Miller Syncrowave 300P AC amperage problem.

[trigg3r]

OK so I have an Interlas built 300P from about 1980'ish and with the exception of the usual HF stuff (now sorted thanks to @NotANormalCoder ) has been fine for the last 20 or so years. Only a hobby machine for me really, so no really heavy use anymore.

However, all of a sudden whilst I'm welding in AC, the weld pool is uncontrollable and the power seems to 'spike' destroying tungstens and anything in its path. I've tried it with two different torches, with panel and remote current and contact via a pedal and still the same. Like huge instant variations in amperage.


DC welding is fine and HF seems fine, ground all good.

Anyone any ideas or seen this kind of behaviour, before I rip the thing to bits?

Thanks in advance.

 

[Jim Davey]

Give the machine a damn good clean. Metallic dust can cause havoc in those machines but if it still persists after that its worth replacing the snubber capacitors on the welding terminals. After that it's an overhaul of the big PCB (about £95) and check of the big flywheel diode in the rectifier. After that if that is no good then its replacement of the SCRs in the rectifier.

 

[trigg3r]

@Jim Davey Thanks for the reply. I'll get into it tomorrow to clean up. I've never ventured into the black box containing the PCBs, maybe it's time for a spring clean.

 

[trigg3r]

OK so I checked the boards visually this morning. From what I can see there's no obvious damage or failure evident.



Time for some more digging.

 

[trigg3r]

OK so after taking the front vent panel off, it was pretty ****ty and dusty behind there.

This is a picture after the blow gun!



I take it the caps in the photo below are the scrubbers you mentioned earlier @Jim Davey ?

 

[Jim Davey]

No, the snubbers are on the welding terminals. Check the insulation on the terminals themselves isn't compromised.

 

[bwservices]

Check the ac balance pot isnt on the way out.meter it and check the resistance is changing steadily

 

[trigg3r]

AC Balance pot checks out OK.

 

[trigg3r]

@Jim Davey Bit of an update. Cleaned the machine down, all the caps and resistors mentioned in the service manual look fine after a visual inspection. No loose connections etc. The manual also states that even if an SCR doesn't check out, look at the associated cap/resistor first, so i assume the SCR's are fairly robust? So, what's the general consensus? Big PCB overhaul first? If so, who do you recommend?

Thanks in advance.

 

[Jim Davey]

Check the gate resistances on the SCRs first. They should all be pretty close, fairly low ohms, around 20 from memory but don't quote me.
As for who to do the PCB test/overhaul you'll do no better than Malcolm at MS Welding in Leeds.

 

[NotANormalCoder]

Either a failing thyristor, gate leads to thyristors or invisible PCB fault, or even the Hall sensor. This gismo is on the left hand side (facing the machine), about half way up and is a smallish PCB with one or two heavy welding cables passing through the centre. It's job is to send a signal to the main PCB, relating to the actual output current. The main PCB then adjusts this if required, to match the current requested. If it goes dodgy, it is possible the machine says 'Hey the output current is very low, let's give it a bit more'. It is fairly rare for these to fail but checking the connections to it would be a good thing. While you're there, take a good look at the rats nest of heavy cables in that area. I have sometimes found shorts between the connections there that give random problems.

HTH Dave

 

[trigg3r]

OK, apologies for a delay in responding.

Having spoken to Malcolm @ MS Welding (extremely helpful guy too ) I've plodded through a bit of testing and found the following..

SCR's measured as follows from L to R from front.

1) 0.98 Mohms (20M scale) - 1.320 Mohms (2M scale)

2) 1.98 Mohms (20M scale) - No reading (2M Scale)

3) 1.43 Mohms (20M scale) - No reading (2M Scale)

4) 1.03 Mohms (20M scale) - 1.432 Mohms (2M scale)

Obviously the infinity reading are due to scaling, but that's a fair difference on the 20M scale?

All gate resistances are around 23 ohms.

Not entirely sure how to test the gates are working. Touch the gate lead to the anode (+) of the SCR apparently? Can anyone shed some light on this and is it necessary?

One more thing. When I first got the machine, the gold chassis mount type 10ohm 50w resistor shown in the picture below was obviously added at some point from the last scrubber capacitor heading into the loom on cable 18. This has now failed and will be replaced, but wondering if anyone has any ideas about it?

In the mean time, i'll replace the resistor and give the machine a try.

Edit - Resistor replaced - same fault.

Cheers