Old Model 250 ac/dc HF Problems

[Jmcghee]

Hey guys-

I recently picked up an old transformer machine, Model 250 ac/dc HF tig & stick. I wired it in (100amp breaker) gave it a quick bath, snugged everything up, hooked up a bottle, & started welding. I put it through it's paces, and everything went great. Spent probably an hour or so trying everything... tig, stick, ac, dc, hf start & continuous, mild steel, aluminum, etc and it all worked perfectly. The only adjustment I made was on the HF spark gap, the manual called for .008" & it was way over that. I also replaced all the tig torch consumables with new. After getting acclimated I had lunch, came back a couple hours later planning on getting started building a small welding table. I'm not sure what happened in that time, but the machine is no longer happy in DC. AC works perfectly, but in DC I'm getting a pretty strong vibration that directly relates to what the amperage is set at (150 amps vibrates hard enough to shake a screwdriver off the top of the machine). It also won't really operate in DC. It'll try to strike an arc, but just barely.

I'm way out of my depth as far as diagnosing this & was hoping somebody might be able to steer me in the right direction? The machine sat for about 10 years before I got it, but everything seems to be visually fine. There's no obvious burned/frayed/loose connections, leaking caps, etc. Anyone have any idea(s) what might be going on?

[Noobie]

It's gotta be something with fan doesn't it? Or indirectly affecting the fan, (rectifier is right there). I'm definitely not an expert but I can't imagine what would make a 350lb welder vibrate like that!

[Aeronca41]

I'd be surprised if the fan has enough mass to cause that much vibration of that much weight. And a fan problem doesn't seem likely to affect only DC. Maybe a shorted diode? Has to be drawing a lot of current somewhere to shake it that much. You can check the diodes with a multimeter. Is the rate of vibration around 60 Hz, similar to the hum you get when welding on AC?

[Jmcghee]

The fan runs just fine and you’re right... the blade has nowhere near the mass needed to vibrate the machine like it is.

Honestly I can’t tell if the rate is the same as on ac... it purrs like a kitten on ac.

How would I go about checking the diodes?

[Aeronca41]

Please post your ser no. so we can look at the correct manual. Are you familiar with things electrical? Do you have a multimeter?

[Jmcghee]

Serial # is R379956. I do have a multimeter & what I’d describe as a base level knowledge of electrical (I built/wired my shop, phase converter, 460v transformer, machines etc). I know just enough about welders to know that this is something I need help with.

[Aeronca41]

Is this a Miller Dialarc 250 AC/DC? Their serial numbers for that machine don't start with R. Is this an Airco?

[Jmcghee]

It’s neither a Dialarc nor Airco... I believe it’s the same/similar to a Dialarc but older?

[Aeronca41]

Yes looks like "kinda" a Dialarc. Can't find a manual for it. Give a call to Miller tech support and see if they can get you a manual from the serial number. They have posted or emailed manuals when I couldn't find them on the web site. Very helpful guys.
Really need to see what the inside looks like--need to see if they are "hockey puck" diodes, stud mount, or whatever, to tell you what to look for.

[Aeronca41]

Took a look at the pix and manuals you sent. I was a bit surprised that Miller didn't have parts lists and internal exploded diagrams of the HF version you have, but I guess back in the dark ages that may have been in a different manual they didn't send you (or don't have any more). Please don't be offended if I get into too much detail that you may already know; just not sure where you are at electrically and want to be clear. First and foremost, be sure the welder is unplugged. Then, take a screwdriver [that you don't care a great deal about :-) ] with a well-insulated plastic handle and short across the terminals of all of the capacitors in the machine. You don't want to get zapped by a stored charge in a cap. Sounds like you have enough electrical knowledge to identify the capacitors. If not, please ask.

The diodes are clearly shown in your fourth photo. They are the round things mounted to the vertical heat sink plates with the braided cables coming off of them. You will have to disconnect one end, either the stud or the cable, of each one to check them. I would probably opt for the cable, since there may be heat sink compound on the stud and you probably don't have any lying around to recoat them when you bolt them back on. So, disconnect one at a time so you don't mix up any connections, and check with your ohmmeter. Put one test lead on the stud and one on the cable; you should read either a very low resistance, or very high (OVERLOAD or OL on many meters). Then, swap the leads. If you read high before, it should read low now, or vice-versa. If you read the same with the leads either way, the diode is bad. I suspect you are going to find one that reads very low both ways (shorted diode). A high reading both ways indicates an open diode, but that would not cause the excess current that must be flowing to make the vibration you're talking about. The good news is this is a pretty simple machine electrically, so there is not a lot to go wrong. Ya gotta love these old battleships! With some very simple and generally low cost maintenance, most of these things will still be running long after all of us are dead and gone.

See what you find out.

[Jmcghee]

Well I followed the above test, and it looks like all 4 of the diodes are bad. 3 of the 4 had no continuity at all (0 reading on the meter) in both directions, and one of them was shorted in both directions. I have to imagine something else in that circuit cased them all to blow?

[ryanjones2150]

I don't know what year your machine is, but when I was working on my old Helga, the folks at Miller recommended I replace both diode bridges. If you recall, Wayne, Helga was born in 1974....anyway, mine are still working and I generally don't like replacing stuff that's working. The Miller guy told me that it would be an updated part, but easy enough to make it work in my old machine. Seems like I remember him telling me that he expected 20 years out of those old ones. Don't quote me on that, just recall a discussion of the sort.

[Aeronca41]

A zero ohms reading reflects full continuity, not lack of continuity, that is, the diodes are shorted--they have turned into "wires" instead of diodes. That definition is a common misconception. Continuity = zero ohms = shorted. Infinity/overload ohms = high resistance = open. Exactly what I was expecting--only thing I could think of that would cause that much current to be drawn. If this had been any other transformer operated device than a welder, you would probably be needing a new transformer. I doubt very much the transformer was hurt--it would have smelled pretty bad.

What were the readings on the one you called shorted in both directions? Were they high/OL?

There isn't really anything else in that circuit that could cause the diodes to blow other than old age (they do sometimes die of old age) or a power line surge of some significance toasted them--that fact that they're all bad sort of leads to that in my thinking. Seems unlikely they all reached "natural death" at the same instant. Unless HF got to the diodes somehow, but that seems unlikely. There are some small capacitors across the diodes to protect them from transients; I would replace all 4 of those just on general principles when you do the diodes since they only cost pennies. They are the little tan/yellowish flat things hanging near the diodes. I didn't see a parts list in that stuff from Miller, so you are going to be looking for manufacturer part numbers on the diodes, and values for capacitance and working voltage on the caps. There are probably two flavors of diodes, commonly called the "straight" and "reverse" (as in welding polarity), but that term is a bit misleading since all 4 are used for both welding polarities. On welder diodes, it relates to whether the stud is the cathode (-) or the anode (+). It as possible the design of the welder allowed for use of 4 identical diodes, but it's probably more likely you will have two flavors. Look carefully and see if you can find some numbers that probably start with 1N... on the diodes. Also look at the caps for any numbers on them. Miller may be able to give you part numbers for the diodes, but they will be 6-digit miller numbers, which you can probably get from Miller4less.com. However, I would expect you will find them cheaper at Mouser, Allied, Digi-key, etc. by the generic number. Probably not always, but usually. Let's see what you find written on them and we'll go from there.

I suppose there is the outside chance that the diodes were taken out by a short internal to the transformer that put the 230 V primary power on the low voltage secondary, but I place that pretty much in the same category as a snowstorm in July (in the northern hemisphere)--yes, it can happen, but mighty unlikely. And, again, I think you would most likely have smelled it.

Good thoughts, Ryan. I didn't see your post until after I had typed all this and posted it. Agree with all.

[Jmcghee]

I suppose I didn't explain it that clearly... the one I'm calling "shorted" had continuity (~.05 ohms, meter beeps) in both directions. The other 3 show absolutely nothing (no beep, no reading of any sort). Not even a little blip when I first touch the leads... the same result as if you touched meter leads to a piece of plastic or to nothing at all

[Jmcghee]

I’m hopeful for the transformer as well. There was no smell, smoke, pop, etc.

Isn’t it extremely unlikely that all 4 diodes failed in a matter of hours without something else causing them to?