Residual magnetism

Date 04-10-2023 22:07

First off, a bit of background. We are welding some 17-4 stainless parts in the annealed condition with various joint configurations. The parts are of a generally thicker nature (2-4 inches). Some of these parts had a serious amount of residual magnetism in them after being machined that we need to weld. We sent some off to a local NDT company that has de-gaussing table that they use for de-magnetizing parts after magnetic particle inspection and prior to being put into service. They were able to successfully remove a significant amount of the magnetism and I am expecting that they will weld just fine now. They were off the charts on my -10/+10 gauss meter even while still 6-12 inches away from the parts. They got that down to -5/+5 in the worst spots.

I did some research prior to having them sent out expecting to find some decent information on what an acceptable level of residual magnetism is, but was disappointed that there was not very much information to be found (and the info I found was contradicting from website to website). I found a little bit of information for some drill pipe kind of stuff, which makes sense due to the nature of drill pipe, but it didnt really go into the detail I was looking for.

Is anyone here familiar with what level of magnetism is acceptable prior to welding? I am looking for a gauss reading that is quantifiable.

Any input would be appreciated. Thanks guys!

By TimGary (****)

Date 04-12-2023 11:58

While this is not the answer you're lookin for, general info -

A quick .pdf search of AWS D1.1, D1.6 and ASME IX shows no mention of magnetism. Not surprising as I don't recall ever seeing weld code restrictions on this.
However, I would not be surprised to see restrictions in various company's internal specs due to previous issues with specific product lines.
Perhaps there should be mention in national codes as residual magnetism definitely causes weld issues, but that would wind up causing undue and expensive additional testing that would hinder production activities.
The only time I recall having to deal with residual magnetism specifications is print notes from the Design Engineer on some specific armor steel weldments for military purposes.
Your particular case sounds like is in need of a thorough root cause analysis and resulting internal specification restriction.

Tim

By jwright650 (*****)

Date 04-12-2023 18:47

I'm curious why the local NDT folks could not get the level closer to zero. A wandering arc is not fun to deal with, so I understand wanting to eliminate the issue for the welders. I don't know what level would start to cause the issue in that I have never checked the levels before or after welding. I hope you will update the thread if you get a definitive answer to your question.

By twistedmetal208 (*)

Date 05-26-2023 15:29

I ended up speaking with our NDT level III and he suggested following the de-magnetization requirements of ASTM E1444. Per that standard residual magnetism after performing MT inspection shall be removed if greater than 3 gauss. This is what we aimed for and ended up not being able to remove all of it sufficiently in all parts. The ones that we could not remove enough gauss we sent off to be stress relieved and that process ended up doing the trick.

While the ASTM E1444 standard was not specific in the answer in regards to acceptable levels for welding, it has seemed to be adequate and we have not had any issues magnetism related problems during the welding.

I think if I am ever confronted with such a problem in the future I would just have the parts stress relieved from the outset. I feel like we learned some good lessons from this experience.

By 803056

Date 06-06-2023 23:09

Did you look at the Inspection Trends archives? there have been at least one article on the subject of demagnetizing parts prior to welding.

Al

By twistedmetal208 (*)

Date 06-12-2023 21:00

I have not. Do you happen to know a rough time frame? I would certainly be interested in finding that information.