Wire Feed Upgrade

[Nige2108]

Hi guys, I have a relatively old basic mig (Kestrel Multimig 100 - no I've never heard of it since I bought it either!!). it hasn't had much service although I did use it about 25 years ago to weld up the front of my wife's mini metro quite successfully.

Recently I have started to use it again mainly for thin sheet metal work but I felt the welds were quite inconsistent even after a reasonable amount of practice. I then turned my attention to the wire feed, which is a bit puny to say the least and having read a number of articles on the forum about this I decided to take the plunge and proceed with an upgrade. I have purchased a 24v cast ally wire feed unit from ebay and am in the process of mounting the unit into the case. See pics. It is a bit tight for space so I have decided to mount the unit upside down in order to fit it within the case, which I think will work ok.

I note the wire feed kit came with rubber insulating rubber grommets. Is it necessary to insulate the unit electrically against the chassis?

I thought I would complete the replacement first, see what the performance improvement looked like and then depending on that, install an independent power supply which from various threads seems as though it might be a worthwhile upgrade.

Within the manual there is a wiring diagram which I have amended to include my proposal for wiring up a new power supply. My thought following a response I read from @eddie49 in a related thread is to utilise a 24v laptop power supply and possibly a PWM controller in place of the existing one.

My question is would someone be so kind as to check my wiring diagram (attached) before I go full steam ahead with the wire cutters?

Many thanks and look forward to your advice. Nigel

 

[Popcorn]

To make sure Eddie see's it you need to add a @
@eddie49

 

[Nige2108]

Thanks Popcorn. Good to know, I've now added it. Think I may have posted this under the wrong heading. Is there a way to move it? Thanks

 

[Popcorn]

Nothing wrong with your heading and it's not in the wrong section bud, just probably not been seen yet. Don't forget at this time of year there's lots of other stuff occupying people. I'm sure Eddie will be along as soon as he gets online and sees it.

 

[eddie49]

Hello Nigel and welcome to the Forum !

Upgrades to the wirefeed assembly and motor power source are often a cost-effective way of improving a hobby-type MIG welder. The cast-alloy wirefeed unit is a good choice. Yes, it does have to be totally isolated from the chassis. The MIG wire is "live" at the torch tip, and all the way down to the pressure roller, the motor/gearbox, and on the spool. If any of those parts come into contact with the chassis of the welder, that will result in the chassis being connected to one side of the 15v-to-30v DC welding power, which will be either the positive for solid wire with gas, or the negative if set up for flux-core and gasless. [ In addition, the welder case will - or should - be connected to AC mains earth.] If the metal case of the welder then comes into contact with a workpiece ( e.g. the bodyshell of a car ) that has been attached to the clamp on the work return cable ( known as the "Earth" clamp ), there will be an arc at that point.
You will need to increase the clearance around the motor body, and use the insulated mountings, or change the bracket to plastic.

The existing speed control circuit on the PCB looks quite substantial. It appears to use a 2N3055 power transistor, and I think there is an 8-pin integrated circuit which could be a timer chip, so it may already be a PWM controller. As you have planned, it would be worth testing the welder with the new wirefeed assembly to check if that has already brought sufficient improvement. You may also want to consider cleaning or replacing the torch liner.

This type of machine uses the main welding supply to also drive the wirefeed motor. Depending on the setting of the power range switches, that voltage will vary, between about 15v and 30v. It will also change during the welding process. When the arc is struck the voltage will drop, and it will then vary depending on the length of the arc. Using an independent stabilised DC supply avoids these fluctuations and provides a consistent wire speed.

Finally, adding a PWM speed controller to replace the original - which is often just a linear current control - allows a wider range of speed and provides full torque at all speeds. As mentioned, this last step may not be needed in your case.

To be able to comment on your proposed wiring changes, I would need more information about the PCB. I suspect that it uses a Triac, rather than a relay, to control the main transformer. Could you please show the original wiring diagram, a full picture of the PCB, and read off the part number from the component mounted on the black finned heatsink ?

 

[Nige2108]

Hi Eddie, thanks for your prompt reply and valuable insight. I will modify the wire feed drive mounting according to your suggestions to ensure no part of it is in contact with the chassis. I recently replaced the liner with a length of curtain wire. The wire reel had been sitting around for the last 20 or so years so I unwound it until the wire became shinny before feeding it through, although I think I really should discard it. I have released the PCB and taken some more photos which will hopefully give you some further detail. The part number on the component mounted on the heatsink is 8TA26.600B (the 8 could be a B). Also attached is the original wiring diagram. Thanks again.

View attachment 266537

 

[eddie49]

Yes, that is a "B" in the part number. The BTA26-600B is a 25Amp 600v Triac, used here instead of a Contactor to power-up the main welding transformer. The thick black wires with push-on tags will be the switched mains feed out to the transformer. There is a 6-pin optocoupler which drives the gate of the Triac when you press the torch trigger. It was nice of them to include a circuit diagram in the manual! However, without a detailed PCB circuit it will take some sleuthing to add the external 24v power supply.
The 8-pin chip is a voltage comparator/op-amp. I'm sure it is being used to send variable-width pulses to the base of the 2N3055 power transistor, which drives the wirefeed motor, so this is already a PWM speed controller.
The blue plastic block is a small encapsulated AC mains transformer, which provides an always-on low voltage power supply to run the logic that triggers the Triac. It does not run the speed controller and wirefeed motor, power for that is tapped off from the main welding output and sent to the PCB on pins 5 ( +ve ) and 1 ( -ve ).

 

[Nige2108]

Okay thanks for the info, so would you say my wiring diagram for the power supply looks about right? I've cut the feed wire (5) and connected the +ve from the transformer and the -ve into wire 1 (not sure if this can just be connected and the original wire left in place or would it have to be cut?). The trigger to the relay is then taken from the torch wire 3 (+ve) and wire 1 (-ve)

 

[eddie49]

Cutting wire 5 may result in not being able to switch the welder on. The trigger switch has only one wire. Some machines, and all Eurotorch machines have two wires. When pressed, the switch joins the wire from pin 3 to the positive welding power feed. ( The switch is probably a brass leaf spring that contacts the brass block in the torch handle supporting the swan-neck.) However, the power transformer is not switched on yet. So I think that the wire on pin 5 is part of the trigger circuit for the Triac. Before the trigger is pressed, is there a voltage on pin 5 with respect to pin 1 ?

 

[Nige2108]

There is a voltage of around 1.8v between pin 5 and pin 1 before the trigger is pressed, increasing to between 18-30v (depending on what power setting is selected) when the trigger is pressed. Would that suggest the connection of the new power supply might be between 3 and the potentiometer by cutting that connection and then connecting the new power supply (+ve) into the potentiometer? You are correct regarding the torch. I have included a photo of the torch internals for your info. Thanks again Eddie for your assistance.

 

[Nige2108]

I've been measuring some more voltages from the pcb against wire 1 at the rectifier.

Pin 1 at the pcb against pin 1 at the rectifier 2v off trigger, 20v on trigger

Pin 2 18v off trigger then on trigger at the following wire speeds 15v@ position 1, 14.5v @ position 2, 14.2v @ position 3, 13.6v @ position 4, 12.5v @ position 5, 11v @ position 6, 8.7v @ position 7, 4.8v @ position 8, 0v @ position 9&10

Pin 3 18v off trigger, 0v on trigger

Pin 4 2.2v off trigger then as pin 2 on trigger at the various wire speed settings.

Pin 5 1.8v off trigger then 18-30v (depending on what power setting is selected) on trigger

 

[eddie49]

Nigel, thank you for the voltage readings. They are helpful, but still do not explain everything. For example, the 1.8v that is present on pin 5 before activating the torch switch is not enough to trigger a Triac, but that value may in fact be pulses synchronised to the 50Hz mains.

The internal view of the torch shows a gas valve as well as the trigger switch - since the welder doesn't have a gas solenoid.
The circuits on this PCB are more complex than is typical on a hobby-type welder. Normally there are two distinct and separate functions:

1. On/off control for the main welding transformer. That section typically uses a small mains-to-12v or 24v transformer which energises a relay on the PCB via the torch switch. In this case, an electronic AC switch - a Triac - is used instead of the relay. If the torch has only one thin control wire, the main welding feed in the torch cable is "borrowed" as the return for the relay coil circuit. I have not worked out what triggers the Triac here; it should be the wire on pin 3, but that goes to the large transistor.

Usually the small transformer only runs the relay. On this board, the AC output of the small transformer is rectified by the 4 diodes below it, and smoothed by an electrolytic capacitor. It appears to then drive the low-power part of the PWM pulse circuit based on the 8-pin voltage comparator chip. The high-power part, i.e. the 2N3055 motor driver power transistor, gets it's power from the main welding transformer, when that is on. The transformer/rectifier positive output connects to the PCB on pin 5, but it is actually pin 3, fed from the trigger switch, that supplies power to the collector ( the case ) of the 2N3055. Current flows out from the emitter, via a large diode, to pin 4, then through the motor to pin 1. This means that motor current is passing through the simple trigger switch, and up the trigger wire.

2. Usually the motor speed control section of the PCB is entirely separate from the trigger and on/off circuitry. Here, as mentioned above, the two functions appear to share low-current power. To be able to safely connect an external 24v supply, it may be necessary to cut a track on the board, but this can only be confirmed by reverse-engineering a circuit diagram for the board. This means identifying the major components and sketching a diagram by following the tracks.

Alternatively, you could abandon the PWM controller on the board, and connect a new controller to the external 24v supply. E.G. :
https://www.ebay.co.uk/itm/12V-40V-...-Knob-Adjustable-Speed-Regulator/162943733493
Both connections of the new 24v wirefeed motor would be wired to the poles of a two pole two-way ( DPDT ) relay. The two normally-closed contacts would be shorted, to act as a motor brake, and the two normally-open contacts would join to the new power supply. The relay coil could be 240v, paralleled with the main transformer, or 24v, connected to where the motor originally was. The original PCB will be retained, but the position for the speed control pot on the front panel would be replaced by the new one.

Have you tested the welder with the original set-up plus just the new wirefeed assembly? Regards, Eddie

 

[Nige2108]

Eddie, many thanks for your reply. Your understanding of electronic circuitry is truly impressive. it's all a bit of a black art to me I have to admit. I am still working on mounting the wire feed and will have to reposition the umbilical cable outlet as the new wire feed unit being larger than the original will require some offsetting horizontally to ensure correct alignment. I have switched the mounting brackets to plastic as you suggested and have checked for continuity through the mounting screws which is giving me a zero so I guess that's all good. I like the sound of your suggestion to use a new controller and relay for the new power supply. I have in fact already bought a 240v relay which I was thinking might have been a bit premature. Anyway I will get on with the wire feed installation, give it a whirl and then report back. I will certainly need some further guidance on the wiring of the power supply if that's next up. Thanks again.

 

[Nige2108]

I have now completed the wire feed and had a chance to try it out on some 1mm sheet. The wire feed does now seem much more consistent and certainly the motor has more power to push the wire through. I did some voltage checks across the motor both off load and with the wifes help on load while welding at and above the mid point in the wire feed range (range is from 0 - 10). They are as follows:


Wire speed 6 - 8-8.7v (off load)
Wire speed 6 - 7v (on load)
Wire speed 8.8 - 18.8v (off load)
Wire speed 8.8 - 16-17.5v (on load)

I'm thinking this voltage drop on load is not too bad?

Now as for the welding and trying not to be a poor workman blaming his tools I'm still think the improvement in weld quality is not that great. I accept I'm not a hugely skilled welder but on the other hand not a complete novice either. I also understand 1mm steel sheet is perhaps not the easiest material to be working with however it is the consistency that seems to be the problem.

The welder has four power settings 28A @ 100%, 40A @ 50%, 55A @ 25% and 80A @ 12%. I find at the 28A setting it is difficult to start the arc and when I do the penetration is poor even moving very slowly. At 40A starting the arc is more consistent although the penetration is almost a bit too much and results in a flat weld and often burning a hole especially near the edge of the workpiece. I generally find the best results are achieved when the wire feed is set at around 6.

The other common problem on both power settings is often when starting there is a pop, pop, pop sound and the wire is kind of pulsing and breaking (although the wire reel is rotating consistently during this), unable to really start the arc. I wondered if this might be an earthing issue although the clamp is right on the work piece (30mm wide steel strip 100mm long). Occasionally I can achieve the fried bacon sound and a reasonably good weld results.

So I'm not sure where this leaves me now really. I'm not sure if it's my lack of technique or the lack of performance from the welder or perhaps a combination of the two. Part of me wonders if I should buy a better quality machine to eliminate one of those uncertainties. If i then still have problems I will know its down to my technique!

As always your thoughts are welcome.

 

[eddie49]

Thank you for the progress report. From your measurements, the voltage at the wirefeed motor is dropping about one volt at lower speed,and about two volts at higher speeds, when under load by starting a weld. This is not a large change. However, due to the periodic sample rate of a digital meter, as it converts the analogue voltage to digital, stores the value, and displays it, its not possible to say what the instantaneous drop is, it could be more.
Due to this voltage drop, the wirefeed speed will drop. Even if the torch height is kept perfectly level, decreasing the wirespeed will cause the arc to lengthen. The voltage across a longer arc has to increase, so the wire will speed up. This stuttering process then repeats. The popping at the start of a weld could be due to this unstable voltage supply to the motor. However, it could also be settings and technique. I would suggest decreasing the wirespeed, holding the torch closer to the work, and tilting the nozzle either forward or back ( push or drag the weld pool ).
With a step-switched MIG, it is not easy to find the "sweet spot". Modern inverter welders, with a fully-variable output, make this easier. Between the voltage steps, adjusting the wirespeed will control the current. On thin metal it is normal to have to use a series of overlapping tacks, with pauses to avoid burn-through. It is often easier to make a start with thicker metal, e.g. 2mm.
Before deciding whether to add a new power source for the motor, you could start a new thread, with pictures, in the Welding Techniques section, to get advice about how the welds look.

 

[Munkul]

I'm all for trying to improve things, but 100amp migs are always gonna be 100amp migs. If its machine gunning and you've ruled out the obvious causes, then it's probably a power source past its sell by date...
You could fiddle with it til the cows come home and pull your hair out with it... Or you could get a bigger and better machine and wonder why you didn't do that in the first place.
Not trying to be critical of your work just that I know what misery it is trying to get a crap mig set to work, it's a rod for your own back.

 

[a111r]

True but it's no SIP. It looks suspiciously like an early Cebora re-brand. The board's more complex than most and it now has a 'nice' feed motor.
Transformers usually keep working, until they fail (over heating, then shorting) and then it's game over.
That torch, though....

 

[Hutcho]

True but it's no SIP. It looks suspiciously like an early Cebora re-brand. The board's more complex than most and it now has a 'nice' feed motor.
Transformers usually keep working, until they fail (over heating, then shorting) and then it's game over.
That torch, though....

Clarke did a version like this - same torch - and the board looks very similar to the body shop welders

 

[Nige2108]

Thanks guys for all your comments. I agree that ultimately you get what you pay for and I recall going back to the 1990's when I bought this machine it was only about £100 so it is way behind in terms of current MIG's. If I had a substantial project like a car resto or something I could justify the cost of a decent hobby machine however at the moment I'm really just doing the occasional job here and there. A few months back I had an outside job welding some 4-5mm angle and 12mm bar. My brother in law who used to weld with arc suggested I get an arc welder and I did buy a Parweld MMA arc machine which is a lovely little unit but mastering that technique is another learning curve I am currently at the bottom of. It did however prompt my interest in welding again and a desire to look at improving the old welder having read some of the threads here. Not sure looking back on that decision whether a new inverter mig might have been a better way forward but I guess it will be sellable if I decide that is the preferred option.

Anyway since my last post I have continued playing with the MIG and trying to improve my technique and with some increased effort in the metal preparation and further practice I am beginning to see some improvements. I know this topic area is not for discussing technique so forgive me for adding a couple of photos of what I've achieved with 1mm thick steel scraps but I am a bit more encouraged by the improvement from a few days ago. I'm also now seeing the wire length fluctuating between the torch and the weld pool and change in the motor sound as I'm welding which I'm assuming is due to the variable power to the wire feed motor.

As the wire feed power supply is really the last variable I have decided to press ahead and order the 24v laptop adaptor and speed regulator suggested by Eddie. This was only £14 or so and at least if I complete this I will then know I have done everything to eliminate wire speed variables. I have a 24v relay which I have temporarily connected to the wire feed motor connection and it triggers when the torch button is pressed. So I thought if I connect the adaptor supply into the 240v connector block from the mains on the welder then switch the output using the relay it should then run the wire feed motor without hacking into any of the existing wiring.

 

[Nige2108]

Hi all, just to conclude this thread I have now completed the power supply installation and I am pleased with the result. The wire feed and arc during welding is so much more consistent than it was before. Of course it's never going to compare with a modern unit but I have been able to produce some reasonable welds on 1mm sheet and I am now able to concentrate a little more on technique. Attached is a photo of a lap weld on a couple of pieces of scrap. I think it is a reasonable weld and the penetration from the other side looks quite good. Many thanks to all those who contributed to this thread particularly @eddie49 on who's guidance I carried out the mods