Blown my welder up.

[WorkshopChris]

There appear to be three large bolted connections, for zero, plus, and minus outputs.

I think this is mains in via a MOV to the AC side of the diods, the other connection can be seen in the above shot.

 

[Retired]

Hi,

Thanks Dave H; quite a while ago I sat at the computer with Bron and we ran a colour blind test; I think there were 24 test cards; Bron got all 24 correct whilst I managed 3. I'm absolutely hopeless with colour; even if I had any type of special lens it wouldn't help me because I simply couldn't recognize the colour anyway; I can see colour OK but putting a name to them I find impossible at times; purple; blue; lilac grey; just different shades to me; when does red turn to orange; brown and green I can tell the difference by shade but not name them; traffic lights go amber/red each time I approach so it's the top light to stop at; yes I can see they are a lot different but again I can't name the colours. I'm told the only thing that cannot be described by text alone is colour? The little coil winders are very useful indeed and cost me nothing to make using bits out of the scrap bin.

Good questions eddie49 and thanks for asking. Yes two full wave bridge rectifiers; one positive the other negative? I haven't given this any thought just assuming both to give DC output which is then smoothed by the Electrolytics? It's interesting to learn your thoughts though; I've spent little time looking at the inside of the welder or circuit board/s; I've done quite a bit of electrical work over the years but I'm actually a mechanical engineer so my main object is to get the machine up and running again without going in depth as to what component does what; if the new caps also expire then obviously I'll need to dig deeper; I'll install the new caps then put everything back together before powering up. I'll certainly not be blinded by exploding Electrolytics which is a very real danger.

Thanks WorkshopChris; if indeed you've missed something then I've missed a great deal more.

All I really need to do is to unsolder the blown caps and solder the new caps in position taking care to get their polarity correct but I've done plenty so it won't be a problem.

It's highly likely others too will have blown up their welder but then taken the welder for repair; I've posted this just for interest because if all goes well and it's just the capacitor fault it will be a very cheap repair but I'm not encouraging anyone to simply jump in because a novice could end up being electrocuted or causing the problem to expand.

I'm more busy removing laurel stumps at the moment whilst the weather permits; I'm not in a panic for the welder; I'll eventually get the welder working again and resume practicing welding; if it can go wrong in fact if it can't go wrong it surely will for me.

Kind regards, Colin.

 

[Hopefuldave]

I keep looking at this I can't work out what I am missing, the board looks to be multi-layer but if all 6 caps are common to the centre section in the underside photo then should all the caps be orientated to either positive or negative to share this common point?
Might just be me missing something due to meds.

If the caps are in series to share the applied voltage they have to be nose to tail (electrolytics being polarised, they tend to burst if wired backwards), so one bank with have their positives to the common "centre point", the other their negatives - that point will be at (close enough to) half the applied voltage and probably doesn't have a load on it.
On inverter welders it may be the centre point of a high-frequency transformer used to convert the rectified mains power to welding current, I'd hve to see a circuit diagram to e sure

That's the place to put the neutral on a 415v in VFD if you're trying to make it run on 240, but as mentioned it does shorten the life of the cap's as they're only charged alternately (one on positive peaks, the other on negative) and ripple current is way higher than they're expecting...

Dave H. (the other one)

 

[minimutly]

I don't think it is just you.... the only way that I can reverse-engineer a PCB circuit diagram is to hold the board in my hand and turn it over and back again, whilst sketching each step.
I would still recommend tracing the AC mains input wires, I'm sure they end up on this board, and then measuring the DC ( hopefully DC ) that is going to get applied to those new capacitors before soldering them in. I would not want to have them go bang again in front of my nose.... but then again, 200 or 300 v DC needs to be respected too.

[/QUOTE]
I do exactly that but hold it up to the light to see through it.
Once tried to fix a 240v ups inverter - the caps went bang, flames and all. Changed them, same thing again....killer ups

 

[sako243]

@Hopefuldave is correct the centre pads appear to be the common rail between the stacked voltage-sharing caps.

Incidentally if you bump them up to 220uF it's a much more common value and the components should be cheaper. Given it's almost certainly just the smooth for the rectification stage of the inverter it's unlikely to make much of a difference to it's operation.
RS have lots of caps rated >450V which you could replace them with.

 

[Hood]

@Hopefuldave ....
Incidentally if you bump them up to 220uF it's a much more common value and the components should be cheaper. ....

The originals are 1200uF, are you meaning bump to 2200uF or was the 220uF what you meant?

 

[Hopefuldave]

@Hopefuldave is correct the centre pads appear to be the common rail between the stacked voltage-sharing caps.

Incidentally if you bump them up to 220uF it's a much more common value and the components should be cheaper. Given it's almost certainly just the smooth for the rectification stage of the inverter it's unlikely to make much of a difference to it's operation.
RS have lots of caps rated >450V which you could replace them with.

I'd be wary of doing that, unless it has a soft-start circuit (e.g. a humongous wire-wound resistor that's bypassed by a relay once up to voltage), the inrush current would increase with the capacitor value and could get to rectifier failure...
I know my (somewhat heftier) TIG and VFDs have soft-start, his may not?

Dave H. (the other one)

 

[sako243]

The originals are 1200uF, are you meaning bump to 2200uF or was the 220uF what you meant?

My mistake - I looked at the photo and missed the 1... Did think 200uF was a bit of an odd size.

 

[Retired]

Hi,

Thanks for all your interest and replies.

Yes the originals are 1200 uF at 200V; I wouldn't mess around with changing the capacitance at all; I've upgraded the voltage previously on other projects but given the complexity of the welder circuits I'll stick to original specification; RS caps should be decent quality and as only the two caps have blown which are connected to the rectifiers my fingers are well and truly crossed replacing the caps will sort the problem out? I'll have a look later at the RS data sheet for the new caps and try to compare with the originals.

As usual I'm very busy; I was in the garden early removing laurel stumps and have just knocked off for a welcome mug of tea; time I got off my backside again and stop being lazy; being retired is full time hard graft.

Kind regards, Colin.

 

[Retired]

Hi,

I've just been looking at the original capacitor spec;

Shelf life 1000h?

My welder had been unused stored under a bench for at least one year being brand new? I'm just curious.

Kind regards, Colin.

 

[R Kraft]

Electrolytic capacitors are made with a formed coating of oxide on the foil plates.
The oxide is formed by the electrical energy applied to them,and is an insulator.
Power is applied to them at the factory as a gradual increase of voltage.
When they are stored unused, the oxide coating starts to decay, and the capacitor will allow power to flow through it instead of storing it ,and they heat and blow up..

A stored capacitor can be used if it has a reforming voltage gradually applied to it.

 

[eddie49]

Shelf life 1000 h ? That is just 42 days.
Frankly, not something I have ever heard of or bothered with, but it does seem to be true:

https://electronics.stackexchange.c...trolytic-capacitors-have-a-limited-shelf-life

In particular, there is a suggestion from Panasonic that out-of-date aluminium electrolytic capacitors need to have the layer of aluminium oxide on the foil "reformed" by applying the rated voltage through a 1K resistor for 30 minutes. If this is not done, then powering up in equipment that has been idle will still re-form the oxide, but without the resistor the current may cause overheating and an explosion.
How soon, after the first power-up following storage, did your welder last before it went bang?

 

[Seadog]

In particular, there is a suggestion from Panasonic that out-of-date aluminium electrolytic capacitors need to have the layer of aluminium oxide on the foil "reformed" by applying the rated voltage through a 1K resistor for 30 minutes.

That's what I understand. But the article I saw suggested initially limiting the current to 20mA, I think it was, and then slowly increasing the current over a period of a few hours.

 

[TigTigger]

If you apply the rated voltage to a capacitor through a resistor it will charge up to the rated voltage and then the current will drop to almost zero, just a small amount of leakage. So there is no way you can increase the current over a period of hours. If you want to pass current through the capacitor you will have to continually charge and discharge it as current only flows when there is a change in applied voltage. Which is exactly what happens when you use it as a smoothing capacitor in a power supply when under load.

 

[Seadog]

True. I'm trying to recall the article which I read several years ago. The capacitor is connected to a current limited supply, the applied voltage being the rated voltage. Current is limited to about 20mA. The reforming effect is achieved by the applied voltage, not the current. The capacitor is cycled between charged and discharged.

I'll have a dig around to see if I can find the information again.

Edit: Found it. The text -

"
[1] Do the doubler first and check the volts are fairly near, although they won't be precisely the same.

[2] Make the Zener circuit, then set VR1 for 20V on the 16V range. [ you'll see why in a bit ].
Measure from Q1 emitter to 0V. Check the other voltages, make sure you ain't got a cranky Zener.

[3] Make the rest. Put a short circuit instead of the cap. and set VR2 to give 10mA on the meter.

[4] With a 10R resistor instead of cap. tweak VR1 again to read 16V from the -ve to +ve test terminals on the 16V range. Then check the other ranges.

The -ve test terminal actually sits at some +4V because of the Q2 limiter action, hence the 4 extra volts in [2].

Poke a GOOD biggish cap. in and set up. [ Not in that order ]

The meter should whip up to 10mA and stay there. When the cap's charged up it will swiftly drop to zero.

As this contraption is sort-of linear you can make a fair stab at finding out what a cap. really is.

CV = It ...

Capacitance x Volts = Amps. x Seconds

Or: C = ( I x t ) / V

Therefore:

Suppose we have 25V, Amps. = 0.01, time to when the needle starts to drop, 6 seconds.

We have (6 x 0.01) / 25 = about 2400uF.

Have done this with caps. marked 56,000 uF, 25V and then with my Peak capacitance meter and it's pretty accurate.

EDIT S1 is a 12P 1W rotary switch, stopped off @ posn. 8 ..

S2 on mine is a ON [ON] 2P 2W just flick it to discharge the cap. DVM should drop to 0V, then cycle cap. again if it's not been used for a long time.
The 'slow' drop from 10ma to 0 gets sharper as cap. improves."

 

[Munkul]

Hi Colin,

Such a shame you had a nearly new unused machine, and blew it up first try! Unfortunately not as uncommon as it ought to be. Seems to be a thing amongst some of these imported machines. Having said that, they all break sometimes.

seems like you're a handy sort with electronics, fingers crossed you can repair it without too much hassle/cost.

Something that may or may not make a difference - I always would test a new (to me) welder on DC first, not AC - given than the cyclic nature of AC is harder on an inverter than DC, and gives all the components a "warm up". It's maybe just an illogical feeling without any scientific grounding, but I always started my ancient ESAB up on DC first, since it was a bit fussy on AC and wouldn't always arc up first time - so tested on DC first, especially in the winter when moisture in the workshop was everywhere and that carried over to the other TIG machines I've used, and even the brand new Lorch I have now.

 

[Retired]

Hi,

Thanks everyone for your most interesting posts and a special thanks to Seadog for adding so much useful information.

During my vintage radio restoration days radio forum members I believe used a "Variac" to give the kiss of life to electrolytics that appeared to be dead; I used to have two Variacs but sold them on when I packed in restoring. I always found it less hassle and far simpler to snip out the old and in with the new.

1000h shelf life is actually very little; a supplier could possibly have these caps on the shelf in store exceeding this? I think I just got unlucky which is normal for me and why I'm not at all upset; I'm so used to things going wrong all the time its a way of life for me; it's actually more interesting than upsetting because I learn so much; many would despair having their new welder blow up but I just turn it into yet another project rather than see it as a problem.

I'm unaware if circuit diagrams are readily available for welders; I have thousands of vintage radio circuit diagrams; I've not had chance to browse the web in fact I'm fully retired and still working flat out. It's my birthday today aged a young 72 and first thing this morning I was up the mountain digging out more laurel stumps which I've been doing over the last three mornings; I've finally got them all out and dumped at the tip this afternoon hence my late reply. Bron and I live on a steep valley side making standing upright in the gardens difficult; I've been carrying up trolley and bottle jack plus other items like a very heavy 5' long crow bar; our rear garden we call our mini park; there is always work to be done so workshop time comes a poor second best.

Three very hard mornings grafting; these remaining stumps I've removed today.

A few of the stumps/roots.

When I do finally get into the workshop I find I feel unsettled thinking I should be elsewhere; the work seldom lets up here.

Thanks Munkul yes it's a shame indeed and hopefully I can resolve the problem shortly; what is more annoying than the welder blowing up is the wasted workshop time lost; I was so keen to have a go at TIG welding then this had to happen; it's not the end of the world but does rather shatter my concentration; I'd spent so much time getting the welder and all the associated kit together including collecting the new welding helmet then getting the petrol tank ready including metal spinning the ends which takes quite a bit of time and struggling to get into the workshop only to be stopped by exploding capacitors; I'm sure if I tried to shoot myself I'd miss. I note your comments regarding starting the welder in DC mode which makes sense; it would help if I could simply turn on the welder allowing the caps to fully charge leaving it running for a while before welding but I'm unsure of the circuits and would the caps actually charge with the welder only switched on with the fan running; I see there is a relay on the board and wonder if this only activates once welding gets underway hence the caps only charge up then? I'm self taught on electronics in fact electrics only learning on a need to know basis; I can tell the difference though between a cap and a resistor; the main concern with this kind of work is personal safety and safety of others; I'm used to having a radio chassis on the bench under power knowing full well the dangers and many of these old chassis were actually live to mains. I used to have an isolation transformer; 100W lamp limiter and also Variac for powering up these chassis many having a multitude of faults due to the chassis unused for many years.

Because this welder is relatively new the fault most likely is isolated to just this one board at least I hope so because I want to learn welding not increase my electronics knowledge. I didn't know Lorch made welders; I fully restored a very rare Lorch Schmidt engineering lathe a few years ago it being German but perhaps the name Lorch is just a coincidence. I did think I'd be in with a good chance because these Parwelds aren't exactly what I'd call cheap but then I've never ever trusted anything with a circuit board and never will; steam power didn't rely on capacitors and I'm a fossil knowing about such things; I was about 5 years old before we even had electricity in our home but we did have gas and a coal fire.

Kind regards, Colin.

Kind regards, Colin.

 

[Seadog]

I still have a couple of Variacs, and I know that is the way that we laearned to reform electrolytics, but current limiting and applied working voltage is the better way to do it.
Consider a very poor condition capacitor. Even at a few volts it could have the potential to draw lots of current and let the magic smoke out. I had a big old capacitor explode once. On a 350v rail, open chassis...It left a nice dent in the ceiling and someone with brown trousers

 

[R Kraft]

Chinese inverters seem to all use approximately the same power board.
Inrush is usually limited by a large resistor and the large capacitors charge to the operating voltage.
The +/- 24v of the driver board runs the welder in standby mode, until the thing is switched on, and then a relay bypasses the inrush resistor.
Not usually real complicated circuits until you get to the amplifier and feedback circuits for the voltage and current out.

 

[Hood]

Hope you had an enjoyable birthday Colin, hopefully you will get a nice present tomorrow when you fit the caps and the welder springs back to life