board were good (as were all of the other diodes it said to test), so in theory there should be power of some sort getting /somewhere/, yet I'm not seeing any. Which is weird, given there's continuity on all of the transformers.

Think I'll pull the boards and check for bad solder joints.

[rlitman]

You're sure it's not something simple, like the wrong wiring in the plug?

[adatesman]

Yup. That was the first thing I checked. Even pulled the 120v plug and used a 220v plug/outlet. Same result.

[rlitman]

Hey, I had to ask.
Ok, what about checking fuses? After that, the power path in an inverter goes from the line to the rectifier.

[adatesman]

I appreciate you asking. Easy to miss the simple stuff.

Speaking of which... Fuses? If there's fuses in there I don't see them, either on the machine or mentioned in the manual. And yup, that's how I recall these things working... Line to rectifier, then off to the electronics. Which is why I'm surprised to not be seeing voltage anywhere.

I did notice one thing on the bottom of the Power Interconnect board though... Looks like it overheated at R1, R2, R36, and R37. There is still continuity from the leads to elsewhere on the board, so the board itself is ok. No telling if the resistors are good without pulling them, near as I can figure.


Not visible from the top:


Where it is on the board:


The Inverter Control and Boost Control boards look fine on both sides, and the power switch appears to be functioning properly. Also there is a 24VDC relay on the Power Interconnect board that I can hear clicking on a moment after turning it on, so clearly power is getting somewhere. Tired of working on this though, so will punt tracing those circuits until tomorrow.

[adatesman]

Actually, rather than trace the circuits I decided to push this back on the guy I bought it from. I have messages from him explicitly stating that it works, he's not trying to unload a dead machine, and that he was unwilling to move on price just because he didn't have a torch/etc to verify or me that it works. Seems a pretty solid case for me in terms of small claims court, so it's not really worth my time to poke at it further since it's clear that at least one of the boards is dead.

Thanks for the help, Everyone.

[shovelon]

What did you pay? If you don't mind me asking?

[adatesman]

What did you pay? If you don't mind me asking?

Well, seeing as it won't turn on, I'm gonna say "too much".

Seriously though, the guy is doing right by me so far and still responding to texts and emails, and I just sent him a video showing what I did to test it. Local repair shop wants $100 to evaluate it / perform diagnostics, and we're discussing what to do next.

[soutthpaw]

On a side note, I'm very relieved to see that the boards are through-hole rather than surface-mount, as it means replacing any failed components is easy. If I can find them, that is...

Only good on a blue machine. If it was Chinese made, everyone and their mother would be here trashing the through hole design. Just see the recent Chinese welder teardown thread. Double standard, me thinks...

[adatesman]

Only good on a blue machine. If it was Chinese made, everyone and their mother would be here trashing the through hole design. Just see the recent Chinese welder teardown thread. Double standard, me thinks...

For me it's more showing my age... Grew up with through-hole and Radio Shack actually having everything you'd need to build a computer from scratch on the shelf, and IMO through-hole tends to be much easier to troubleshoot/repair using easily obtainable parts. Well, back when Radio Shack actually had parts on the shelf... This surface mount crap? Sure, I could order parts from Mouser or Digikey, but the boards are usually designed to be simply replaced rather than fixed so are hard to troubleshoot at the component level. For whatever reason, this bugs me and I'm always happy to see through-hole.

[kctgb]

Good luck, the circuit board in my dynasty fried itself one month after the warranty ran out. $1500 to fix the damn thing, I sold it after it was fixed. I don't like buying expensive inverters any more after my experience.

[WorldPowerLabs]

I do miss the Radio Shack of old, where you could actually buy useful components (not just 50 cents worth of TTL gates sold for $9). For prototyping, certainly, I prefer through-hole construction. If I'm repairing equipment, however, I find it slightly faster to repair SMT -- desoldering a component tends to be quicker, easier, and cause less potential damage to the PCB (because you're not trying to heat up the mass of a plated-through hole). Need to remove and replace a 100-pin chip? No problem -- just grab the hot-air tool, remove the chip, clean up the pads a bit, and install new (couple different methods to do that). Need to replace a (non-socketed) through-hole chip with many pins? Unless you have an excellent desoldering station, it's a pain in the...

When I design products, I'm often forced to use a mix of SMT and PTH (pin-through-hole), which is not ideal from an assembly cost/complexity standpoint, but it's often inevitable because of physical constraints (for example, I refuse to use surface-mount versions of connectors or switches that will see heavy use, as the solder joints on these sorts of components are often their weak point when they're subjected to physical abuse)...

Surface-mount technology was driven by the need for smaller, faster electronics -- not because it's always "better"...

For me it's more showing my age... Grew up with through-hole and Radio Shack actually having everything you'd need to build a computer from scratch on the shelf, and IMO through-hole tends to be much easier to troubleshoot/repair using easily obtainable parts. Well, back when Radio Shack actually had parts on the shelf... This surface mount crap? Sure, I could order parts from Mouser or Digikey, but the boards are usually designed to be simply replaced rather than fixed so are hard to troubleshoot at the component level. For whatever reason, this bugs me and I'm always happy to see through-hole.

[adatesman]

Well, true. But if you don't have a hot air gun, SMT gets annoying fast.

In any event, I just dropped the welder off at the local independent welder repair shop so hopefully I'll find out sometime soon what's wrong with it. Gotta say though, I'm not all that thrilled with his initial response that "you do know that these things often die if they're left to sit unused, right?" If that's not a reason to go with a much cheaper import machine, I don't know what is.

[7A749]

Hopefully you'll get some news other than the main and booster boards are fried $1800 to replace with no guarantees the problem will be fixed afterwards.

The main thing about inverters is that they are in series, interdependent systems. If one link in the chain fails, everything else goes down, and several other components can go with it in the process. The guy who sold it to you likely knew it was junk. I got burned on an eBay deal once a long time ago like that. This was long before they would give you back your money if you simply didn't like the color, etc. Anyways, the guy really went out of his way to state it was an AS IS deal, but still implied it was a working unit. It wasn't, and I bet my right arm he knew it.

Even if you had extensive knowledge repairing PCBs, if you miss one thing in the power circuit, and fire it back up, you'll likely take out all the parts you just spent all that money and time fixing, not to mention other stuff could go as well. I mainly work on older Miller inverters so I'm not much help with something as complex as this.

My word of caution from this point would be NOT to throw parts at it once the service center finds out what's wrong. Unfortunately, a lot of service centers do little more than replace parts these days, and if dude tells you such and such board is bad, you go buy one and put it in yourself, if he didn't catch other issues, you just blew up a $700 board.

It's almost a paradox. Tough call. I'm betting the diagnosis won't be cheap either way. Depending on the age, you may be better off taking this one on the chin and just forking out the dough for something else. I've been there too man. It can happen to anyone.

Please let us know what you find out.

Oh, and guys. No need to bring the whole import/circuit board design argument to this thread. It isn't gonna help the OP either way at this point.

Thanks.

[adatesman]

Well, I've heard back from the repair shop and they weren't able to get it to do anything either, even after calling Miller for help. Their only option was to swap in 2 new boards for $1500, which may or may not work and may or may not blow out the new boards. Waiting to hear back from the seller to see what he wants to do.

[7A749]

Damn. I was afraid of that.

Hope the guy offers you some resolution. Some would say go looking for him, but that never usually ends well.

Now if my long lost uncle happened to be Michael Corelone....

Good luck with it.

[adatesman]

Yeah, here's hoping....

Thanks for the help, everyone!

[ccawgc]

Did they check the 800 volt buss? if only line voltage, the boost circuit is not working,
Unfortunately it could by just the boost pcb or like the last two I repaired it took the boost and the interconnect and the power module. To get it to come to life.
Just a soft click and hum points to the boost circuit not working. If it is working you should hear some electronic switching sounds. Clicks and squeaks.

[adatesman]

To be honest, I don't know what they checked as I kind of tuned out upon hearing they had no luck and $1500 since IMO the machine is going back.

But "click and hum" is all it does; no other sounds. Also, looking over my notes I stopped testing things long before that point in the Tech Manual.