Using 95/5 Shielding Gas During Low-Alloy Filler Rod Conformance Testing

Given these constraints, the welder at the testing lab now has been charged with the task of producing a defect-free weld. But there’s just one problem. Although 98 percent argon/2 percent oxygen shielding gas yields a stable arc, the arc energy is relatively low due to the small percentage of “active” components in the shielding gas mixture. To ensure a spray transfer and good bead wetting, the welder’s first inclination will be to increase the arc voltage setting on the power source – and there is absolutely nothing wrong with this approach.

The voltage range allowed in the testing lab’s WPS is 27 volts to 29 volts (and for comparison, AWS A5.28 allows the voltage to vary from 27 volts to 32 volts for .045 in diameter wire). So does this mean that if the weld does not pass X-ray it must be 100 percent the welder’s fault?

Not so fast, my friend! As the output voltage is increased, the output current will also increase if all other variables are kept constant (wire feed speed, CTWD, etc.). So in the present situation, when the output voltage is turned up to a mere 27.5 volts, the output current runs right up against the maximum of 360 amperes imposed by the AWS A5.28 filler metal spec.

Because of this, the output voltage was not permitted to be turned up to the “sweet spot” and the effect was that the arc was erratic and the puddle was somewhat sluggish. This made it difficult for the welder to manipulate the puddle and wash the molten metal to the sidewalls of the V-groove joint.

But most importantly, the resulting weld had scattered internal porosity, with the defects being numerous enough to result in a failed X-ray radiograph test. Even if the conditions were perfect, 98/2 shielding gas can still provide welds that are susceptible to internal porosity due to the resulting low arc energy and narrow weld bead papilla.

However, in this case there was no technique in the world where manipulating the torch angle, weave width, or weave frequency would alleviate this porosity problem on a consistent basis. The lack of arc energy resulting from this 98/2 blend, combined with the constraints outlined above, has created a situation where the weld puddle is not fluid enough to allow the trapped gases to escape from the weld puddle before it has solidified.

Thankfully an accomplished welder by the name of Dan Gaul suggested a simple, yet effective solution that seemed to remedy this problem. Based on his previous experience, he recommended that a 95/5 shielding gas should be substituted for the 98/2 mix that was causing so much consternation.

Of course! It’s widely accepted that further additions of oxygen have been shown to result in a more stable arc, as well as a reduction of the threshold current to spray transfer. And, upon close examination of AWS A5.28, one can see that when performing conformance testing of an ER90S-D2 electrode the use of an argon shielding gas containing anywhere from one percent to five percent oxygen is permitted.

However, a five percent oxygen mixture is not without its drawbacks: With an increased level of oxygen in the shielding gas, a higher degree of oxidation of the weld metal occurs. Because of this, alloying elements such as silicon and manganese are lost between the welding wire and the puddle, resulting in a reduction in weld metal toughness.

If a shielding gas with five percent oxygen is used for low-alloy steel welds, care must be exercised to select a solid wire with enough deoxidizers that will ensure that a minimum Charpy V-notch impact energy of 20 ft-lb (at a test temperature of minus 20 deg F) is achieved.

In the end, the five percent oxygen blend presented the advantage of a more fluid puddle. And the benefit of this gas was the means for the trapped gases within the weld puddle to escape before solidification. The radiographs of the multiple pass groove welds made with 95/5 gas showed significantly less porosity. The ER90S-D2 solid wire could now be tested with the same restrictive parameters, but without the same heightened fear of weld defects.

So what is the take-home message here for the fabricator?

Put away the handcuffs. Just simply state in the contract documents that testing should be conducted “in accordance with AWS A5.28”, rather than stipulating that welding should take place with a particular blend such as “with 98 percent argon/2 percent oxygen shielding gas”. After all, the procedures utilized during conformance testing will usually not reflect those that will be implemented in the fabricator’s shop.

In other words, the heat input, wire feed speed, etc. will be different, so why require that the shielding gas be identical? If the American Welding Society (Miami, FL) has seen fit to provide this latitude during testing, it must have been provided for a reason. This will allow the testing process to advance smoothly, and leave some wiggle room for the lab to make adjustments as the situation warrants.