GMAW-P on P4 Material 2

[Mr168]

Does anyone here have experience using a pulsed GMAW process on P4 (Gr. 11, 12) material?

Using Lincoln equipment and running a downhill technique, we were unable to get a porosity-free weld using ER80s-B2 wire and a 90/10 Ar/CO2 mix.

When running the same process and comparable parameters on carbon steel with ER70s-6 wire, we could achieve perfect welds every time.

However, once the ER80s-B2 wire was used, we could not achieve a sound weld any time we ran over previous passes. No problems with an open root, but any time we pass over existing weld metal, even when ground/smoothed to bright, clean metal, we get porosity that is not open to the surface. We see it sporadically where the weld interfaces with the previous weld metal, and other times right in the middle of the weld bead. When running an open root pass, we did not see any porosity in that part of the weld, but would see porosity/inclusions when back grinding and welding the other side, or when welding the hot and fill passes.

We got the same results in the flat position, trying the ER80s-B2 wire on both carbon steel plate, and the P4 plate. We tried different heats of wire, different manufacturers, different gas bottles, different torch angles, different pulse wave forms, different flow rates, different interpass temperatures, different cleaning techniques... none of which eliminated the porosity/inclusions.

Needless to say, I'm baffled as to what the cause of this may be. Does anyone have any suggestions or input?

 

[metengr]

You might want to alter the shielding gas from 10 percent CO[sub]2[/sub] to only 2-3 percent. This really sounds like a shielding gas mixture problem especially if you are seeing this on both carbon and P-No 4 base plate.

 

[Mr168]

Met: Lincoln Electric is pursuing this for us, and that is going to be one of the first trials (both 95/5 and 98/2 mixes). What surprised me was that the ER70s-6 ran flawlessly on the carbon plate, but the ER80s-B2 would not. I did not think that the chemical composition between the two wires were different enough to merit changes as drastic as those we witnessed. We were forced to run a substantially reduced wire feed speed in order to achieve stable arc characteristics, but the more perplexing part was that I would have anticipated the porosity/inclusions to occur during root pass welding with the backside exposed to atmosphere, not on subsequent fill passes.

 

[stanweld]

E70S-6 has added deoxidizing elements of Mn and Si whose purpose is to provide additional protection. While Cr and Mo act as deoxidizers, that is not their purpose.

Are you using Lincoln's STT process and calling it pulsed transfer?

 

[Mr168]

No, this is a traditional pulsed GMAW process using their 25M suitcase feeder.

 

[stanweld]

I have had best luck GMA welding Cr-Mo alloy steels using 95Ar-5CO2 or 98AR-2CO2 mixes.

 

[stanweld]

Wire quality (cleanliness, seams, etc.) can also affect the incidence of porosity.

 

[DVWE]

You've got contamination somewhere, sounds like bad wire to me. I run GMAW, although traditional spray not pulsed, with the exact same shielding gas all day long with no problems. Try a different brand of wire (I use Bohler Thyssen).

 

[Mr168]

That's what I don't understand. We also run it with traditional spray and globular transfer with the same gas without issue. With the wire, we tried two different heats of one manufacturer which I despise, and then tried a vacuum sealed Metrode product from Euroweld which we have always found to be top notch. I'm also having a difficult time grasping that if it were a wire issue, why we would not be seeing the contamination on the root pass.

 

[DVWE]

Are you using the same machine to run all the plates and different wires?

 

[stanweld]

Pulse frequency, either too low or too high, can also promote porosity.

 

[Mr168]

DVWE: Yes. We have changed gas hoses, gun, and cable. Also had a Lincoln rep come in to return the programs to the factory defaults and update to the newest firmware.

Stanweld: We have tried multiple wave forms and trim settings to no avail, though it wouldn't surprise me if it were somehow involved.

 

[TubeWelder]

When I've run this wire in this application it was using Miller Equipment. Nonetheless, from the viewpoint of the guy behind the lens - you've got a combination of several factors, if I was running the process I would cure it in the following manner: use tri mix, change out your hoses (they might contain contamination), change to a ceramic liner, verify your wave form and power supply (erratic or spiking with other equipment coming on line and affecting the carrier will cause a problem), get your wire in a hermetically controlled environment and verify zero moisture content, check your purge valve and regulator to verify a tight fit and no environmental contamination, play with the speed and gas shielding to get the perfect combination for your application. Good luck - it can be done and is a very economical and efficient way to accomplish a high quality weld and ascetically pleasing product.

 

[Mr168]

Tubewelder: Thanks for the input. As someone who has spent a good bit of time behind the hood (and still do), I am still leaning towards the gas mixture being the key. I ran quite a bit of pulse on pulse aluminum some years ago as well as GMAW-P on carbon, so this entire situation threw me for a loop.

A few of your points we already addressed:

Electrically, we're ok. We are hard wired in with a bank of machines that are not in use while we are running this equipment, and it is a dedicated line to the test shop.

Wire was hermetically sealed and used fresh out of the package.

We are on the second regular and set of hoses, no issues there.

Running the 90/10, the parameters are as close to perfect as possible. Going in either direction on voltage (trim) or wire feed, performance begins to decline. It currently runs with almost no appreciable spatter whatsoever.


I too have faith in this process, and was really hoping to get a procedure qualified with it. Our application requires 100% PT, burn through is a concern, and they wanted a downhill procedure for production rates, so this process was ideal over a short arc procedure where lack of fusion would become a real concern given the abilities of our welders.

 

[Mr168]

Just as an update, Lincoln Electric has been doing some experimenting on our behalf. Using new equipment and a controlled environment, they too are getting porosity using 90/10, 95/5, and tri-mix. None showed any improvement over the samples we had originally produced.

 

[Guest102023]

I agree with Stanweld about the Mn & Si content in the wire. ER70S-6 will have more of both deoxidizers than ER80S-B2. Changing the shielding gas is an expensive remedy that covers a multitude of sins.
I would approach this by changing one variable at a time, in the following sequence:
(assuming there are no cleanliness issues)

1 - Different wire (a premium brand).
2 - Pulse parameters
3 - Travel speed, electrode angle, stickout
4 - Shield gas

Just a thought - could the heat of base metal be somehow implicated? Good luck - this is one more reason I got off the shop floor!

 

[Mr168]

Brimstoner: I'm in agreement. We isolated all of those variables. The pulse with these parameters with those silicon and manganese contents will not suppress the porosity, nor allow it to gas out before solidifying. The only time they were able to eliminate the porosity was by running the pulse so hot that it could not be welded out of position.

Even the "high quality" versions of these wires have the reduced silicon and manganese levels because they're designed primarily for wire fed GTAW applications. They have a 80S-B2 wire with a slightly higher silicon content and significantly higher manganese content on the way. That's really our last resort.

 

[Mr168]

Just as a follow up, the manganese content was the key. We successfully qualified the procedure, and the puddle was noticeably more fluid. The 90/10 gas provided satisfactory results.

 

[metengr]

Thanks for the follow-up.

 

[DVWE]

Mr168,

Thanks for the follow-up. Reviewing this thread, I assume that you increased the Mn content? What range are you having success with?