Fixing a GYS Smartmig 162

[KemppiFrog]

Couple of thoughts: If the two whites are a low voltage feed to the board and the transformer is not working/powered up, then your wire feed would not work? Anything on the display?

You are a bit far for a look in.... I am just south of Rochefort.

 

[Sam_]

You are a bit far for a look in.... I am just south of Rochefort.

That's a long way for sure, thanks for the intent though.

If the two whites are a low voltage feed to the board and the transformer is not working/powered up, then your wire feed would not work? Anything on the display?

True, that's why it must be a overheating sensor. There is no display, just a pair of switches and a knob to adjust wire speed.

Just tried after work to measure again the line coming from the board and... tension was present but very low(20 something AC). Last time I did got 240V. Tried to plug the transformer directly to line and tadaa ! Nice buzzing/humming sounds.

The great news are: it works ! Transformer is not melt and that really is a relief. Now I have to hunt down the problem on the board. Will try again tomorrow night, but I can for sure "fix" it by plugging the transformer directly to the main switch, it should work as well but that way I wouldn't have overheating protection which must be done by the board.

The board is pretty simple but really difficult to remove (one way plastic retainers). Maybe just a faulty relay.

 

[ms08]

Might be a dry joint on the board - worth flashing over with a soldering iron first? - unless something obviously burnt out.

 

[Sam_]

Yep, bad solder or anything. I'll get the board out of the welder and see what's faulty. Board is quite simple, very few components. Will try that tonight if I can

 

[KemppiFrog]

Yep, bad solder or anything. I'll get the board out of the welder and see what's faulty. Board is quite simple, very few components. Will try that tonight if I can

You were talking about the board fixings. They work like barbed fish hooks. If you have a pair of those little electrical pliers, with bent ends, put a bit of lift pressure on the board, while pinching the part of the plastic that sticks through.
KF

 

[Sam_]

If you have a pair of those little electrical pliers, with bent ends, put a bit of lift pressure on the board

Very hard to reach, I made my way using a flat screwdriver and a fair amount of patience.

Board is out and doesn't look bad :

If was suspecting the relay (bottom right with the sticker on it) to be faulty, but turns out it's the relay for the wire feeder motor. It clicks when I pull the trigger.

The red on left is the wire that goes to the transformer and should be line. It has 240V with line, therefore grounded. On the bottom side of the board, it seems to be connected with the relay.

Everything seems to work: fan, feeder, even the overheat light. If I disconnect one of the overheat switch wires, the light turns on. I need some time to think it over, but maybe the easiest thing to do is just bypass the board to power the transformer.

The board is quite simple, no microcontroller, the chip on the board is an "operational amplifier". Interesting thing to note though, while the (+) and (-) signs are cleary marked for the feeder motor, it turns in reverse. If I switch black to + and red to - it turns the right way.

 

[KemppiFrog]

Off the cuff.. The rust coloured block is the transformer for the board. So it has it's own power supply. The blue bit is an over voltage protection and the bunch of diods to the right of it are a rectifier. Presumably you have pulled the fuse to check continuity?

 

[Sam_]

Yep, the fuse was the very first thing after opening the welder, but unfortunately it's good.

Top left black "bar" is also a relay, according to the datasheet it should be like that:

I'll test it tomorrow. But as there are no other relays and the first one is the feeder it might be it

 

[Sam_]

Got some progress today: Applying +12V to the coil of said relay makes the transformer buzz, so the problem is that the relay doesn't get fed with appropriate power.

Might be related to the fact that the wirefeeder turns in reverse. I have asked GYS for schematics but didn't get any answer (I wasn't expecting much from them...).

Maybe I'll have to resort to reverse engineering schematics from the board but I'm not sure I'll be able to do that :/

 

[KemppiFrog]

Not had a reply about the price of a board. I will ask Ashley if he can help.

Just wondering how they switch four power outputs on that board, other than the switches on the panel do it and the wires are routed through the board.

 

[Sam_]

Just wondering how they switch four power outputs on that board, other than the switches on the panel do it and the wires are routed through the board.

The power comes from the board and then goes into the line switch which in turn selects a winding. The neutral comes into the other switch that also selects one of the other winding on the transformer.

 

[Sam_]

I don't want to leave the tread as-is without conclusion, so:

I managed to "fix" it 2 weeks ago. I don't even know what I actually did. I applied power to the relays to hear the "clac" and then... tada ! When pressing the trigger the transformer is now powered and I can weld.

I did test if the thing would work with the trigger on, put all what it did before was to unroll wire while touching the ground clamp with no spark and no nothing.

So case closed it now works. I didn't find any option to add [SOLVED] to the tread, so if someone could do it, that would be nice.

 

[ms08]

Sticky relay which once operated now works again. At least if it happens again you know where to look and possible replace.