cup size 6 is hot glowing @ 150A

[babak]

Hi there,
these days I am pushing 150A current limit on my stainless steel tube roll forming line.
The welder machine is inverter based 250A machine, with;
Alumina Cup size 6
tungsten size 2.4mm (3/32")
Argon flow 10 lit/min (more than 20 Cfh)
everything is fine below 150 amperes, but just beyond this limit they all go wrong;
the arc sound changes,
the arc brightness changes,
tungsten tip starts melting,
cup gets overheated and starts glowing and after a while the arc blows.
According to the Millers handbook, my parameters are normal except that it suggests using cup size 8, but my question is: is it just about a bigger cup size which would make such a difference?

 

[AndyStobbs]

I'm no expert but it sounds as though your not removing enough heat from the tungsten.

 

[hotrodder]

A larger diameter cup is, well, larger. The sides of the cup are further away from the white hot tungsten and things that are physically bigger need more energy to heat up hence kettles take longer to boil when full

I don't understand how you can be melting a 2.4mm tungsten (or killing ceramics) @ 150A DCEN though... 2.4 tungsten should be good for ~ 250A DCEN and i don't have any problems with 2.4 tungstens or ceramics (usually a #7 and 6ish litres/min argon) at 200A AC which is harder on the torch in general but then my kit is water cooled

I'm guessing the torch is air cooled? What size is it? What sort of cycle times are we talking about?

 

[Brad93]

As mentioned above you are running straight polarity right? As in electrode negative and work positive?

 

[babak]

hotrodder;
are you talking about a continuous welding? because this is a production line. now, limited by that 150A barrier, I am welding continuously @ 2 m/min speed and stopped after each half an hour to change the tungsten.
the torch is Wp-18 (as written on it), it is water cooled I suppose, since there are 2 coolant tubes inside it (in-out), but the the ceramic and the tungsten are never get cooled of course.
about the polarity, yes, it is DCEN, no other choice!

 

[hotrodder]

Sorry should've mentioned i'm welding manually so obviously more stops/starts to grab a fresh length of filler wire etc which makes a big difference. The watercooler on my kit is controlled by the machine i.e. starts up when you strike an arc and switches off when the machines been left idling for 2 or 3 minutes and it's not uncommon for me to have the cooler running continuously for well over 1/2 hour. Okay i'm gonna be nowhere near as productive as a machine but from a heat POV i'm talking ally, AC, 200A and a smaller torch so i don't think the comparison is that unfair?

A wp-18 is manual torch typically rated at 350A DCEN 100% duty cycle. The machine specific version (wp-18p) is 500A @100% duty cycle.

The tungsten is watercooled albeit indirectly... with an aircooled torch the tungsten will continue to glow after breaking the arc for far longer than it does with a water cooled torch since coolant keeps the torch temp lower/more stable and the heat path is tungsten, collet, collet body, torch.

At 2m/min talking 60m of weld between 'proper' breaks but i'm guessing we're not talking about one continuous 60m weld? Assuming i'm not wrong about that how many arc starts are we talking about (what i ment by cycle times) and how are you prepping the tungsten- included angle and tip diameter? Not gonna make a 'night or day' difference but what flavour tungsten, 2% thoriated or something else? If ceriated then i'd expect a change to thoriated or 2% lanthanated to improve things with regards to tip life

What water temps are you seeing? The water return is used to cool the power line so as welding amperage increases the coolant has to deal with more heat at the torch head and the power line, could the cooler be running at/beyond it's capacity/dutycycle?

 

[babak]

hotrodder,
so, by all means, you have never noticed that ceramic glowing, right?

about your questions;
it is a one go 60m welding (sometimes more), and the machine is to weld continuously 2 shifts a day but it's still on paper!
about the tungsten prepping,
*red tungsten (Thoriated)
*tip dia = 0.4-0.6 mm
*cone angle ~ 18-20 deg (half angle ~9-10 deg)
As I understand, larger tip diameter yields longer life but makes more amps and larger HAZ, so it is a challenge to find the optimum point.

other parameters:
*Arc length (distance between tungsten tip and weld seam) = 2-2.5mm
*tungsten extension (distance between tungsten tip and cup) = 3-4 mm

about the water, it is a 400 liters reservoir circulated by a pump continuously. you are right, it has to cool down both, I check the temperature of the return water by hand during work to make sure the flow is enough. it never exceeds 30 Celsius.

 

[hotrodder]

Gotten 'em hot enough for the surface texture to change (very slick 'glaze') but never noticed 'em glowing. While the amperages were higher cycle times are miniscule compared to what you're talking about though. I was (wrongly) assuming we were talking shorter lengths and multiple arc starts with tungsten changes after ~ 60m of weld i.e. specialist sizes/low volume

Tungsten tip geometry effects arc geometry which in turn effects the weldpool geometry and it works kinda backwards in that sharper tungstens result in a more diffuse arc and a wider, shallower weldpool. The HAZ is outboard of the weld and fusion zone i.e. got hot enough to alter things but didn't melt. The HAZ is more about heat input (Kj/mm) and so the formula for calculating it talks about volt, amps, travel speed and process efficiency. AFAIK no dinstinction is made for tungsten shape, TIG has an efficiency of 0.8 in the formula regardless

As my world is manual welding this might not translate so well as i'm gonna subconciously react to the weldpool which is something a machine can't do (i need a LOT more tea breaks though!). Within reason (and very low amperage/delicate stuff aside) i've found that larger tip diameters improve tip life with no drawbacks. Tip angle has a greater effect on arc geometry but it's effects on the weldpool is something i compensate to automatically as i'm reacting to what the weldpool is doing/what it looks like. My thinking is that your included angle is too small... arc stability @150A should be a mute point, i would have thought 30 - 35° included angle would be the absolute min. Arc length also jumped out as a tad excessive to me but having never welded at 2m/min i don't know if 2.5mm max is actually excessive, whether your kit has voltage feedback for control of arc length or how well that works. As a general rule the arc length should be as short as possible

Any chance of a decent close up pic of an overloaded/melted tungsten?

Given how much material we're talking about to evaluate parameter changes i would be talking to the machine manufacturer if i was in your position, they should know them better than some random bloke an welding forum!

 

[babak]

here is the melted tungsten photo. you see there is almost nothing remained of the tip, but, it is quite interesting for me that this tungsten was working well @ 125 A with this condition! I let it work to see what happens next but until I stopped the machine to change the strip it was working. we talk about how important tungsten preparation is, so this question comes to my mind that, is it really that important?

of course it is much better to ask the machine manufacturer, if it is not Chinese! unfortunately there is no real support for this machine from the manufacturer.

 

[hotrodder]

Ahh, wasn't thinking cheapnese/chinglish.

After seeing the pic the first thing i'd do is buy some new collets and in future throw the tungsten away before they get too short to be clamped by the collet properly. That collet is FUBARed and the tungsten is soooooo short there's naff all contact for conduction of leccy and heat

I'd also avoid non branded collets regardless of price... decent manufacturers use a tellerium copper alloy which is much harder/tougher/heat resistant than plain copper. Genuine weldcraft etc aren't that much more but last a LOT longer

With manual welding prep is often overcomplicated on forums by people that've seen diamond tungsten grinders and the like. If the prep angles differ/the tip is slightly ****** etc a human driver can/will compensate or stop and correct the issue. A machine on the other hand will carry on at the same speed/amperage it was told to and (potentially) trash the part...

 

[wheresmejumper]

im probably wrong,but to me it almost looks like the tip is still there,but has been contaminated with a chunk of metal

 

[babak]

I will change the collet and tungsten as you said, hopefully things get better.
and, I checked miller website for tig torches. wp-18 is now called w-350 (after miller-weldcraft joint) but I did not find anything about machine specific torches, as you mentioned on wp-18p. the question is, what is different in a machine torch?

 

[babak]

im probably wrong,but to me it almost looks like the tip is still there,but has been contaminated with a chunk of metal

NO!... and yes! the tip has been melted to a thin hairy neck, but there is a powder like substance stuck to it. I am never an expert tig welder, but I think it is not contamination, maybe it is something due to poor Argon quality (as I have learned from tig handbook by Miller).
then, how pure Argon gas should be?

 

[hotrodder]

Machine torches are on the miller site under 'automation', overview here that shows the differences... http://www.millerwelds.com/pdf/spec_sheets/AY37-0.pdf wp-18p is now called w500 since the brand consolidation and the take down of the old (and much better with regards to torch info) weldcraft site

Gas purity is called out as either as a decimal number or a percentage for example grade 4.8 argon is 99.998% i.e. four 9s and an 8 is fairly typical. I think 4.6 is about as low as welding grades goes?

The tungsten doesn't really look melted to me either but i'm not sure what's going on... if there's a gas problem i'd expect all of the tungsten that isn't in the collet to be a blackened, charred mess from oxidation etc. The shaft looks sooty rather than oxidised to me while the tip looks 'frosted' and covered in deposits. Here's a more typical pic of how the tip usually degrades

If the shaft is oxidised rather than sooty then may be a gas problem of sorts going on too but that might even be due to the FUBARed collet as the argon passes around collet as it goes through the collet body or at least is supposed to- buckled collet isn't gonna help...
Blockages/leaks etc are FAR more common than dodgy gas

 

[Shox Dr]

Are you using a known make of tungsten. I had some unnamed Ceriated and they would form nodules on the end no matter what I did.

 

[derekm]

I'm a self taught amateur who welds mostly ally... but as I go up in amperage I go up in tungsten, cup size and increase gas flow accordingly. I found that it appeared to work.

However, with a back ground in physics I understand that DC plasmas involve ion flow and greater current means a stronger ion/plasma flow, and to stop air entrainment by this increased plasma flow, this will need a greater shielding gas coverage.

Try bigger cup and increased gas flow.

 

[babak]

seems that it was all about Argon purity. I found another supplier and everything changed to good!

 

[Robotstar5]

Thanks for letting us know you got it sorted.

 

[babak]

and this is the result.. tungsten after 2 hr working @ 150 A, with 2 interrupts of 5 minutes.. 2 hours and still working!