Welding 410SS to A36 structural steel

[KFmecheng]

As part of a larger project I'm working on, I need to specify the welding of a 410 stainless steel plate to an A36 structural steel I-beam. The I-beam is partly embedded in a concrete structure. The purpose of the stainless steel plate is for wear and corrosion resistance. A self-lubricated bearing pad slides on the plate. This is a modification to an existing structure, so all of the work will need to be done in-situ. I am not a welding expert, so any guidance you could give would be very helpful. Some of the concerns I have are:

1. What is an appropriate filler material?
2. Is pre-heating and post-weld heat treatment required? If so, how can that be done without damaging the concrete?
3. How can I ensure that the plate isn't distorted due to the welding process?

Thanks in advance.

 

[metengr]

In lieu of using 410 SS, why don't you consider an AR plate material, like AR-400? This is an abrasion resistant (AR) steel plate that can provide the same benefit as 410 SS and is readily weldable to A 36 without PWHT.

As far as corrosion resistance, I don't know all of your specific concerns. However, if you are selecting the 410 SS for this aspect of the job, I would give serious consideration to the AR-400 plate alternative.

Regarding welding, using AR-400 plate, an E7018 fillet metal is appropriate and welding at 50 deg F minimum preheat is all you need. Using 410 SS, bigger headache because of preheat(400 deg F) and post weld heat treatment concerns.

 

[KFmecheng]

As far as corrosion resistance, part of the plate will be continuously submerged in fresh water (a river), and part will be above water. After welding, the plate will need to be machined in-situ (the area the plate is in will be de-watered at the time) to achieve some fairly tight tolerances and a surface finish of 63 microinches. We need the plate to keep this sort of surface finish for a design life of at least 50 years.

Perhaps I should have said hardness rather than wear resistance, since the plate won't see that much wear. There will only be a few cycles per year of the part that rides on it. However, we want the plate to be harder than the lubrite bearing pad to prevent them "welding" together. The bearing hardness is 223 Brinell. The wearing plate should be 50-100 points harder.

 

[metengr]

I believe the AR-400 plate is a good option.

 

[KFmecheng]

I'm concerned about the corrosion resistance of AR-400. Do you have any information on that? The original design, built 40 years ago, used Cor-Ten material. It has not held up very well. Part of the problem is that some road salt can get into this area as well, which has accelerated the corrosion. We really need something in either stainless steel or else with equivalent corrosion resistance. If welding of stainless steel can't be done, we may consider bolting it on.

 

[metengr]

Another option to consider would a be duplex stainless steel that still affords good weldability to A 36 (minimal preheat and no PWHT) and does have better corrosion resistance in comparison to AR-400 and Corten. The duplex stainless in this service can be locally work hardened to provide increased wear resistance. Cost will be the driver in comparison to AR-400.

 

[stanweld]

Other materials to consider are some of the "super martensitic" stainless steels based on cast alloy CA6NM. These are much more weldable than the martensitic 410 grade.

 

[KFmecheng]

What would happen if we used 410 and didn't do the heat treatment? These welds are not really structural. They are just holding the plate onto the beam, but they won't see significant stresses.

 

[metengr]

Because of the hardenability associated with 410 SS, the fillet weld toes would have increased susceptibility to crack initiation and propagation. The source of the tensile stresses could be a combination of residual after welding and flexing of the bearing plate in service. In addition, having the fillet welds partially immersed in water sure doesn’t help matters.

 

[DAVIDSTECKER]

Have you thought about making a pocket for the plate to sit in buy boxing it in with flat bar or angle iron?
You can butt the bars up to the plate and weld them in on site so there would be no gap.

 

[GRoberts]

If you decide that 410 is the way to go, you could avoid any preheat or PWHT in the field by buttering the areas to be welded on the 410 plate in the shop with 309 filler metal (preheated to 400F), then give the whole plate a PWHT. The weld to the A36 in the field using 309 filler metal without preheat or PWHT.

Any way you manufacture this, be carefull about distortion. A welded plate seldom comes out flat enough for a bearing plate without further machining unless it is very thick. If you welded a stub beam to the plate in the shop, you could also preheat and PWHT in the shop, then machine the plate flat, and weld or bolt the stub beam to the existing beam with much less distortion. If you can weld the beam to the plate directly in the shop, there isn't a need to use the 309, although it will work well. 410 or 410NiMo filler would also work depending on PWHT temperatures.

 

[metengr]

Be careful with dissimilar materials in this application - aqueous environment is a source for galvanic corrosion.

 

[KFmecheng]

Would a material like 17-4PH be any less susceptible to cracking in this application than 410 if not heat treated?

 

[EdStainless]

I favor GR's idea to either butter the contact points or weld stubs/tabs in a shop where you can preheat and PWHT. Then make the easier welds in the field.

You could use 17-4/17-7/15-5 or any other PH grade. But you will need to weld tabs and then heat treat the plate.

Are you sure that a duplex SS will not work? What is the mating material? I presume that this is a flood gate or some similar structure. Is there anyway to force this to be flushed with water? The dirt/salt along with a wet/dry spach zone is a tough environment.

= = = = = = = = = = = = = = = = = = = =
Corrosion, every where, all the time.
Manage it or it will manage you.

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