Welding 410 SS 2

[weldtek]

I'm looking for suggestions on making a successful weld in 410 material. I recently completed a test plate in 1" thick material which was welded using the SMAW process with a 400 F preheat, followed by a 600 F DHT and a 1400 F PWHT but unfortunately we had problems with very coarse grains resulting in mechanical issues.
It has been suggested that we change our PWHT procedure from a ramp up rate of 400F / hr, to increasing the temperature as quick as possible to 1000 F, hold at 1000 F for an hour and ramp up to 1400 F at 400 F/hr and cooling from a 1000 F to around 600 F fairly quickly. Any thoughts?

Thanks,

John

 

[GRoberts]

Where were the coarse grains, and how did they affect the properties. How did you verify coarse grains are your problem? Since martensitic stainless is transformation hardening, grain coarsening is not normally an issue like it is for ferritic stainless.

 

[weldtek]

We lost a bend and a tensile coupon, and the impacts were very low. The lab tech says the grains appear to be columnar and very large. I haven't seen the specimens yet.

 

[EdStainless]

Is this material martensitic or ferritic? What is the C level?

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[GRoberts]

What temperature are the CVNs? This materail has genrally poor impact toughness in any state. Also, sub-critical PWHT will not affect the grain size. You have to get hotter than that to start grain growth. If the grains really are that big, they were that big before you gave it a PWHT.

 

[weldtek]

This material is S41000, C is .08 - .15, I think this particular plate runs ~ .12.

Impacts were run at -20 F.

 

[GRoberts]

I don't think getting decent impacts from 410 at -20F can be expected. We have done room temp CVNs before, and didn't quite make 20ft-lb for the lowest HAZ value.

 

[weldtek]

At this point I'm not concerned about getting impacts, I just want to get the bends and tensiles to pass. I'm wondering if I need a lower interpass temp to ensure that grain growth does not become an issue?

 

[GRoberts]

If you lower the preheat/minimum interpass temperature, you might run into cracking problems. 400F is an appropriate temperature for preheat on 410 stainless materail. The only way to reduce grain growth (if that is your problem), is to lower the heat input of the weld. The DH treatment and PWHT will not do anything regarding grain growth.

 

[metengr]

The 410 ss that is subjected to PWHT in accordance with ASME B&PV Code requirements should pass bends and tensile tests resulting in a successful weld, excluding impacts at this point in time. My guess is that the coarse grain growth in the base metal HAZ was from the prior thermal mechanical treatment that was done to the plate. What was the incoming hardness and heat treatment of the 410 ss plate?

I do not agree with lowering the preheat or interpass temperature or using an aggressive ramp rate to reach the proper PWHT temperature. The grain growth has more to do with prior cold work. Go back, and check the MTR and associated heat treatment of this plate.

 

[weldtek]

Thanks for your help.
The test plate material MTR indicates that it was annealed > or= 1400F for 35 min hr per inch of thickness and air cooled. C 14.4 BN 156 and a grain size of 9.

I've obtained new test plate material C .11, annealed at 1300 F min, grain size 6.

 

[weldtek]

The AISI has a publication entitled 'Welding of Stainless Steels and Other Joining Methods' which suggests for Martensitic grades a preheat of 500 F followed immediately by a PWHT of 1350 F - 1400 F with controlled cooling to 1100 F ( 50 F / hr) and air cooling from 1100 F.

 

[GRoberts]

The plate with a grain size of 6 has bigger grains that one with a grain size of 9.

http://metals.about.com/library/bldef-Grain-Size-Measurement.htm

What location were the large grains your lab-tec was looking at - HAZ, weld, base? If base metal, your lab-tec and the MTR reporting a grain size of 9 are in conflict.

 

[metengr]

weldtek;
I believe the problem has more to do with the original heat treatment of this plate. Was the plate given a proper austenitization (1700-1800 deg F) and air cool heat treatment before tempering at 1400 deg F? I have seen where the plates are hot formed, and tempered without going thru an austenitizing heat treatment.

What I would do is to go back and have a quench and temper heat treatment performed to assure 100% formation of martensite. At this point, the plate can be tempered at 1300 deg F for 1 hour to assure adequate tempering.

I have qualified 410 ss with bends and tensile testing using the heat treatment mentioned above with no problems. Impacts are another story….

 

[EdStainless]

I have to go with Metengr on this one. I suspect the original heat treatment. The only thing that you could have done to have much effect on this is the weld heat input.
At you carbon level you should get finer grain size and resonable toughness, but don't expect much.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[PAN]

weldtek,

What is the specification of your welding electrode?

 

[0707]

I would weld your 410SS material with AWS E309-15 or E310-15 with a 400 F preheat and without any PWHT.

 

[weldtek]

Thanks to all for your help. I wanted to update you on our effort. It was brought to my attention that the electrode chemistry is critical, in particular with respect to carbon, manganese and silicon, and, that these elements should be at or close to, .1, .5 and < .3, respectively. When I checked the MTR for the consumable we used I found that, although the levels of these elements were well within the spec, they were not close to the levels stated above. We located some electrode with this recommended chemistry and the test passed all testing. Charpy's were not included. I'm not saying the plate didn't play a role here, but, with one bend coupon breaking in the weld it seems that the weld was suspect. I wanted to pass this info on as I wasn't aware that one needs to discriminate to this degree when procuring electrodes for this application.

 

[weldtek]

Correction Mn should be > .5