Welding Tips and Tricks

I recently purchased a cheap Chinese 200A tig. This one is branded as Cobel http://www.ebay.co.uk/itm/321042647185. I was quite impressed by it considering the price. Unlike many others this one uses a fabric sleeve on the torch cables so the cabling is very flexible. The foot pedal is also surprisingly solid but I would prefer a rocker type pedal. I changed the spring in the pedal as the original felt too soft to me.

Being curious by nature it wasn't long before I had a look inside. The electronics don't look too bad and the design is pretty basic so anyone with decent electronics knowledge should have a reasonable chance of fixing any faults. The boards are conformal coated to protect them from dust and damp and all plugs and many components are supported with hot glue which is good. One problem I did find is that the bolts on the bus bars between the power boards were not very tight. If left they could cause problems so I tweaked them up.

Like many other cheap Chinese tigs this one is fixed at 60Hz AC so I had a look at the controller board (mounted on the front panel) and found it uses a very simple R-C oscillator circuit to generate the 60Hz. One quick resistor change later and it now runs at 125Hz . I rather expected to have to change a component on the AC balance side to compensate for the different timing but it turns out that it handles the higher frequency just fine.

As the AC balance can handle varing frequency it would also be a simple job to fit a front panel variable resistor to give variable frequency control. Currently I am not to keen on doing this as it would invalidate the warranty. It is unlikely that anyone would spot one component changed on a board but a new knob on the front panel may raise some eyebrows!

Now for my question: what would be the best frequency to use if I have it fixed? I can probably select anything from 10Hz to about 250Hz. I don't know how good the driver circuit for the mosfets on the H-bridge is so I would be cautious about pushing it past 250.

In the future I may investigate making the pulse switchable to a high speed (currently 0.5-20Hz) but I don't know how well the inverter section will handle that.

For most applications I like somewhere between 150 to 200 there times when its nice to have the 400 on my dynasty 350 but I hardly ever use anything above 200. Balance is no more of a key to me then the frequency is.

Technically you would have probably invalidated your warranty when you looked inside, definately you would have when you swapped the component.
Keep the original and swap it back if the machine goes 'pop' or 'bang', or whatever noise chinese machines make...
I'd be interested to learn what effect the higher frequency has when welding, as the machine i'm using at college is a transformer type and is fixed at 50 or 60hz.

Landyman wrote:Technically you would have probably invalidated your warranty when you looked inside, definately you would have when you swapped the component.

Of course. To be honest I doubt if the warranty is worth the paper it's written on. This came from a box-shifter that doesn't know much about welders so it is highly unlikely they would spot the mod anyway.

I'd be interested to learn what effect the higher frequency has when welding, as the machine i'm using at college is a transformer type and is fixed at 50 or 60hz.

From my (limited) experience the arc becomes quite a bit more controllable and focused. At 60Hz it wanders around a lot. At 120Hz it is more stable. Looking through Jody's videos, he recommends 100-120Hz as a good 'all round' setting so I'll stick with 125 for now.

on my miller at work i find 150 works great. at home on my everlast i like 125.

Hi - I recently purchased the same Cobel TIG as it was all I could afford for how much work it will get - on DC it seems OK though the end amps on 4T won't go down low enough- but on AC it suffers from arc wander & I really got annoyed trying to figure if it's something I'm doing or the welder - I was amazed to read you were able to alter the frequency - would you consider posting details of how you did this - maybe with a picture of where on the board the resistor is & what value needs to be substituted - due to the warranty I've not had the courage to take the cover off mine ! as although it's cheap I could'nt afford to replace it just now. I can solder etc just fine & would like to get mine running at a similar frequency as this seems to be in the " sweet spot " most guides recommend as a start setting on adjustable sets till you develope some experience with a particular model. How are you getting on welding aluminum since changing frequency ?

I forgot to take pictures when I had it apart. If you really want to do this mod let me know and I'll take it all apart again. Unfortunately the front panel board is a pain to get at. The conversion itself was pretty easy. It was just a case of changing one resistor.

Upping the frequency does seem to reduce arc wander but keeping the arc short also makes quite a difference. The picture is of the first proper ali welding job I've done. The welds aren't perfect but they do the job. This is on 2mm (12G) sheet. I was pleasantly surprised by how little distortion there was.

I am really keen to make this modification to my own Cobel welder - so PLEASE for myself & anyone else with this welder let us in on the secret ! ( pictures would be great & what value resistor ) I don't know if it was your electronic skills or a bit of luck / experimenting that got you to the 125 Hz - but it would seem for a fixed frequency welder this is about the best all round - I had assumed that fixed frequency sets simply came out as whatever the mains frequency input was - not realising the frequency is being created electronically- I wonder why the manufacturer does'nt set it a bit higher - ( if it does'nt create heat or other problems in long term use )

Congratulations on your 1st ally weld - looks fine to me - if that is your 1st not just with the Cobel you must be pretty good - I won't be sharing my 1st attempt with anyone soon ! - despite buying a "Tig Finger" & Jody's favourite 2% Lanthanated tungsten & watching him make it look easy I still made a big pile of bogey - can't call it a weld ! luckily I mostly do stainless steel with mine - though I wish the darn Chinese had made some attempt to mark the amps on the front - must be the only welder you set by guess work - it surely would'nt have cost much to print a front panel with amp settings - when I get a chance I must get a clamp meter on a lead & try to get to roughly mark the knobs - don't help they don't appear to linear - still for the price I can't really complain - I would love a Miller Dynasty - but at UK prices I think you can get 10 Cobel's for 1 Miller !
All the best.

noddybrian wrote:I am really keen to make this modification to my own Cobel welder - so PLEASE for myself & anyone else with this welder let us in on the secret !

OK. I'll try to open it up on the weekend and take some pics. I can't believe I didn't take pics in the first place. I normally do when I do that sort of thing.

I don't know if it was your electronic skills or a bit of luck / experimenting that got you to the 125 Hz

It took me a little while to figure the circuit out because the layout isn't very logical. For instance the AC balance is on one end of the board but the 60Hz is generated on the other end. Once I figured out it was a basic R-C oscillator it was just a case of deciding on a new resistor value. I simply halved the existing resistor to double the frequency.

I wonder why the manufacturer doesn't set it a bit higher - ( if it does'nt create heat or other problems in long term use )

That one baffles me as well. My only guess is that they were designed by someone who had no real idea about welding. In electronics terms, 60Hz is pretty darn slow and I don't think doubling it will cause any issues. I certainly haven't noticed any heating issues.

Congratulations on your 1st ally weld - looks fine to me - if that is your 1st not just with the Cobel you must be pretty good - I won't be sharing my 1st attempt with anyone soon !

Thanks. That is my first real tig welding job. I have done lots of mig and some gas welding. I did quite a bit of practicing on scrap. I also have to admit I went back over a couple of the welds with just the torch to re-flow them and tidy them up. I also won't say how long this job took me or how many times I reground the electrode
So far the best way I have found is to hit it hard, especially when starting the weld. Really stomp on the pedal to get the pool started then keep plenty of power on and try to go as fast as possible. If you back off too much the arc won't punch through the oxide layer properly and it all goes to hell. I really haven't got my left and right hands synchronised properly yet and feeding the rod through my fingers is something I can't do at all. I just keep the rod long and take a break when I run out and my fingers start smoking

though I wish the darn Chinese had made some attempt to mark the amps on the front - must be the only welder you set by guess work - it surely would'nt have cost much to print a front panel with amp settings

Yes. That would help. Howeverf that would also mean the manufacturer would have to fit the knobs more accurately...

Here's a thought....

The resistances here are hypothetical...

If the resistor in the R-C oscillator is 3.3 Kohm, you could replace it with a 1 Kohm resistor in series with a 2.2 Kohm potentiometer, and have adjustable frequency for just a couple dollars more... At the pot's minimum, you'd approximately triple the frequency, and at max be very near the factory preset. (And, yes, I realize the frequency relationship to the resistance isn't truly linear, but I am speaking in generalities without knowing the values of the components in the ocillator.)

Just food for thought.

Steve S

Yes I did think of that. The only reason why I haven't done it is because it would totally invalidate the warranty. You can hide a resistor change but it is difficult to hide a hole in the front panel and a new knob.

By the way your guess was fairly close. If I remember correctly the resistor was 4k3 and I fitted a 2k2.

Good point. I was thinking, "As long as you're in there...", but you're right. I can change a single 1/4 or 1/2W resistor, and only a top professional will notice.

And, yes, I was guessing.

Sorry it took me so long to sort out a photo. I have been away. Anyway, here's the modified board. The arrow points to the resistor. The resistor was 240K 1/4W and I fitted a 120K 0.6W. A 1/4W resistor would fit better but I only keep 0.6W in stock. The board is soldered with high melting point lead free solder so taking the resistor out is difficult and it is very easy to end up lifting a pad. I would recommend piggy backing another 240K on top of the existing resistor. It would also make removing the resistor easier if you ever want to restore it to standard.
Note 240K is an E24 value. If you don't have E24 series resistors, a 220K should give about 130Hz or 270K would give about 118Hz.

Thank you for suggesting "piggybacking"

Not everyone has the skills to properly desolder a joint on a PCB, but most can solder a parallel resistor in and achieve the same result.

Thanks a million for this information - it takes the guess work out for those of us who are not quite as electronically gifted and is such an improvement for so little expense - I'll be looking to modify mine this week.
Cheers & happy welding.

Update on modifying Cobel welder - so far not successful in replicating Les's results.

Following the picture showing R52 & the value required kindly uploaded by "OP" I took the cover off the welder & while looking at the board saw it was actually easier to solder onto the back of the board as the resisor had quite long " legs " protruding due to less than perfect Chinese finishing !- as I did'nt want to fix the frequency after discovering that although increasing arc focus raising the frequency also reduces the effective heat input ( stated in Jodys video welding a large casting where he laso added helium to aid penetration available on limited amp set - he also reduced frequency down to 50Hz ) I took the logical next step by adding a potentiometer and fitted it in a small enclosure on the front of the set with a short piece of 2core cable connecting it to the board threaded through one of the numerous air vents - this solved the problem of adding a hole in the panel. The pot is a switched 1Meg - I have the switch in series with the pot & a 220k 1/4 watt resistor in series as well - so when the pot is all the way anticlockwise the circuit is open - turning the pot then completes the circuit at 1.22Meg & progressively reduces resistance down to 220k. sounded right to me !

So cover back on - all excited & lets weld - well the results are strange - the welder behaves normally with pot turned off - frequency does rise as resistance is changed but so does AC balance & by quite a bit - also with the pedal plugged in as you go past 3/4 pedal both frequency & balance change - to the point of virtually losing the positive side of the wave & the arc going almost silent - so I checked & double checked location of R52- correct - not accidently flowed solder across to anywhere else - bit dissapointed.

Next day - have a think - cable is relatively near output connections & although there is no attempt to protect board from interference could it be inducing feedback to oscillator circuit from proximity of HF ? Ah I got it this time ! replaced cable with screened & just to be sure put pot in metal enclosure - cover back on - here we go ( but not in a Shelby Stanga way !! )now that made it worse ! due to limited time currently for playing I've put it back to standard & everything is back to normal - just 60Hz - but predictable results. I realise I've not done exactly the same as the "OP" but cannot see why this does not replicate his results - if anyone with electronics knowledge has any input on this i'd be gratefull for their thoughts.
Bottom line - from my experience to date - anyone else with a Cobel set with any warranty left - think twice before messing.

Hey guys! So I have pretty much this same welder but is laid out a bit different and I'm having troubles doing this mod and I really want to do it! I'm not sure how but if somhow you guys can help that would be great! Does anyone know where the two ends of resistor r52 go to? Or to the OP how did you find it what where you looking for and what was that resistor I between?

I posted some pics so you guys can get more familiar with how this machine is laid out. It's a luxor 200p first thing I noticed is your main board is part of the front POT plate now if you can see in the pics mine has a separate POT panel on front and it's then wired to the main board. But I took it out and I'm just not sure where it is this thing is hard to trace back but there's one pic I posted with two arrows. The left one point at qlmk which is where the AC balance POT terminals are and the right arrow is pointing to the AC/dc switch. So do I trace back on these cuz I am finding only one little 10uf CAP in line right after a resistor but it also goes to a voltage reg in near the back so I'm not sure that's it. Any ideas?

Welcome to the forum Nissan20det.

OK - here's my take on it - the OP found this resistor & substitute value I believe by experience as he's something of an electronics guy / tech & as such found the circuit relatively easy to figure out - he was happy to change the frequency permanently for better arc focus at the expense of maximum heat input - I was not happy doing this as I'm often at or past the current limit on the set for what I use it for - for this reason I attempted adding a variable resistor as I had no desire to lose the low frequency capability - I never seen a fixed frequency set other than 60hz - so you got to figure this is done for a reason - he claimed that other parameters were unchanged by the modification & I won't argue this point as I did'nt exactly replicate his method - when I tried it it created all sorts of other problems & as the machine at that time was under warranty I never pursued the experiment so have been using it as standard since - while it's been acceptable for the price it has many quirks & is too small for my future needs so it will be replaced as funds allow.

Your machine is a generic WSME 200 from China & is one of many clones sold under various stickers / colors - it too has some quirks as it is reverse engineered by them from a batch made for a recognized brand - none of the factories making them I believe have an original schematic & all have some minor issues though none that stop it being used - they don't however behave entirely as they should though it varies as to the batch / factory exactly what they do - I spoke at length with the guy that imported some to this country with a view to selling them - he had 10 machines at a university that had a welding shop for extensive testing & they discovered various oddities - they did an updated schematic for the factory with a view to a sort of Mk 2 model with the faults ironed out - some were ready to be sent over but interest faded as there is only a limited interest for these machines on Ebay - most people having seen them are keen enough to spend a little more - the guy gave up the idea & stuck to plasma cutters which he had confidence in - there is nothing wrong with those WSME's as a starter set if that's your budget - but they are rarely seen here now - I think 1 guy on Ebay still has some but he does'nt care about feedback - he'll move anything there's a profit on - bad news is there is no similarity at all with the one in this post - the only other welders coming out of the same stable are a simplified model sold at a greatly inflated price by Eastwood & I believe the early models from AHP - have seen another color / sticker but can't remember what it was - their circuit designs share nothing much in common so unless you have the ability to draw out the schematic from the board on your machine & figure it out your at a dead end - you could try private messaging the OP - but I think he was a one hit wonder with this post - don't recall seeing him active on here since - so bottom line is you have a welder that will weld OK & is actually quite reliable / robust for it's price - but to attempt to change frequency is probably not practical or worth messing with it - if your only reason for doing it is arc focus / wander - just live with it - experience / seat time will help & the biggest thing is tungsten & shape / quality of grinding - all these nasty imports seem to suffer with chronic arc wander & are way more sensitive to grinding than anything I've used before - treat it as a stepping stone - if you like Tig as a process & outgrow it's abilities sell that welder on & move up a price range - after all what can we really expect from a £400 machine - with pricing here you could get just over 10 of these for the same money as a Dynasty ! - time to look at Everlast or Rtech.

Good luck with the machine & your future projects.

Holy crap I just wrote a four paragraph essay and all got deleted sweet stupid fing iPad ahhhh. AnyWho will try this again. Okay I agree with some of the stuff you're saying. Now most the problems and little issues people have with their welders that are like mine for some reason this Luxor brand doesn't have any of them. Thing welds like a Dynasty other that current capability. And what I noticed about the electronics in this is that they are more thought out then the ones like yours. And the schematic thing isn't quite true. I actually got one from my eBay supplier not that it helps me because it's not one for mine. But it is definitely one of the earlier drawings of these boards,Honestly it really could be one for your machine. One thing that I noticed when looking online at pictures of the electronics in your guises machines is all the boards say PK something and all of my boards say QK something. Now the schematics I have are for the PK boards. Here enjoy https://onedrive.live.com/view.aspx?res ... g-dEJ3g4GYlol. Now this was very far off from mine. this one only has a couple Zener diodes and Mine has like 100 really weird things. And you're right for me to really figure this out I'm going to have to reverse engineer this which I kind already started I have about a fifth of the board done and jeez that took hours. My father-in-law is an electrical engineer and he is been helping me with this but the wrong schematic doesn't help me. Although my eBay guy is working on getting me the build schematic for this machine from the manufacture. He said he could get it cuz he owes me( he one forgot to ship the pedal and two when I got the pedal it was wired wrong. When I pushed the pedal down amps would go down when I adjusted the POT it would change starting amps not Max amps like it should.lol. Anyone that uses the schematic if it is your machine let me know I want to know who's it is. And your welcome haha. Also if anyone knows what they're looking at on the schematic and wants to help me try to figure this out let me know and I'll post my reverse engineered diagram I made on digikey!

P.S. To read that schematic click file then download

@ Nissan20det

Well I never had anything Apple - but I don't know anyone that likes them - was a craze for some phone here they raved about when launched but is such a POS everyone has got shot of them & want's Samsung now ! luckily I don't follow trends - I got a PC given to me that was being thrown out of an office 10 or 12 years back - that's still running fine & I got an old Nokia phone I was given second hand by a mate - maybe a bit older when Deb was in hospital as there's no phone up my workshop - yet another example of the older stuff was made better & outlasts the modern stuff !

Thanks for the schematic link - I've not studied it enough to decide if it is mine & my particular model does'nt use every component there's space for on the boards for so there must be model variations using a common case / board - I think that's how you end up with that overpriced Eastwood thing.

I'm happy to hear your Luxor is free of quirks although some of them on those models don't show up that much when welding - only when extensive testing against a known load while being looked at with an oscilloscope - or it could be you have one that is from an original drawing or the 2nd generation model - the thing is there are huge numbers of factories out there ( entire districts in fact just making electronic products ) - none of them do much if any testing of the product - they simply receive the drawings & a target production quota - chances are they don't even know what it does let alone have anyone there that could weld with one to try - that's not something they're interested in - quality control is usually when the importer has more returns than he will accept & at that time they try to fix the issue for the next batch - not how I think business should work but that's the reality - big brand names that produce over there are very choosy about which factories they use & often have a representative there regularly or permanently ensuring that their product is being finished at a standard they are happy to put their name on - the issue with Chinese production is not their ability per say - just the price the importer is dictating & if they are working from legit designs given to them officially by the creator or as is often the case someone sees a product selling on Ebay & just shops around for the cheapest quote to clone it - at this point someone has to reverse engineer the item to give them the plans & often this is where things go bad.

Even if you manage to draw out your schematic just remember that the circuit as a whole may not respond as predicted by a simple component change without affecting other things - in my case the AC balance changed dramatically & once the foot pedal was connected that in turn created strange symptoms - it may not be worth the effort just to achieve variable frequency unless all you want is control on very thin aluminum - I realize many can't justify the cost but be honest what can you really expect from a Tig machine that's under £400 - & if you fix the frequency then there's the seizure inducing pulse rate & the list goes on - I'm not unhappy I got mine - it's done a few handrails &generally been useful - but now I'm convinced I need one as I don't have use of the big machine I used to borrow I'm looking to move up to a 250amp machine in the better import range ( £1500 ) - this has frequency up to 250 hz - pulse is still abit slow at 25 - but I can live with that.

Good luck & enjoy welding whether you mod the machine or not.

right on, well said
best engineering can turn to crud when it reaches production compromises and cost/profit

Ya speaking of addons my board has kinda the same thing so the luxors have 4 models a 180 DC tig in diff shell then a then a 200amp no pulse, 200p with pulse and a 250p so what I see Is nothing missing but there's a vertical board of the main board you can see in pics but it's most definitely a addon card my guess is the pulse feature, addon makes it easy to manufacture no pulse model, same board couple jumpers and addon card amounted and pow pulse. Just a guess but that's how I would do it. One question I have the pulse feature on mine goes down to 2hz now I would like to drop to .25hz on low end and maybe 350hz on high end instead of 300 do you guys thing I would have any problems if I did this mod? Would be very very easy to do but slow I think I'm fine but why bout faster?

There are far better qualified guys here - including Otto who may chime in as he is familiar with this post from the beginning - but as a layman I'd say most of the similar import machines I see have pulse down to 0.5 or so though this is not always precise - I see no problem there but assuming yours is the " normal " sort using MOSFET's I'd be very wary of going up anywhere near 350 - it's very rare to even go 250 on this setup - from what I know the MOSFET is very sensitive to heat based on the on/off time - if you don't allow enough dropout / off time the rating is much lower in amps - trying to take them up to that sort of frequency is asking for trouble & overheating issues - also I still believe component substitution in isolation that on paper should only effect one function often influences the circuit as a whole in other unexpected ways - the importer guy here had 10 machines for testing at a university with numerous electronic / tech / nerd types on the case & they struggled to get the results they wanted - I realize your particular model maybe an original or Mk2 not an inferior clone - but I still think your asking for trouble messing with what appears a brand new unit under warranty - I don't want to put you off & if you succeed then fair play & maybe post the mods / results - but those machines weld quite passably out of the box once you get used to them - I'd live with it untill you outgrow it & move up - as standard they are actually a pretty reliable if basic welder.

Oh, I'm not putting my dog in this fight...

There are so many variations in circuitry even in seemingly "identical" models, and I agree it's simplistic to think changing a resistor in an R/C oscillator won't have "downstream" effects.

The original poster seemed to have achieved his desired result, but where's the "year later" review?

I wouldn't attempt it, myself, based on the limited information provided here.

Your mileage may vary.

Steve S

Yep, there is probably a reason that feature is not on the cheaper units. If it was that easy, they would include it and it would be a strong selling point for the expenditure.