Failed Vickers Hardness survey 3

[CandCQC]

I performed a test using a GMAW process in the 1G position. The matl was SA 516-70 and the plates were beveled at 37.5 each, leaving a .125 land. The root was open .093. I used a solid lincoln wire(ER70S-6). I used Argon(95%)Oxygen(5%)mix for a shielding gas. I failed my vickers hardness survey due to being above my max limit of 248HV. I did employ a temper bead technique by staggering from side to side with the last weld of each layer in the middle of the sample. Any thoughts on how I can pass this test without a PWHT process?

 

[vesselfab]

GO TO A ER70 S-3 WIRE

6 IS HARD

 

[metengr]

CandCQC;
Your technique that you described in your post is not temper bead. I have several questions;

What was the thickness of the plate? Did you use preheat?

How did you conduct the hardness testing? Was the unacceptable hardness in the base metal heat affected zone, the weld metal?

What is the basis for the hardness limitation of 248 VHN?

 

[SJones]

248 HV is very ISO 15156 for H2S service. Apart from the preheat, what arc energy was used? Can you 'turn the juice up'? Is the plate chemistry amenable to achieving that hardness restriction with typical GMAW arc energies?

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[CandCQC]

Metengr---I used a 1.50 thick plate;No pre heat, the plate was 73 F at the start and I limited the interpass temp to 400 f max. The testing was performed by an outside source. It was performed with a 10Kg load. The 248 HV max is a NACE req referenced to an internal spec for my customer. I am told by their NACE guru that this can be done with no PWHT but their own weld engineers are having trouble with this as well in a tig\stick process.
SJones---I used a GMAW spray arc. I ran a range of 253-350 amps however most of the welds fell into a range of 290-320 amps. I used 24-25 volts. As far as the plate I believe it is capable of producing the results that I am after but I am having trouble. Any input in regards to this is greatly appreciated.
vesselfab---I will look into that wire. Thanks.

 

[metengr]

CandCQC;
I would use a 150 deg F preheat for a coupon of this thickness. Second, I would increase the interpass temperature to 600 deg F instead of 400 deg F.

 

[DVWE]

I agree with the preheat and bump to 600F on the interpass. You could also increase your heat input:

Voltage * Amperage * 60/Travel Speed.

You can increase this just by having the welder slow down a bit and keep your settings the same.

 

[vesselfab]

i would preheat to 300*F, if a vessel you would have to any way.

I don't know if it will help hardness, but we use 75 CO2/25% argon for our shielding gas for GMAW.

 

[weldtek]

CandC,
I didn't see where you answered metenger's question as to where the hardness exceeded 248. Weld? HAZ? both?

 

[stanweld]

Section VIII, Div.1 requires a 200 F minimum preheat fot T > 1 1/4" and < or = to 1.5" without PWHT. Also increase interpass temp. per metengr. ER70S-3 will also produce lower hardness in the deposited weld metal as per Vesselfab.

 

[CandCQC]

Sorry metenger & weldtek. The hard spots were in the HAZ at the cap pass.

 

[SJones]

The classic location!

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[jrupert24]

Based on the 248 HV requirement, it seems you might be working with the NACE MR0103/Rp0472. Definitely want the 200F preheat, high interpass temperature and temperbead as suggested. We have changes to ER70S-3 for the weld metal as well, however, it doesn't seemlike that might be your issue. Make sure your final passes are not flush with the surface of the test block. You should run at least one extra layer and maybe two making sure the toe of your temperbeads are not more than 1/16" from the edge of the welds touching the base material. This will temper the HAZ in the cap pass most of the time. ASME Section IX-07 has a good diagram of temperbead welding in section 462.12.
Another thing you may want to consider is that not all heats of 516 Gr 70 are equal and if it contains residual or intentionally added elements such as Cb, V, and Ti you may face some difficulties. NACE SP0472 and NACE 8X194 have good information on base material chemistry. One last suggestion would be to weld your test plate and cut one section to send to the lab first for a hardness traverse. If you encounter a problem, you still have the rest of your test block to run additional passes or see if a half bead technique is needed vs the temperbead before sending for the final hardenss traverse. I am spoiled here as we have our own lab for Vickers hardness traverses.