TIG welder possible triac problem?

[Chris Wilson]

After 12 years of sterling low current usage my trusty and dry stored Migatronic 220 MTE Dialog has thrown a wobbler. I run it on UK single phase mains, on a 80 amp supply. I very rarely run it at all hard, I am a race car engineer and most jobs are thin ali work, and thin stainless. The AC "over voltage on the torch socket" warning light is on all the time, even when idle and just powered up in AC mode. On AC the current is uncontrolled at the torch when the torch button is pressed. If I turn it to DC with the unit powered up it instantly draws MASSIVE current off the single phase mains, and emits an alarming low hertz buzzing. This is without the torch switch being activated. I just have to switch it to the DC setting. I took the covers off (I am a licenced radio ham, so have some electronics knowledge, and understand good high voltage and high current safety techniques), and had a quick poke about. R1, a 10 ohm wire wound over a grooved  ceramic former resistor with a slidable tap, sitting on top of the cooling fan, has heated to failure. The thing is about a foot long and an inch in diameter, but fed with very thin wires. The section from M3 pin 3 to M3 pin 10 is the section that's had extreme heat, and the ceramic former is shattered. The other "half" shows no sign of distress. The overheated section has failed open, and is dangling into the fan blades. I am trying to understand the schematic and work out what would have caused one section of R1 to heat so badly. I suspect a triac driver issue or one of the 2  thyristors in the secondary has failed, but I am unsure. . Anyone familiar with the workings of these things? All the (Type I) control panel functions seem alive and well, the current digital display ramps up and down, gas control, blah blah still responds. It seems to be a control switch / board or rectification issue, (I think....). Thanks! The manual with schematic is at http://www.gatesgarth.com/manual.pdf

[Fank1]

Do a quick check of the SCR.s they fail shorted most of the time.
Just check continuity across them with the power disconnected you should read open in both directions.

[Chris Wilson]

Sorry for late reply, that damned nuisance thing called "work" interrupted things I have now stripped the machine partly down. The thyristors T1 and T2  are fine, as are all components on the control board, as far as I can tell. The choke (some regions call it the reactor when it comes to welders) is shorted between windings, but not to the frame. there's also a sign of an external arc over across from one of the external bolted joints that bond one half of the inductor to the other. This must have jumped at least a quarter inch air gap. Am I right in thinking this choke could see some very high voltages induced? I have a shudderingly expensive quote for a new one, which Migatronic would need to have built, and a slightly less shuddering quote to rewind mine. I need to have some sort of handle as to why it would short, I pulled apart the cotton covered extremities, hoping a short might be at an end, but of course, it's not and must be within the RH winding as shown in the photo of the brute. It's 11  by 11  by 7 inches and about 55 kilos.

The BY259, C1 and the 4 off in parallel 1uF caps (shown as 2.2uF on the schematic) to pins 2.3 and 15 of m3 are not shorted. I t appears the damage is limited to the shorted L1a and L1b choke windings.

Thanks for the tips Fank1.

[Fank1]

The only two things I know of that cause that big of a spike are lightning and a collapsing magnetic field.
Did the power fail when you were welding?
Also be sure and check the AC / DC switch before reassembly.

[Chris Wilson]

No, I have never suffered a power failure when welding, or even when the welder was plugged in. Migatronic inferred this failure was not at all unknown, the technician I spoke to suggested it as the first port of call when I told him of the symptoms. I was assuming that it saw high voltages in the windings. It has never been stored in the damp, in fact it's in a de-humidified workshop where bright steel and in buid engine and gearbox components are kept. It has a life of luxury I have metered out the switch contacts and all seems as it should be according to the schematic, and no false shorts or opens in any of the 3 positions is occurring.   Thanks for the reply again Fank1

[tautech]

Migatronic inferred this failure was not at all unknown, the technician I spoke to suggested it as the first port of call when I told him of the symptoms.

A very telling reply, probably the reason they told you they would have to have one made is that they had sold all their spares! 

I would unwind it and attempt a repair....you have nothing to loose.
Then immerse it in a transformer lacquer for good measure.

Just be sure before you do that is not a failed component elsewhere.

As Fank1 says back emf can be your friend or enemy.

[G7PSK]

The choke stops the HF from getting back to the rectifiers so if that has shorted out the hf which is also quite high voltage will got to the electronics in all probability. My experience with tig welders has been that virtually all problems stem from the hf (used for arc initiation and stabilisation)