Parweld XTM 201i "went bang"

[Tommy Tig]

Hi everyone, apologies to bring problems as my first post but i'm in need of some more experienced folks than myself to help diagnose this.

So I have a Parweld XTM 201i, its been a great little machine and never had any problems with it, until a few weeks ago it went pop upon turning on the machine (no trigger pulled, just plugged in and turned on made it go)

Now, being that I have no warranty and am also somewhat electrically experienced, I had the cover off to find a group of 3 resisters completely obliterated. I have no schematic for this machine and parweld wont even reply to my emails when i ask politely as a loyal customer for information that would cost them nothing to give.

I suspect one of the following two scenarios (or perhaps some strange combination of the two) has occurred.

Scenario 1: Dirt caused a short. Why do i suspect this? because there was some evidence to suggest the sleeving on the ground/earth lead that runs across the board (which was quite filthy i must admit) had conducted some of the blast (blackened and heat evidence - faintly visible in picture)

2: Something has gone wrong with a one of the IGBT's (CHINESE ONES.....IN A MACHINE THATS COSTS ~£1000!! - Stands back in utter disappointment!) or perhaps one of the gate driver circuits. I have, however, checked as best I can and found all but 2 of the transistors read good, 2 are showing open on all connections but I wont know if thats "true" until i test them off the board.

I guess what I would really like to know is, what values would these 3 resistors likely be, and does anybody have the foggiest idea how I can go about saving this from the bin!

Thanks to any help in advance!

 

[Erie Fred]

A bunch of electronic components are made in China.
Not much you can do about it.

 

[Tommy Tig]

This is true. But they did explode on this occasion...

 

[Barking Mat]

Can you get a new board ? You say parweld don't reply to you, have you phoned them, if so name and shame them.

 

[eddie49]

Hello and welcome to the Forum.

I think that one or more of the IGBTs either failed short-circuit or was incorrectly driven into full conduction and then blew open when the second "leg" of the inverter was switched on. When a device is shorted internally it can conduct the 325v DC supply back into the gate circuits, which is probably where the three resistors were.

This recently-revived Parweld thread describes a similar failure:

Parweld XTT182DV no power

I have an XTT182DV inverter set, which is currently not showing any signs of life. A preliminary investigation is showing that I am getting AC voltage to the first pair of rectifiers and when the DC side of the rectifiers is disconnected from the main PCB there is 240V DC. When the DC side of...

www.mig-welding.co.uk

 

[Kram]

CHINESE ONES.....IN A MACHINE THATS COSTS ~£1000!!

Most components are made in China, they are pretty damned good at it, if the importer is willing to pay for a quality product. I suspect Parweld are not.

I phoned up Parweld customer service to ask if a supplier was sending out fake tig gloves. No, they are genuine, authorised distributer. Oh, please send me some without holes. After repeated calls I had 3 pairs of holey gloves, only suitable for grinder use.
I phoned the again to ask, how can they expect a customer to then go and buy anything more from them? Tig kit is not cheap.


Anyway, I remember on ebay there was a "£99 fixed price repair any chinese inverter" listing. It was mentioned a few times on here but I dont remember any users reporting back good or bad...? Theres a thread somewhere.

 

[Tommy Tig]

Can you get a new board ? You say parweld don't reply to you, have you phoned them, if so name and shame them.

So far only Email attempts I must admit, I will try to find time over the next few days to sit on the phone for however long it takes. I will only name and shame if they are completely unhelpful, however, after what I've heard from others I'm not holding my breath for them to bend over backwards for me...

Hello and welcome to the Forum.

I think that one or more of the IGBTs either failed short-circuit or was incorrectly driven into full conduction and then blew open when the second "leg" of the inverter was switched on. When a device is shorted internally it can conduct the 325v DC supply back into the gate circuits, which is probably where the three resistors were.

This recently-revived Parweld thread describes a similar failure:

Parweld XTT182DV no power

I have an XTT182DV inverter set, which is currently not showing any signs of life. A preliminary investigation is showing that I am getting AC voltage to the first pair of rectifiers and when the DC side of the rectifiers is disconnected from the main PCB there is 240V DC. When the DC side of...

www.mig-welding.co.uk

Yes I have read through that thread, full of useful information. I've managed to order a pack of 10 K75H603 for £12 from aliexpress. These transistors are not as common as other ones I've seen used in similar machines but I'm not gonna argue with the price! I can easily replace all 4 transistors twice over with 2 to spare BUT i still don't know where to start with the resistor values...

Most components are made in China, they are pretty damned good at it, if the importer is willing to pay for a quality product. I suspect Parweld are not.

I phoned up Parweld customer service to ask if a supplier was sending out fake tig gloves. No, they are genuine, authorised distributer. Oh, please send me some without holes. After repeated calls I had 3 pairs of holey gloves, only suitable for grinder use.
I phoned the again to ask, how can they expect a customer to then go and buy anything more from them? Tig kit is not cheap.


Anyway, I remember on ebay there was a "£99 fixed price repair any chinese inverter" listing. It was mentioned a few times on here but I dont remember any users reporting back good or bad...? Theres a thread somewhere.

I actually honestly believe that the cheap medium range machines from china entirely necessarily at all bad. Put it this way, Ive had an "andelli" 250 mig in the past and used it daily for 5+ years.... cost me £300, welded beautifully. And here I am with a dead £1000 machine with only one feature that the chinese cousin didnt have (burn back), which in my opinion is a luxury not a necessity. Don't get me wrong, those bottom of the barrel 'gasless only' machines with chocolate wire feeds and cheapest possible components probably just shouldn't exist, but the quality goes up drastically when you spend a bit more on a machine with real features and quality running gear like well built wire-feed system etc...

Your experience, unfortunately, is far less rare than we would like, now that everything is made as cheaply as possible it can be hard to find anything of value even when you're paying alot for it. I've had Tig gloves from the cheapo places and lucked out big time, still got pairs going years later!

Cheers all, any info on figuring out these resistors would be helpful, the mystery continues!

 

[Kram]

Interesting they used 3 resistors in parallel, makes me suspect its designed as a fuse with 0 ohm, it seems to be quite common to use as fuses.

If its not a fuse, then using three resistors to spread the heat, rather than using a single bigger component?

With three, the value must be 3x greater than the intended actual value (if you use a single item).

Your experience, unfortunately, is far less rare than we would like, now that everything is made as cheaply as possible it can be hard to find anything of value even when you're paying alot for it. I've had Tig gloves from the cheapo places and lucked out big time, still got pairs going years later!

Its silly. If they would have thrown a decent pair in the post, then I could say, may have been a bad batch, reasonable brand etc.. but the fact their customer service, twice, sent unacceptable items, tells me they really do not care for reputation., or they dont have anything better!

 

[Tommy Tig]

Interesting they used 3 resistors in parallel, makes me suspect its designed as a fuse with 0 ohm, it seems to be quite common to use as fuses.

If its not a fuse, then using three resistors to spread the heat, rather than using a single bigger component?

With three, the value must be 3x greater than the intended actual value (if you use a single item).


Its silly. If they would have thrown a decent pair in the post, then I could say, may have been a bad batch, reasonable brand etc.. but the fact their customer service, twice, sent unacceptable items, tells me they really do not care for reputation., or they dont have anything better!

I thought about it being a way to spread thermal load but the fuse theory is interesting! If I could somehow find out the values of these resistors id have a good shot at repairing this machine.

UPDATE: I have now removed all 4 Transistors (K75H603) from the board and have tested them again. The results are the same, 2 are dead, 2 appear functional but one of the two "good" ones has a ruptured case so I will replace all 4 with the new ones when they arrive.

Now I just need to find some way of either knowing the resistor values or testing different values, Im at a bit of a loss!

Thanks

 

[Kram]

Can you get good photos, both sides of the board. It shouldnt be too difficult to reverse engineer. What components are either side of these three resistors?

 

[Tommy Tig]

Can you get good photos, both sides of the board. It shouldnt be too difficult to reverse engineer. What components are either side of these three resistors?

Okay so first few images are of the whole board as best i could capture (both before and after removal of Transistors and Heatsinks:


Next we have a shot of the 7 MOSFETS which I have left alone as they all seem to check out....


And Finally, this is the cap thats connected to either side of the 3 resistors (some sort of soft start/smoothing circuit??)

 

[Kram]

I know its not easy, but good photos are needed (additional light around the big parts in shadow) or you will have to look carefully and detail each component and connection in this area of the pcb.

In that case, simply between two legs of a capacitor, I would suggest they are drain resistors, to remove any charge when its powered off. They will have been high value items to minimise power consumed.
I am thinking that dust has accumalted and allowed them to arc over and explode, which I cant see being related to the IGBT's failing, may have been a seperate incident.

The larger pcb trace of the two is marked as high voltage/live. Where does the smaller trace go to, is it grond or to another component?

@eddie49 any ideas?

 

[Tommy Tig]

I know its not easy, but good photos are needed (additional light around the big parts in shadow) or you will have to look carefully and detail each component and connection in this area of the pcb.

In that case, simply between two legs of a capacitor, I would suggest they are drain resistors, to remove any charge when its powered off. They will have been high value items to minimise power consumed.
I am thinking that dust has accumalted and allowed them to arc over and explode, which I cant see being related to the IGBT's failing, may have been a seperate incident.

The larger pcb trace of the two is marked as high voltage/live. Where does the smaller trace go to, is it grond or to another component?

@eddie49 any ideas?

Thanks that makes sense actually, although it does make me wonder about the IGBT's failing..

Its really hard to tell where that smaller trace goes but I will have a closer look in the morning and try to get some coherent pictures...

 

[Tommy Tig]

Also worth noting this board was very filthy and covered in dust (largely grinding dust) so while it may look clean now, it certainly wasn't, and i think that's one reason i suspected a short of some description....

 

[Tommy Tig]

If anybody knows of anyone who owns an XTM201i that wouldnt mind whipping the cover off and taking a clear photo of those 3 resistors, id be endlessly grateful!

 

[Tommy Tig]

Some clearer pictures of the capacitor that the 3 blown resistors sit between. The tracks are all connected to the main caps which suggests they are indeed drain resistors to discharge the main caps when machine is turned off (I think!)

The cable sheathed in white behind the white cap is the DC+ from the diode bridge.

I have emailed my local parweld stockist asking very nicely if they would take the panel off one of their machines and send me a photo....but unless someone there is feeling very nice I doubt they will be bothered to do it.

Again if anyone knows anyone who has one of these machines it would save my bacon here just to have a simple clear photo of the board.

Cheers all.

 

[Tommy Tig]

UPDATE:

Got an email response from Parweld, they didn't supply me with much useful information but did say they could sell me an entire new board (i've asked how much and we will see what they say, I suspect an arm and a leg)

Also got a more helpful email from my local supplier who amazingly found out the exact resistors i need. '202' which would mean 2K resisters and going by the physical size they are '2125' smd 1w. I will get hold of some and replace them along with the transistors and update once i've turned it on and either blown it up again or fixed it.

 

[Kram]

Far too low value for drain resistors so they must have another purpose, to rule it out, measure if there is continuity between each leg of the white capacitor and the DC bus - each leg of the big electrolytics. If they were it would be 240w going through the resistors.

The white cap is likely a filter as its an AC type but for what, as resistor values still far too low for 240v - that would be dropping 86w

Can you see where the traces go from the white capacitor? Look like they go under the electrolytics.

 

[northwest]

In defence of Parweld, who I think have let you down here, when I had an issue with a Parweld machine they were very helpful and patient with me (as it was my ignorance that created the problem). Kudos to your local chap.

 

[Tommy Tig]

Far too low value for drain resistors so they must have another purpose, to rule it out, measure if there is continuity between each leg of the white capacitor and the DC bus - each leg of the big electrolytics. If they were it would be 240w going through the resistors.

The white cap is likely a filter as its an AC type but for what, as resistor values still far too low for 240v - that would be dropping 86w

Can you see where the traces go from the white capacitor? Look like they go under the electrolytics.

Right, finally got to spend a bit of time in the shed. Either side of the Resistors (and legs of ac cap) are each pair of Transistors and each end of the Primary Winding on the output tranny. So this definitely related to the 2 blown Transistors. At this time, after testing a million little components and all the Zenors I could find, Im still fairly confident this is a case of a short (likely between one of the big caps and the resistors) So as it stands I will go ahead with the plan to replace all Transistors and the 3 Resistors, put it all back together and see what happens.

Could be a fairly length wait for the transistors to turn up from China but I did manage to find a pack of 10 resistors for like £3 from a UK seller.

In defence of Parweld, who I think have let you down here, when I had an issue with a Parweld machine they were very helpful and patient with me (as it was my ignorance that created the problem). Kudos to your local chap.

I'm not even cross with Parweld over this, the more I consider what their product (or any welding plant) goes through and the environment it operated in, its impressive that this isnt more common across the board. They DO appear to laquer the boards quite well, although obviously a certain amount of grinding dust and vibration will eventually always get through in the end...

I was a little put out by how long it took to get a reply from them but at the end of the day they DID get back to me, albeit with the suggestion of buying an entire new board which is likely going to cost more than its worth for a machine like this, we will see.

As for my theory on this being caused by grinding dust, I guess its a lesson to all of us that it wouldnt hurt to have the cover off once a year and blow out the cobwebs... Although I must say I've seen machines FAR more full of dust and crap that were still running just fine...

Cheers chaps, will update as soon as I have anything further to report!