Thread: polarised views on polarity

The welding world seems to be filled with opinions and anecdotes - finding hard facts can be a challenge. When I first got a DC welder, the question of polarity naturally arose, and I discovered that just as many swore that you should run DCEN as those advised using DCEP.

The same issue arises when trying to work out the difference between 6012 rods and 6013 - so many opinions, but so few facts

All that said, the somewhat convincing explanation of the effect of polarity that I've heard is that the electron flow is from neg to pos, so the positive side receives the flow of electrons and thus gets hotter. The result is that DCEN gets more penetration and heat into the work, DCEP puts less heat in, and can be easier to maintain an arc.

Similarly, I'm guessing that flux core MIG uses DCEN as there's no cooling of the wire by shielding gas, which you get with gassed MIG, which runs DCEP.

Unfortunately, I can't offer any authoritative references - if anyone can do so to support their argument I'd be keen to see them

i'm taking a stab at saying it has something to do with the shielding gasses around your weld pool. the ions i think their called? and their heat transferring characteristics.

This will answer some of the question without getting to bogged down in the detail
Figure 8-4.—Effects of polarity on the weld.
There's lots more on the same site if you really want detail

Originally Posted by WelderMick

OK, so I'm a guy who likes to understand stuff... I wasn't always like that, in fact, at school I don't think I displayed an inch of curiosity. But now that I'm older and realise that I'm as wise as I should be I find myself wondering about the important things in life...

For instance... the whole polarity versus heat input thing.

I don't think there is any argument that the majority (apparently two thirds) of the heat is concentrated at the positive electrode when TIG welding. So when welding DCEN (torch connected to negative and your work grounded to the positive) you get best transfer of heat into the weld.

Why is it so? and why is it two-thirds?

with stick welding it seems opinion is somewhat divided and there are reputable sources that quote conflicting information (e.g. Miller recommends DCEP for more penetration with stick, whereas others quote the reverse). I realise some rods are designed for a specific polarity, but still... this topic seems to pop up reasonable regular on the various forums but I have never found a satisfactory explanation as to why polarity affects the heat input into the job.

Why is it so? Can anyone explain?

Cheers

- Mick

Hi Mick ,

In simple terms, the cathode (the negative) gets cooled. If you need to understand the ins and outs here is a technical paper.

http://eagar.mit.edu/EagarPapers/Eagar109.pdf

Regards
Pete (Tokentools)

Originally Posted by RustyArc

I discovered that just as many swore that you should run DCEN as those advised using DCEP.

Yeah, in fact it was one of those threads, maybe a year ago now that has had me thinking about it off an on ever since then. Personally, for stick I used to be a DCEP man, but since then have learnt to be more flexible if things aren't working.

Originally Posted by brenton21

i'm taking a stab at saying it has something to do with the shielding gasses around your weld pool. the ions i think their called? and their heat transferring characteristics.

From what I understand, the different shielding gases have different ionisation energies and hence different arc plasma temperatures. For instance helium runs hotter than argon, so if you need a bit more punch for a given amperage you can get it by swapping or mixing gases. I don't think it explains why polarity makes a difference though.

[Edit: Actually Brenton... after reading a bit more from the document Pete linked, it seems you are spot on. It is the thermal conductivity of the shielding gas that makes the difference. Helium for instance is a much better at transferring the heat from the arc to the metal than argon. The fact that it has a higher ionisation temperature helps, but doesn't really account for how much better it is.]

Originally Posted by dubman

This will answer some of the question without getting to bogged down in the detail
Figure 8-4.—Effects of polarity on the weld.
There's lots more on the same site if you really want detail

Thanks Dubman... I couldn't really find what I was after in all that. I guess this is the sort of stuff that is quoted a lot, but I was after the deeper physics of it all.

Originally Posted by inverter_weld

Hi Mick ,

In simple terms, the cathode (the negative) gets cooled. If you need to understand the ins and outs here is a technical paper.

http://eagar.mit.edu/EagarPapers/Eagar109.pdf

Regards
Pete (Tokentools)

Hi Pete, now this I think is the heart of it and it seems pretty logical and is actually the first time I've seen something definitive by way of explanation.

From what I read in the paper and from your brief introduction, it seems that the voltage pushes the electrons from the cathode (-ve electrode) toward the anode (+ve electrode) and as they jump the arc they ionise in the shielding gas. The process by which electrons ionise and leave the cathode can be thought of as being similar to evaporation, so whatever the electrons are 'evaporating' from (i.e. the -ve electrode) gets cooled (just like water), whereas the reverse is true at the anode.

Thinking about it more, I think I understand why the heat transfer with stick welding is more debatable... with DCEP, in theory there is more heat being generated at the welding rod tip rather than the work, however, this heat doesn't stay on the rod as the molten metal quickly transfers the heat over to to the work. In addition, as the weld solidifies, the heat of this condensation is also transferred into the work. So whether you're running DCEP or DCEN probably doesn't make a whole lot of difference at the weld end (which has been my experience), but can certainly affect the heat at and hence vapourisation of the welding rod (and hence how the arc behaves and deposition rates). Is this right?

RustyArc's point about polarity and flux core wire is a good one... is it just the lack of cooling shielding gas, or is there some other reason to run DCEN. And why isn't DCEN recommended for all MIG anyway? I don't have a mig and don't use one... it's just idle interest on my part.

Cheers

- Mick

Hi Mick ,

Great analogy in regards to the electron evaporation, I like it. As to the mig wire that is flux cored and it's electrode negative polarity I beleive it is to help the flux shield vaporise closer to the job as opposed to the tip as the job is hotter. If it vaporises nearer the tip it will lose it's effectiveness.


Cheers
Pete

Originally Posted by WelderMick

So whether you're running DCEP or DCEN probably doesn't make a whole lot of difference at the weld end (which has been my experience), but can certainly affect the heat at and hence vapourisation of the welding rod (and hence how the arc behaves and deposition rates). Is this right?

Having Googled about the place looking for a definitive answer, and still finding statements both ways, I shuffled down to the workshop this evening and ran a series of beads with 6012 rods at 115A alternating between DCEN and DCEP to see if there was a clear difference in appearance, then cut the weld to see if I could identify a difference in penetration.

My conlcusion was that on average, the bead profile was pretty much identical between the two - variations in width and height were more to do with slight differences in my technique, and the amount of heat left in the metal from my last pass (I was using 50x50x3 angle). My attempt to identify differences in penetration were a fail as I simply couldn't see any deliniation between the weld metal and the base metal. Perhaps using thinner steel and looking at the back might be a better way.

The only difference I noticed was that with DCEN, the arc appeard to blow the pool backwards a bit, whereas with DCEP and the sam rod angle, the arc was more under the rod and the pool was right up near the tip.

The other interesting observation was that when using 2.5mm Tyeasy rods, it was impossible to maintain and arc with DCEN, while I had some limited sucess with DCEP. My conclusion is that I wouldn't recommend Tyeasy in either polarity.
Tyeasy® Series

Originally Posted by inverter_weld

I beleive it is to help the flux shield vaporise closer to the job as opposed to the tip as the job is hotter. If it vaporises nearer the tip it will lose it's effectiveness.

That sounds more plausible than my half-arsed theory - though it'd nice to find some reference.

It does raise the question - what problems would arise if running DCEN with gas?

Originally Posted by RustyArc

My attempt to identify differences in penetration were a fail as I simply couldn't see any deliniation between the weld metal and the base metal. Perhaps using thinner steel and looking at the back might be a better way.

Or flap disk the cut and use a penetrating dye. Or maybe just leave a very thin root gap, so you can see how far through its penetrated.

Originally Posted by RustyArc

The only difference I noticed was that with DCEN, the arc appeard to blow the pool backwards a bit, whereas with DCEP and the sam rod angle, the arc was more under the rod and the pool was right up near the tip.

So maybe DCEN for stick is bit better for out of position welds as the arc force can push the weld around a bit better?

Cheers

- Mick

I found another link http://canteach.candu.org/library/20053426.pdf which has some interesting stuff about welding arc physics (well at least I think it's interesting)