Weld cracking in dissimilar steels

[Kelvin]

This is the tig procedure I purchased from AWS.

Holy cow, $200 for a set of instructions? Is that typical? I've paid less to psychics for winning lottery numbers.

[ronsii]

Holy cow, $200 for a set of instructions? Is that typical? I've paid less to psychics for winning lottery numbers.

Welcome to the world of 'specs' and 'standards'

[12V71]

Welcome to the world of 'specs' and 'standards'

Like buying site prints to make a bid?

[ronsii]

Like buying site prints to make a bid?

That too

[shovelon]

Holy cow, $200 for a set of instructions? Is that typical? I've paid less to psychics for winning lottery numbers.

Tig is not a prequalified process in AWSD.1 and D1.6. Since qualified procedures are required it is quite cheap to buy from AWS compared to qualifying the procedure yourself. Usually CJP coupons are welded to a set of parameters. Then sectioned, bend tested, tensile tested, and to apply to PJP joints macro testing is done to support the CJP s qualification. If passed the minimum acceptance criteria the time, material, and lab testing all before reports are written can cost anywhere from $2000 to $4000 depending on the joint, post weld heat treat or not, and material involved. I processed one this summer welding grade 50 steel to 304 and 316 stainless with various bevel configurations independently and the costs were over $15,000. Well worth the cost to complete a multimillion dollar project for my customer.

Don't get me started on qualifying alum joints. Almost nothing is available for purchase.

[Kelvin]

Tig is not a prequalified process in AWSD.1 and D1.6. Since qualified procedures are required it is quite cheap to buy from AWS compared to qualifying the procedure yourself. Usually CJP coupons are welded to a set of parameters. Then sectioned, bend tested, tensile tested, and to apply to PJP joints macro testing is done to support the CJP s qualification. If passed the minimum acceptance criteria the time, material, and lab testing all before reports are written can cost anywhere from $2000 to $4000 depending on the joint, post weld heat treat or not, and material involved. I processed one this summer welding grade 50 steel to 304 and 316 stainless with various bevel configurations independently and the costs were over $15,000. Well worth the cost to complete a multimillion dollar project for my customer.

Don't get me started on qualifying alum joints. Almost nothing is available for purchase.

Wow, that's an eye-opener. Thanks for the info.

[husq2100]

Finally got them welded out.

This was my procedure:

Bolt up with the 7 bolts (not the arp studs I will use on vehicle), light tension.

Pre heat to 300c, tack a each bolt (7 points around dia)

check heat, weld out 50mm runs, checking heat as I went. I did not let it get below 270c and did not weld if it was over 310c.

after root run was complete a quick wire wheel and check temp, bring back to 300c and repeat for second pass.

remove bolts as fast as possible

post heat up to 500c, and kept there for 30 mins. After that I kept a bit of heat on the end flange and housing so it would not wick the heat out too fast. Then when the weld was down to 250 let it air cool.

[shovelon]

Finally got them welded out.

This was my procedure:

Bolt up with the 7 bolts (not the arp studs I will use on vehicle), light tension.

Pre heat to 300c, tack a each bolt (7 points around dia)

check heat, weld out 50mm runs, checking heat as I went. I did not let it get below 270c and did not weld if it was over 310c.

after root run was complete a quick wire wheel and check temp, bring back to 300c and repeat for second pass.

remove bolts as fast as possible

post heat up to 500c, and kept there for 30 mins. After that I kept a bit of heat on the end flange and housing so it would not wick the heat out too fast. Then when the weld was down to 250 let it air cool.

Excellent! Nice to see a plan come to fruition.

[ttoks]

bugger, I was hoping I was going to able to see it lol

What are your thoughts on post heat treatment? Spec say 1 hour, ttoks is saying 8-12 (huge difference) and id be struggling to keep it there for even an hour?

Sorry to come back to this so late.

I was thinking of the WPS for a main steam B3 weld (36" internal diameter with a 6" wall thickness) the WPS for that is 12 hours for the round trip, bringing up to temp, holding for an hour and back down which takes 8-12 hours, I overlooked that small detail in my first post

I would still be careful with that with the PWHT you were able to give it, 500C is far below what B3 normally needs, but Shovelon's guidance here is something to listen to, he has alot more experience with qualified WPS's than I do.

With that said, I've seen improperly heat treated B3 welds fails in spectacular ways in service which is what makes me nervous.

[Kelvin]

Sorry to come back to this so late.

I was thinking of the WPS for a main steam B3 weld (36" internal diameter with a 6" wall thickness) the WPS for that is 12 hours for the round trip, bringing up to temp, holding for an hour and back down which takes 8-12 hours, I overlooked that small detail in my first post

Holy cow, 36" ID pipe with 6" wall, what is that, like Schedule 10,000? It boggles the mind to even consider how much power could go through such a steam pipe...any idea what temperature / pressure the steam is? I bet it would cut through a pinhole like a water jet!

[ttoks]

Holy cow, 36" ID pipe with 6" wall, what is that, like Schedule 10,000? It boggles the mind to even consider how much power could go through such a steam pipe...any idea what temperature / pressure the steam is? I bet it would cut through a pinhole like a water jet!

just as a visual, this one is only a 12" pipe going into the turbine steam chest (the main steam like splits into 6 12" lines going into the HP turbine), Pressure is 2400 PSI at 600C (1100 degree's F), as for power, as for power, 740'000 horsepower at the generator, probably closer 1million in thermal energy going through a main steam line.



honestly it's probably more than 6", i've never actually measured it but a 9" grinder will only cut about 1/4 of the way through.

[Welder Dave]

That looks set up for narrow gap welding but the extended lip in the middle is to allow room to do the root pass? Is the root Tig welded or Mig and automatic or manual? I suspect there's some corrosion allowance as well. A friend of mine worked on some 400 ton cokers and the piping was 4" double extra heavy wall. They were using automated orbital Tig welders and had to be done to nuclear spec... meaning if the X-ray wasn't perfect, it couldn't be repaired. They'd have to cut the entire weld out and start over.

[Woznme]

That's a "J" prep. Allows access to the root with the stick rod while minimizing the number of fill passes compared to a bevel prep. There will be a standard gap in the prep.

[Welder Dave]

It's a U groove (double J groove) which is a type of narrow gap welding. Was more curious how the root was done.

[ttoks]

It's a U groove (double J groove) which is a type of narrow gap welding. Was more curious how the root was done.

root is done with TIG with an 6" long tip cup to be able to reach all the way into the prep.

[Welder Dave]

That sounds like it requires some experience to do.

[ttoks]

That sounds like it requires some experience to do.

the welding itself isnt actually all that hard, the hard part is that they're always pre-heated to 280C (540F) so not burning your hands is the biggest challenge.

also there isnt really any corrosion happening, the pipes are almost always hot, and a thick (like 3/32 thick) layer of scale forms on the outside, i've never seen one get pitting ect, failure mode is always cracking from heat cycling as the boiler comes on and off line for repairs or cleaning, a new bit of pipe generally lasts 20-40 years before they start cracking.

[Welder Dave]

Interesting, I know a lot of vessels add corrosion allowance. When I was working with a friend on a 2 -1/2" thick vessel welding the nozzle and repad it was pretty toasty. Used 2 tiger torches (weed burners) to preheat it and then poured pounds and pounds of 1/4" 7018. It was still hot when we came in for the next shift. It didn't help the repad was cut too big. Supposed to be 20" but cut for 24" nozzle. Shop figured it was cheaper to fill with weld than have a new 2-1/2" thick repad made. Repad also had to be cut in half because there wasn't enough room to weld the nozzle with the repad tied to the flange like most repads are. Time went by fast welding it though. Just pouring rod.

[William McCormick]

Hey all,

I started a thread here a while back

https://weldingweb.com/vbb/showthrea...r-rod-for-709M

but am getting closer to doing the job. Ill be welding some 709M (basically 4140) to the end of an axle housing flange. I do not know what the flange is made from.

I hope to have a piece of each for test welding.

I have ordered the correct filler rods for he 709M, but im concerned that there may be a problem with cracking and the axle flange material.

When you get cracking in the weld with dissimilar steels, is the cracking obvious to the eye, or only seen by X Ray?

If it were me I would use ER70S-2 and TIG weld it, no preheat other than the TIG torch until both base metals bond naturally. The reason is it is more forgiving and fights cracking, much better especially when you have two dissimilar materials, and with TIG we first blend the two base metals and then add the filler which connects them very well and becomes a material very similar to both metals. Chromoly is a very sluggish metal it starts to approach titaniums sluggishness, it resists heat.

You can also preheat and use a 7016 ARC rod.

You could also do ER70S-6 if you flame spray with Argon and also melt and bond the two base metals in the puddle, that would be a very good weld, the problem is getting penetration and an even weld, as well as having to keep it horizontal. As well as having problems starting hot enough without a run-in tab. Chromoly is a very sluggish metal it starts to approach titaniums sluggishness, it resists heat. That is why I TIG it slow and hot with only a little wire, mostly penetrating and blending.

The Germans make a lot of custom electrical connections using a pipe with machined flanges made of 4140 that have to hold up very well, they use a much harder rod but to be honest I believe they will be subject to cracking. I was asked to examine the weld by a German machinist, it looked like a weld that was loaded and just waiting to crack but he was raving about it. You have to look at the weld as a weak point and because of the structure created by the weld, a hyper-strong point, and not just make the weld strong, but allow it to flex or it will certainly crack next to the weld and not plastically deform.

Sincerely,

William McCormick

[William McCormick]

I guess first off..........this stuff ain't really supposed to be welded http://www.matweb.com/search/datashe...4e502e6f992c7d

If it is welded, taking into account the nearest ASTM steel is 4140, it should be welded in the annealed state http://www.interlloy.com.au/our-prod...el/?output=pdf. If welded in hardened state, it's a bit more complicated according to the specs.

Why you'd weld this stuff with TIG is beyond me, especially such a large piece that's gonna soak up a lot of heat. Most recommendations call for low-hy SMAW.

Sorry, I guess it's SAE 4140

It TIG's well, you just have to preheat with the TIG torch, bond the base metals, and move very slowly for good penetration. The ARC rod has trouble blending the base metals and penetrating even preheated but you can do it. It looks a little cold or you get undercut.

Sincerely,

William McCormick

[William McCormick]

The above post of mine has the weld prep and root run of test piece. This one shows second, fill pass and break. Note, I penetrated edge of hole and bolt. Bolt snapped on removal , this is not an option!

The reason for the funny start near right hand end is when I was welding I did a false start there, that is I started (wasnt thinking) then thought as the real pieces are round, there is no "end" to start on. So I stopped moved in a little and started again (hope that makes sense)

2.4mm tungsten at about 45 degrees sharpened
1.6mm filler
300c pre heat
140 amps for root
160 for second and 3rd passes ( I just looked up my notes as I did this a little while ago, I thought it was 2 passes but notes say 3?)

2nd and 3rd passes were done when test piece had cooled to 300c.

Smear this on the bolt before you weld.

https://www.amazon.com/Rectorseal-73.../dp/B008A3UFB8

It is pretty good stuff.

Sincerely,

William McCormick

[Lis2323]

Smear this on the bolt before you weld.

https://www.amazon.com/Rectorseal-73.../dp/B008A3UFB8

It is pretty good stuff.

Sincerely,

William McCormick

Good idea. Never occurred to me to use anti-seize.

Thanks!


Sent from my iPad using Tapatalk

[Barclay]

DMW cracking is another fabrication issue that can and does result in equipment failure. It usually occurs at the weld juncture where carbon steel or low alloy steels are welded to austenitic stainless steels in high temperature applications.

[shovelon]

DMW cracking is another fabrication issue that can and does result in equipment failure. It usually occurs at the weld juncture where carbon steel or low alloy steels are welded to austenitic stainless steels in high temperature applications.

That is why er309 is recommended. Personally I prefer er312. The both have the high ferrite content to deal reduce sensitization in Dissimilar Weld Metal joints, but I like the higher strength of er312. I will use 309L if the procedure or customer prefers it.

[husq2100]

Sorry to come back to this so late.

I was thinking of the WPS for a main steam B3 weld (36" internal diameter with a 6" wall thickness) the WPS for that is 12 hours for the round trip, bringing up to temp, holding for an hour and back down which takes 8-12 hours, I overlooked that small detail in my first post

I would still be careful with that with the PWHT you were able to give it, 500C is far below what B3 normally needs, but Shovelon's guidance here is something to listen to, he has alot more experience with qualified WPS's than I do.

With that said, I've seen improperly heat treated B3 welds fails in spectacular ways in service which is what makes me nervous.

I went by the material specs and research into stress relieving. I figured your specs took into account the massive thickness and size compared to mine. And as for air cooling after I had my heated sand set up, I was concerned about it cooling too slowly and changing the crystalline structure (from what I read) again, spec said "air cool". I did how ever make sure that no windows or doors open in the shed, to keep heat outboard of the weld (until it 250c) being a tube and it would cool faster. And after all that oxy heat, the shed was hot lol