Low Temperature Welding (-75f)

By Sourdough (****)

Date 07-11-2007 16:02

PRE-HEAT, period......

By bsalonek Date 07-12-2007 01:39

Thank you Sourdough,

I do understand the importance of pre-heat. I

By js55

Date 07-11-2007 18:21

There are a variety of mateiials that can be applied based just upon your service temp. Austenitics (either SS's or Ni's) demonstrate little or no transition to brittle fracture and are a safe choice. Your engineering will involve however more than just temp. Service medium, pressure, fluid velicity, and even service life estimates should be involved.
As for low temp applications with ferritic/bainitic/martensistic type materials (low alloys, generally specifically Ni steels, or CS's) keeping grain size down is generally a must for achieving the impact strength required. This requires welding heat regimes that allow for rapid cooling. Interpass and heat input maximums are the usual control methods.

By bsalonek Date 07-12-2007 01:42

Thank you Js55,

I understand that we have to look at the project in a overall light, with regard to plant life and other conditions. Builting the plant out of SS would be extreamly cost prohibitive. I was looking for guidance in possibly using CS with a hight Ni content that standard A36 structural steel.

By js55

Date 07-12-2007 13:18

If you're looking for low temp low alloys try SA-333. Grade 6 is fine grained carbon steel with a specified impact test temp of -50degF. This one is popular for nukes looking for automatic impact testing. Grade 9 is a common nickel steel commonly used for low temp applications, often LNG. I don't remember off hand what the specified test temp is for 9 nickel. But ASME SA-333 is the standard for low temp seamless pipe. There are alloys in this spec in between the two grades with varying low temp impact testing regimes.

By CWI555

Date 07-13-2007 04:11

A553 would work as well. SA 333 ranges from a little over 2% to 37% percent nickel depending on the grade tested at the temps you've specified, 553 is 7.5 to 9.5% depending on the grade (9% nickel) and is tested at -196C and A333 is -195C (no signifigant difference). Either will have excellent low temp serviceability. Most LNG tanks these days are being built with 553 (as is my current project and my next one to come) It's tested at -196C. The normal minimum design range is -165C. Depending on the grade of LNG, it's liquefication (LNG) point is between -161 and -164C. SA 333 grade 8 is an excellent choice for piping, but most don't like the cost. A358 304SS is used more commonly.

By js55

Date 07-13-2007 13:10

Gerald,
Yes, Grade 8 not 9. Thanks. Been awhile.

By js55

Date 07-13-2007 13:16

Gerald,
What kind of impacts are you seeing at those temps. And what Grade is being used?And if available, do you have any lateral expansion data?
thanks

By CWI555

Date 07-13-2007 16:33

A553 p11 type 1 gr 1. A5.14 ErNiCrMo-3 weld metal averaging 108 joules, Haz averaging 125 joules with lateral expansion in inches between .039 and .059 depending on haz or weld. Base metal alone is considerably higher. SMAW is a little different but not much.

By js55

Date 07-13-2007 17:01

Gerald,
Thanks. The nickel. And quite frankly I'm surprised the nickel steel BM, even as Q & T is better than the 625 filler. Though I've never done cryo impacts on 625. I can see the lateral expansion of the weld being better than the HAZ with a Q & T though, if that is what you intended. If not, it doesn't make any sense to me. Your water Q and then T is essentially gone in the HAZ, essentially turning the Q & T into a normalize. Which might enhance the impact s'trength' over the nickel base filler, as your post indicates. Its all a bit confusing, but the numbers are certainly good enough in any case. Thats the bottom line.
But then, thats why you do the tests.
Very interseting. This one is going in my file. Thanks

By CWI555

Date 07-13-2007 18:57

Haz is on the higher end, weld at the lower end of values. Heat input is kept to a minimum due to the quenching and tempering of the base metal and other reasons.

By js55

Date 07-13-2007 19:21

Gerald,
I'd keep my heat input low the 625 anyway(perhaps not Q&T low but low nonetheless), considering the Nb. But with a Q & T you get the added benefit of shrinking the width of the HAZ. This not only shrinks the width of the CGHAZ and the non Q&T microstructure, but makes it more difficult with a V groove to have your Charpies entirely HAZ without a little contribution from the UBM (conveniently so if you don't tilt your specimens-something to my knowledge only ASME III requires-or used to).

By CWI555

Date 07-13-2007 16:37

should have asked, are you looking at the 304 or the nickel ?

By Sourdough (****)

Date 07-12-2007 16:57

I'm now very curious, being a native Alaskan.......where is your locale?

By dbigkahunna (****)

Date 07-13-2007 23:32

Correct me if I am wrong, but the service temperature will be -76 not the temperature you will be welding in.
Assuming this, the WPS and PQR's should identify the essential variables and materials. If you are being asked to recommend materials for this service temperature, I would say you are in over your head. You need a metallurgist for the recommendations and then Section IX will require impact (Charpy) test for the PQR. If you are the welder, ask to see the welding procedures. Also, -76 is not that low of a service temperature.

By js55

Date 07-16-2007 13:31

If I may be a little picky. Section IX doesn't requre impact testing. It codifies the requirements for testing once the testing has been required by other codes.