Wire Feed Upgrade

[johnrev]

Congratulations on all you've done so far. I have done similar mods myself to several machines and the wire speed mod really does make a big difference. Only thing I had to do was add a couple of resistors to each end of the PWM board potentiometer to get the useful wire speed range more evenly spread over the pot travel as otherwise the useful range came in about 25% of it's travel as you have found. I'm sure Eddie could advise if you wanted to give this a try.

The dimmer I used on my experimental setup was a 10000w unit I bought from Banggood.
Unfortunately I haven't had the spare time to make much use of it but I can say that it does work to control the transformer output voltage as it should. At low setting it sounded very buzzy but slightly off minimum it was fine and control up to max seemed smooth. As Eddie said these are not intended for inductive loads which is why I went for the one with the max power capability I could find. I also noticed it seems to have a rectifier on the board and have seen a youtube vid by 'the post apocolyptic inventor' where he explains the advantage of this. I can't pretend I understood all he said but it sounded like it might help with inductive loads.
The unit I got is:-https://www.banggood.com/AC-220V-10000W-Digital-Control-SCR-Electronic-Voltage-Regulator-Speed-Control-Dimmer-Thermostat-p-1373125.html?rmmds=myorder&cur_warehouse=CN

I have tried adding caps to both a stick welder and mig welder and have been disappointed with the outcome on both machines. All it seemed to do was make a lot more spatter and a horrible crack when striking the arc. Given the time I will try again but the choke is a far more effective method of smoothing an arc as far as I can see so will not worry too much about caps for now.

It's been good to read your progress reports and I'm glad you've had success. All those who say get rid of the machine and get a better one are missing out on a great feeling of achievement when all the head scratching and fettling produces worhwhile results.

All the best

John

 

[Seadog]

Perhaps you needed an inductor in circuit too.

 

[johnrev]

Hi Seadog. On my SIP topmig there was an inductor and I put the caps between the rectier and inductor but results were poor. I tried lots of different numbers of 4700uf caps in parallel but the more I used the worse it got.
The stick welder had two homemade inductors which I tried seperately and together but the thing was also worse with the caps than without. I also found the transformers and chokes started getting hot more quickly when the caps were connected.
I've now got some chokes from Andy (a111r) and I will try again when I've nothing else needs doing. Got a few other projects on the go at the moment.

John

 

[Seadog]

I started to go down the upgrade route about two years ago, on my SIP 190A stick welder. I then lost interest (No space in the workshop so even small jobs turned into big ones).Once I get the workshop sorted out I will recommence the project, so interested to hear about your results.

 

[Nige2108]

Congratulations on all you've done so far. I have done similar mods myself to several machines and the wire speed mod really does make a big difference. Only thing I had to do was add a couple of resistors to each end of the PWM board potentiometer to get the useful wire speed range more evenly spread over the pot travel as otherwise the useful range came in about 25% of it's travel as you have found. I'm sure Eddie could advise if you wanted to give this a try.

The dimmer I used on my experimental setup was a 10000w unit I bought from Banggood.
Unfortunately I haven't had the spare time to make much use of it but I can say that it does work to control the transformer output voltage as it should. At low setting it sounded very buzzy but slightly off minimum it was fine and control up to max seemed smooth. As Eddie said these are not intended for inductive loads which is why I went for the one with the max power capability I could find. I also noticed it seems to have a rectifier on the board and have seen a youtube vid by 'the post apocolyptic inventor' where he explains the advantage of this. I can't pretend I understood all he said but it sounded like it might help with inductive loads.
The unit I got is:-https://www.banggood.com/AC-220V-10000W-Digital-Control-SCR-Electronic-Voltage-Regulator-Speed-Control-Dimmer-Thermostat-p-1373125.html?rmmds=myorder&cur_warehouse=CN

I have tried adding caps to both a stick welder and mig welder and have been disappointed with the outcome on both machines. All it seemed to do was make a lot more spatter and a horrible crack when striking the arc. Given the time I will try again but the choke is a far more effective method of smoothing an arc as far as I can see so will not worry too much about caps for now.

It's been good to read your progress reports and I'm glad you've had success. All those who say get rid of the machine and get a better one are missing out on a great feeling of achievement when all the head scratching and fettling produces worhwhile results.

All the best

John

Thanks John for your feedback. The dimmer you have selected sounds interesting. Just browsing Banggood's site I came across this one which looks similar with a digital control rather than the pot control https://uk.banggood.com/AC-220V-100...XCIUI89BZ33V_-YW422R2-irtXdHvcKBoCgLQQAvD_BwE

I'll wait to hear from Eddie with his thoughts but I'm wondering if there is any merit in (a) using the type of dimmer you suggest and maybe (b) switching the second capacitor. I have a spare 240v double pole switch which would probably do that. That combination might improve control at the bottom end of the power range.

I'm interested in your thoughts on the choke. I'm not particularly knowlegeable on this but am I correct in thinking this performs a similar function to an inductor coil? My MIG already has one of these fitted. This being the case would adding an additional choke improve things further?

Thanks once again for your encouragement.

 

[Seadog]

A choke is an inductor.

 

[johnrev]

Nige you are best being guided by Eddie or one of the other knowledgable guys on here but I can tell you what my limited experience suggests.

Adding a bank of caps to my mig welder was not a great success with the arc seeming less 'smooth' than without. May well be that I didn't give it enough time to get settings correct but I was expecting a gentle buzz from the machine and got a loud crackle instead! The welds seemed less smooth too but I found that the transformer and choke started getting very hot quite quickly so called it a day. I didn't want to damage a good machine by experimenting further. Using a capacitor bank obviously increases the effective voltage output of the welder which may well need taming to get a good weld.

The choke/inductor apparently resists the fast change in current you get from a rectified transformer output. This output is actually a voltage which changes from zero to maximum at 100cycles/sec and each time it reaches zero the arc effectively collapses and has to restart again as the voltage rises. The choke resists this fast rise and fall so the arc has less tendancy to collapse and restart producing a smoother weld. The foregoing is my 'take' on what happens?
On my Oxford mig there are three inductor settings which tap into the choke so producing different degrees of inductance. The difference in the sound and feel of the arc is very noticeable. At the highest inductance the arc feels and sounds silky smooth and the weld looks really nice. Conversely it is also the least powerful weld setting with the least penetration. Understandable I guess since the choke is resisting current flow.

Capacitors do a similar job in that they actually store a charge as the voltage rises and release it as the voltage falls so 'filling in' the periods when the arc would normally collapse and so produce a smoother arc.

As I see it a downside to the capacitor action is that they are components which become pretty much a dead short across the transformer/rectifier as they charge up from zero. My worry was that this was causing the heating of the transformer etc and must have been putting quite a strain on the rectifier too. I don't know if the extra drain on the transformer secondary would increase the stress on the primary side and therefore the 'dimmer' too? I also don't know why my results were so poor using caps. More time and trial settings is needed perhaps.

In fact there's an awful lot I don't know! For instance I got a few chokes from a member on here and tried adding one on a small 100amp mig which was using gasless wire. It effectively doubled the inductance on the machine and the result was...............hopeless! Weld was still harsh and spattery and only effect was to decrease the apparent power of the welder.

Inductance value seems to be quite difficult to guess at when you are winding a coil on a core to make one. Add to that the fact that you really don't know what the ideal value should be for any given setup and we're talking total guesswork here. Nearest I can get is by looking at the ones a111r gave me. They seem to be about microwave oven transformer sized or a little larger and have around 60 turns of 4mm dia.wire on them. These are for 100-150amp machines.

Make of all this what you will and good luck with your experiments. I will keep playing as time allows and will post up any findings which seem interesting.

John

 

[eddie49]

Thank you for the update. You have done a very neat job of installing and wiring the capacitors and the bleed resistor, and splicing onto the aluminium wire crimp.
The results were promising, apart from the dimmer giving up!
However, I am not certain that it is gone. The power control device is a Triac, which looks like a pair of diodes back-to-back so it conducts both parts of the AC cycle. It will only do this when triggered, so a multimeter check will show no conduction. I would suggest putting the dimmer in series with a 100W mains filament bulb wired to a 13A plug, and see if it will still act as a dimmer. If no-go, open it up, it may contain a fuse. I have a "4000 Watt" dimmer, and there is a 20 Amp fuse inside.
At low settings only a small slice of each AC half-cycle will get through, looking like a saw-tooth rather than a sinewave, so as John said the transformer may well buzz.

If we can find a robust dimmer then it could be wired in the MAX feed to the transformer, the switches set to "MAX" and "2" , or bypassed altogether, and the dimmer would control power from zero to maximum. As I mentioned, the Lidl Parkside PFDS 120 A2 is a transformer welder controlled by a Triac, and these two recently-updated threads :

Flux core welder modification plus a big hello

So this is a little bit of two post's into one, I'm new to the forum so a big hello to every one, quite new to welding done some in the past but is been a few years. I'm a sparky by trade, was in the game for over 10 years but been in the building trade all my life, but after the loss of my...

www.mig-welding.co.uk

Mig master 130

Hi. Very new here. I have a mig master 130 (auto jack). A couple of months ago I had issues with the circuit board. Long story short. Got a replacement, but now can't recall where all the wires go. One of the members (jmh474) posted some pictures which solved some - thanks!. Can I please ask...

www.mig-welding.co.uk

describe the "AutoJack Migmaster 130" which has a rather fancy-looking dimmer board inside.
There are professional machines, e.g. Kemppi and BOC, which are not inverters but which use thyristors or Triacs to provide fully variable power control.

The alternative dimmer in a box with a power socket that you found is probably just the same device in a fancy enclosure. The "10000W" one suggested by John is better, because the actual Triac seems to be twice the width of a normal TO-220 device - it has a wide metal tab with two mounting screws. If your dimmer has blown up, and it is the Triac that has failed, then one solution would be to replace that component with a more robust one, such as a stud-mounted Triac ( if they can be found ), or a pair of back-to-back stud-mounted SCRs. There are also Triac "modules", such as the 40 Amp TG40E60, or this one:

SCR module TRIAC module MTC110A (Sansha appearance) | eBay

Find many great new & used options and get the best deals for SCR module TRIAC module MTC110A (Sansha appearance) at the best online prices at eBay! Free delivery for many products!

www.ebay.co.uk

The idea would be to use the original trigger circuit & control pot from the dimmer, but fit a better Triac. Or, design a proper one from scratch....

It is true that full-wave rectified AC does fall to zero at the end of every half-cycle, 100 times per second, and capacitors are there to stop this, effectively "supporting" the voltage at near the peak value, with just a slight ripple on the top. The first power-on of an empty capacitor bank will stress the rectifier, and perhaps the dimmer too, but once the caps are full they are open-circuit. Rapid on/off triggering, to create a series of overlapping weld spots ( the "thin metal technique" ) could be a problem. I think the ultimate dimmer circuit needs to have a "soft-start", which would gradually ramp-up the width of the trigger pulses for the Triac over the first few AC cycles.

The two capacitors now total 44,000uF, and I think a rule-of-thumb is 1000uF per Amp, so that would be good for 44 Amps, or more. You could test with just one capacitor, but I would not expect any improvement. My 200 Amp Eland MIG has 6 x 10,000uF capacitors.

As for an inductor ( choke ), the third picture on your first post shows that your Kestrel MIG already has one, and it is quite a good-sized one too.

 

[Nige2108]

Thank you for the update. You have done a very neat job of installing and wiring the capacitors and the bleed resistor, and splicing onto the aluminium wire crimp.
The results were promising, apart from the dimmer giving up!
However, I am not certain that it is gone. The power control device is a Triac, which looks like a pair of diodes back-to-back so it conducts both parts of the AC cycle. It will only do this when triggered, so a multimeter check will show no conduction. I would suggest putting the dimmer in series with a 100W mains filament bulb wired to a 13A plug, and see if it will still act as a dimmer. If no-go, open it up, it may contain a fuse. I have a "4000 Watt" dimmer, and there is a 20 Amp fuse inside.
At low settings only a small slice of each AC half-cycle will get through, looking like a saw-tooth rather than a sinewave, so as John said the transformer may well buzz.

If we can find a robust dimmer then it could be wired in the MAX feed to the transformer, the switches set to "MAX" and "2" , or bypassed altogether, and the dimmer would control power from zero to maximum. As I mentioned, the Lidl Parkside PFDS 120 A2 is a transformer welder controlled by a Triac, and these two recently-updated threads :

Flux core welder modification plus a big hello

So this is a little bit of two post's into one, I'm new to the forum so a big hello to every one, quite new to welding done some in the past but is been a few years. I'm a sparky by trade, was in the game for over 10 years but been in the building trade all my life, but after the loss of my...

www.mig-welding.co.uk

Mig master 130

Hi. Very new here. I have a mig master 130 (auto jack). A couple of months ago I had issues with the circuit board. Long story short. Got a replacement, but now can't recall where all the wires go. One of the members (jmh474) posted some pictures which solved some - thanks!. Can I please ask...

www.mig-welding.co.uk

describe the "AutoJack Migmaster 130" which has a rather fancy-looking dimmer board inside.
There are professional machines, e.g. Kemppi and BOC, which are not inverters but which use thyristors or Triacs to provide fully variable power control.

The alternative dimmer in a box with a power socket that you found is probably just the same device in a fancy enclosure. The "10000W" one suggested by John is better, because the actual Triac seems to be twice the width of a normal TO-220 device - it has a wide metal tab with two mounting screws. If your dimmer has blown up, and it is the Triac that has failed, then one solution would be to replace that component with a more robust one, such as a stud-mounted Triac ( if they can be found ), or a pair of back-to-back stud-mounted SCRs. There are also Triac "modules", such as the 40 Amp TG40E60, or this one:

SCR module TRIAC module MTC110A (Sansha appearance) | eBay

Find many great new & used options and get the best deals for SCR module TRIAC module MTC110A (Sansha appearance) at the best online prices at eBay! Free delivery for many products!

www.ebay.co.uk

The idea would be to use the original trigger circuit & control pot from the dimmer, but fit a better Triac. Or, design a proper one from scratch....

It is true that full-wave rectified AC does fall to zero at the end of every half-cycle, 100 times per second, and capacitors are there to stop this, effectively "supporting" the voltage at near the peak value, with just a slight ripple on the top. The first power-on of an empty capacitor bank will stress the rectifier, and perhaps the dimmer too, but once the caps are full they are open-circuit. Rapid on/off triggering, to create a series of overlapping weld spots ( the "thin metal technique" ) could be a problem. I think the ultimate dimmer circuit needs to have a "soft-start", which would gradually ramp-up the width of the trigger pulses for the Triac over the first few AC cycles.

The two capacitors now total 44,000uF, and I think a rule-of-thumb is 1000uF per Amp, so that would be good for 44 Amps, or more. You could test with just one capacitor, but I would not expect any improvement. My 200 Amp Eland MIG has 6 x 10,000uF capacitors.

As for an inductor ( choke ), the third picture on your first post shows that your Kestrel MIG already has one, and it is quite a good-sized one too.

Thanks Eddie, as always your insight is invaluable. Your are correct, the dimmer is working when connected to a 100w light bulb. Attached is a picture of the test and the dismantled unit showing the innards for your info. Its a little puzzling why it stopped working on the MIG because it did work initially. I'm trying to remember if it worked after I connected in the caps. I thought it did work after they were connected but can't be certain. Not sure where we should go from this but look forward to your comments

 

[eddie49]

Since the dimmer still works with a light bulb, it is basically OK, and I think it will work again with your welder. Probably the buzzing sound put you off, and I think it should be avoided by adjusting the control to a "High" setting. Assuming the noise came from the transformer, it could be due to the Triac passing a DC current, which would not be good!

I have been reading up about "AC control ( power control, phase control ) for inductive loads", and have learned a lot.
For example:

https://www.st.com/resource/en/application_note/cd00003856-triac-analog-control-circuits-for-inductive-loads-stmicroelectronics.pdf

The basic problem is that an inductor impedes the flow of current, due to the rising magnetic flux inducing a current in the opposite direction ( Lenz's law ). The current lags the voltage and they get out-of-step, which doesn't work well in AC circuits, especially if you are relying on an electronic device ( namely, a Triac ) which is supposed to turn off ( stop conduction ) when the current through it drops to zero. In AC, it does this at the end of every half-cycle, 100 times per second. With an inductive load, when the current drops to zero, the voltage will not be zero, and so when the Triac "opens" ( stops conduction ) this voltage will suddenly be applied across it. The rate of voltage change ( dV/dt ) may exceed the specs of the device, and blow it. To cushion this pulse, a resistor/capacitor combination known as a "snubber" can be paralleled across the Triac. There are robust Triacs that can manage without this, known as "Snubberless Triacs", and they even have a special device name which I had never heard of before: "Alternistor" .
I used to think that the solution was to specify a rugged Triac. For example, the one in my "4000 Watt" Dimmer ( price £4.55 !! ) is a BTA41-600B, rated at 40 Amps and 600 Volts. However, I now know that the design of the trigger circuit is important, to ensure that the Triac actually does turn off at the end of every half-cycle, and back on at some variable time after the start of the next one. These simple cheap dimmers which are just one-wire devices, in series from the mains Live to the load ( as opposed to across Live and Neutral ) are not accurate enough to provide full control. Apparently they are adequate if you set them from 3/4 to fully on.

 

[johnrev]

That's a great article Eddie. As usual I only 'get' about half of the explanation for the problem/solution but that welder control circuit doesn't look to be too difficult to make. Is it worth making a veroboard version for trying out do you think?
I've probably got some of the components already but things like 1w resistors and the exact triacs/diacs etc will need to be sourced if I haven't got alternatives to hand.
Been a long while since I had to order a batch of components so I'll have a look round for a source of reasonably priced bits.

John

 

[Nige2108]

Since the dimmer still works with a light bulb, it is basically OK, and I think it will work again with your welder. Probably the buzzing sound put you off, and I think it should be avoided by adjusting the control to a "High" setting. Assuming the noise came from the transformer, it could be due to the Triac passing a DC current, which would not be good!

I have been reading up about "AC control ( power control, phase control ) for inductive loads", and have learned a lot.
For example:

https://www.st.com/resource/en/application_note/cd00003856-triac-analog-control-circuits-for-inductive-loads-stmicroelectronics.pdf

The basic problem is that an inductor impedes the flow of current, due to the rising magnetic flux inducing a current in the opposite direction ( Lenz's law ). The current lags the voltage and they get out-of-step, which doesn't work well in AC circuits, especially if you are relying on an electronic device ( namely, a Triac ) which is supposed to turn off ( stop conduction ) when the current through it drops to zero. In AC, it does this at the end of every half-cycle, 100 times per second. With an inductive load, when the current drops to zero, the voltage will not be zero, and so when the Triac "opens" ( stops conduction ) this voltage will suddenly be applied across it. The rate of voltage change ( dV/dt ) may exceed the specs of the device, and blow it. To cushion this pulse, a resistor/capacitor combination known as a "snubber" can be paralleled across the Triac. There are robust Triacs that can manage without this, known as "Snubberless Triacs", and they even have a special device name which I had never heard of before: "Alternistor" .
I used to think that the solution was to specify a rugged Triac. For example, the one in my "4000 Watt" Dimmer ( price £4.55 !! ) is a BTA41-600B, rated at 40 Amps and 600 Volts. However, I now know that the design of the trigger circuit is important, to ensure that the Triac actually does turn off at the end of every half-cycle, and back on at some variable time after the start of the next one. These simple cheap dimmers which are just one-wire devices, in series from the mains Live to the load ( as opposed to across Live and Neutral ) are not accurate enough to provide full control. Apparently they are adequate if you set them from 3/4 to fully on.

Thanks @eddie49 I could try the dimmer again or would it be better to go with this one which is a 10000W unit and has a live and neutral input and output https://uk.banggood.com/AC-220V-100...XCIUI89BZ33V_-YW422R2-irtXdHvcKBoCgLQQAvD_BwE I presume the neutral will have to be connected into the connector on the end of the mains lead? This dimmer has a digital power adjustment instead of a potentiometer which they claim is more robust?