WELDING - Bolt 8.8 and S355

[ALANIT]

thread725-449248


Hello,

Since I've found a thread which treated this issue but it is closed now, I'd like to readress this question and add some reflections.

In fact, I've digged the EN ISO 898-1, EN 683-2 and EN 10025-2. I've found for example that the S355, by the IIS formula, that the equivalent carbon is about 0.52 %, and by the same formula some steels, which are suitable for 8.8 class, seems compatible with that (ex.: 25MnB5) and have even a smaller equivalent carbon amount. However, the 25MnB5 integrates BORE in 0.005 % max.

Why it should the 8.8 bolt should not be weldable to the S355 steel, with pre-heating? I'm not a weld expert, and I've always heard about to NOT DO this, but I'm asking: Why ? It seems that it should be possible, with the adequate precautions.

 

[desertfox]

Hi ALANIT

One should never weld nuts or bolts because the heat from the welding reduces its tensile strength, so an 8.8 grade bolt would normally have a capacity of 640Mpa but if you weld it that capacity goes out the window, that’s my understanding, it’s only really acceptable if the joint is not structural or taking a huge amount of tensile load.




“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[ALANIT]

Hi desertfox,

Thanks for your answer, and I agree with you that weld can alter locally bolt mechanics characteristics. But these bolts will be some tie rods in concrete, and not a steel-steel connection, so they will be loaded only in tension and their maximum expected stress is about 380 MPa instead of 640-660 MPa (unitary stresses), so nearly half than its real capacity. No shear load would be transferred by these bolts, since dedicated and massive shear lugs will be casted together. Do you think that would be really hard in these conditions to have a good welding?


I add these information for more engineering context:
Unfortunately, I tried to keep actual anchorage design with the EN 10025-2 steel grades for tie rods, (S355) but they are insufficient by some %. Since the EN 1992-4 covers only EN 10025 and ISO 898 steel grades, I'm forced to jump to 8.8 steel class.
Design changes are possible but they should be avoided, since the industrial workflow will a significant amount of time for a really small change...

 

[desertfox]

Hi again

It doesn’t matter whether it’s a steel connection or not, it’s he tensile loading that matters and at 50% or slightly above I wouldn’t risk it, fatigue might be a concern also if the loads are not static. Why not use the Anchorage design and specify a greater number of anchors?
Can you provide more detail of your situation? A sketch maybe.


“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[ALANIT]

Hi desertfox,

Another thank you for your answer. Unfortunately, the issue is that design is already fixed and cannot change. If not, I would simply increase diameter and/or change anchor arrangement in a more convenient way: I'm not a fan either of 8.8 grade solution, but one should explore all possible ways.
The maximum tensile force is calculated for an equipment under maximum seismic conditions with a pseudostatic method, so I don't think fatigue should be an issue (in normal conditions, no particular tensile forces acts on anchorages, max 20% of 640 MPa). But, it is more easy to avoid problems than demonstrate that there aren't problems (and you're never certain at 100%).


ARRANGEMENT:
The anchor arrangement is very simple, just one row: the problem for updating design would be simply the industrial process which is behind, like plans, approbations, etc...

------------------------

However, finally, I've decided to go for other steel in EN 10025, like S420NL or ML. A more suitable steel which is more adapt for welding.

 

[ALANIT]

UPDATE: the EN 1090-2 §8.2.1. forbid to weld threaded bolts.

 

[desertfox]

Hi ALANIT
EN10025 is structural steel are you still going to weld bolts to it?
The problem once you have welded the bolts is how do you know what strength they have?
A rule of thumb which leads to uneconomical design but safe is keep the working stresses below 25% of the UTS and a fatigue failure is unlikely.

Well it’s your call at the end of the day, you know all the risks and what impact any failure might have : collateral damage, injuries or death etc good luck

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[ALANIT]

Hi desertfox,

I don't understand: in the last message, I said that I've abandoned the 8.8 class and that I've chosen another solution, which is rolled bars in S420ML steel, welded to a S355J2 plate. The S355 is described in EN 10025-2 as a weldable steel, and S420ML (from EN 10025-4) is known to be a easily weldable steel, even more suitable for welding than the S355J2... More, most of actual studs available on the market are made with EN 10025 steels, which are structural steels.

Which issue are you talking about? There is something else that I've not considered?

 

[desertfox]

Hi ALANIT


I wasn’t sure what you meant from the above quote in your earlier post, however I see now that the S420 steel are round bars to replace the 8.8 grade bolts.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[ALANIT]

Hi desertfox,

Yes, the S420 steel will be the new steel class for round bars. If some expert gives me an advice, I follow that advice, I just like to comprehend the physics behind and its real meaning, besides prescriptions and norms

Thanks another one time!

 

[r6155]

ALANIT you said "I'm not a weld expert". Here is your problem.
NEVER, NEVER weld a bolt.
Again: NEVER

Regards

 

Let me augment r6155's statement:

NEVER, NEVER, NEVER, NEVER weld a fastener.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[ALANIT]

Hello,

@r6155: yes, that's why I asked. If I already knew the answer, I wouldn't ask... As I said in my first post, I've always heard that: NOT WELD high grades bolts. But it is not enough to me, and since I was exploring alternative ways, I wanted to know more and reasons. Desertfox has been patient and helped me through.

@ironic metallurgist: I don't agree with that. Fasteners' welding is very common worldwide, and it is the priviliged solution for NPP (almost 100'000 fastenings per unit, generally at least 4 fasteners per fastening, = more than ~ 400'000 welded fasteners per reactor). And they're all welded. You can look up for Nelson or Köco studs (example of commercial solutions). Even Eurocode 2-4 Annex F.3 cites directly fastener welding.

 

You can look up for Nelson or Köco studs

I'm a career welding engineer, so yes, I'll look up stud welding.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[ALANIT]

Then why if you know very well stud welding you say "never weld a fastener"? Really, I don't understand.

 

You asked a question, I gave my opinion.
I don't have time for debate today.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[desertfox]

Hi ALANIT

I found this article about welding of fasteners which might be of interest:-

https://www.portlandbolt.com/technical/faqs/bolt-welding-guidelines/

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[r6155]

@ ALANIT
Anyway, did you think about making a welding procedure test coupon?
I suggest you contact a welding engineer.

Regards

 

[mfgenggear]

I have pass experience with stud welding on sheet metal products,
it very product specific.

https://www.youtube.com/watch?v=-Yofk5RtRzw

 

[mfgenggear]

Alanti
I am forgetting procedures, however I prefer to weld structures in the annealed condition when ever possible, and post heat treat when ever possible. to be more predictable
of the parent material tensile and hardness. when it is not possible then extensive testing should be conducted of actual tensile and shear. in my experience there has been a lot of scrap produced.
when experimenting with no testing conducted. "even when heat treating with materials (as an example) I will run test parts". it has been a successful method. to verify results.
so in summary it will depend on the hardness and tensile requirements of your design. and what is and is not acceptable. for every rule there is an exception to the rule.
but when it becomes an issue of safety. there is only one rule. safety factor.