Home Checking Welding Transformers Print

QUESTION-How do I check my transformer? The welder just blew the main fuses in the bus plug.

ANSWER-There are only a few things that could go wrong with a transformer:
1. A short, primary to ground.
2. A short, primary to secondary.
3. A shorted turn in the primary.
4. A magnetized core.
5. An open primary.
6. An open secondary (!!!).


Of all these, I see number 3 the most. I have seen number 6, which was caused by a new construction method for the secondary.

Here is more detail about these problems:

1. A short, primary to ground.
The primary windings are flat copper wire, varnished, and separated quite well from ground. The transformer would have to be cooked pretty well to have this happen.
This could be a primary to secondary short but it shows up this way because the secondary is often grounded.


Test for this with an ohmmeter between the case of the transformer and one of the input lugs.
In some cases this short may be a higher impedance and may require a "Megger" to test at 500 volts to see if it "flashes over". The megger is an ohmmeter that works at a much higher voltage. Your ohmmeter works at about 3 volts and will not duplicate the ability of the weld power to conduct through water, carbon tracking or jump small spark gaps.

In any case you should get a very high reading, 10,000 Ohms or more, any less and you have a shorted or leaky, (meaning "a path to ground") transformer.
Pri grounded


2. A short, primary to secondary.
The two windings are very close to each other, about 0.050" apart, and separated by a thin phenolic insulator. Long ago this used to be paper, and condensation would easily short this out. We even found one rebuilder had used newspaper for the insulation!

Modern transformers are pressure cast in epoxy, and are much better, but this could still happen.

The safety importance of a grounded secondary should be obvious at this point.

Test for this with your ohmmeter, or better yet, your megger, between one lug on the primary to one lug on the secondary. If this is a transformer with diodes built-in, you will surely destroy the diodes with the megger. Again you should have very high readings, well over 100,000 Ohms.


Pri to Sec shorts

3. A shorted turn in the primary.
This happens more often than you would think. Of course transformers are very reliable, I see hundreds and hundreds working for every one I see with a problem, but I have seen this many times.

This is simple to check, but I have never seen it done by anybody else this way, I don't know why...

The flat copper primary windings are extremely close together, and if they ever vibrate, thet could wear through the thin varnish that insulates them. Modern transformers with the pressure cast epoxy are much better at preventing this, but it still happens. The pressure cast transformers have a big "X" on the side, black = "480", red = "575".

First of all you can't check this with an ohmmeter, as the entire secondary of 40 turns or so is less than 0.01 ohms, and your meter can hardly read 1.0 ohms very well. A micro-ohmmeter would read this nicely, as it would read this as 10,000 micro-ohms. But you won't find a reference for what this should read, and a shorted turm may make it read less by a factor of only 1/40th, too small to determine this with.

However, a shorted primary will make the transformer draw nearly unlimited power (like 50,000 amps!) whereas a good transformer won't draw enough to blow a 5 amp fuse, if the secondary is open. Yes I have heard about the "magnetizing current" transformers are supposed to draw when first put on power, but I haven't seen any blow a 5 amp fuse, so I think this mysterious current is a lot smaller than most expect.

Check primary shorts

To perform this check, make sure there is no circuit on the secondary. Remove the cables to be sure, but if there is no slag across the pivot, you could just leave the gun open. Then attach a 5 or 10 amp fuse in a wire from the breaker to the transformer, and if you like being really safe, do the same in the other leg to the transformer. Apply line voltage and see if the fuses survive. If it blows a fuse, the transformer is bad. Be sure to write what is wrong with it to identify it, because some fool will put it on a machine 3 months later, then call you to fix it. This happened to me TWICE! The guy who was with me both times cut the transformer in half with a bandsaw to stop it from happening again. I was given the transformer slice years later for training purposes.

4. A magnetized core.
This can happen if the current was "imbalanced", meaning more in one polarity that the other polarity. Things that can cause this are:
a) SCR fired a couple half cycles of the same polarity.
b) Sealer or something in the weld stackup caused imbalanced current
by intermittenty being an insulator.
c) Repair or modification of the secondary that improves it so drastically that the first cycle of current blows the breaker.
This happened once on night shift, and they replaced everything, still kept happening, because the transformer got magnetized. I came in on days, degaussed the transformer and they couldn't believe it!
d) Loose secondary connections causing intermittent current flow.

It is easy to degauss the transformer, just fire it WITHOUT current regulation as low as you can go (20%)

Degauss

5. An open primary.
I suppose this could happen, it would be easily checked with an ohmmeter across the two primary lugs. A good transformer is way less than 1 ohm. I usually measure about 0.01 ohms, if I use a micro-ohmeter.

6. An open secondary (!!!).
This doesn't seem like it could happen, as the secondary is a cast copper "U" shaped piece. But a new design made of thin copper tubing, with the copper pads soldered on, will separate if the pads are subject to stress or vibration. While I haven't seen this open electrically, it does leak water major bad.

Please retain my name and company on this page as you share it, because I support my family by sharing my knowledge of welding., and I always need work.

-David Bacon, Update Technology 586 634 6784