Larson Miller Parameter 1

[roniprabowo]

Dear all,
Has anybody hear about Larson Miller hardness parameter for Cr-Mo material. This is the guidance for PWHT in reducing hardness to void any crack.

 

[davefitz]

I have heard of using the LMP for creep life, and creep life is associated with hardness test results , but I never heard of using LRP for hardness.

There are some issues related to proper PWHT for newer alloys, such as P91. If you PWHT at at too high a temp or for too long, you overtemper the alloy and lose creep life, but gain notch toughness and room temp brittleness subsides. This can be detected by a hardness traverse after cooldown. The PWHT temp and time is also related to the Nickel content of the filler wire.

Also complicating the hardness issue is the contradiction between the ASME code apparent requirement ( B31.1) to proceed to PWHT immediately following weld completion , and the need for P91 to cooldown to below 200 F prior to PWHT to ensure complete transformation to martensite. Such a practice without intermediate cooldown will lead to a brittle weld, unless you PWHT twice.

Even worse is the common practice of cooling down to room temp prior to PWHT for purpose of X-rya- the risk of cold cracking is high if left cold for 2-3 shifts prior to PWHT. And to top it all off, nearly noone tests for hardness to confirm it is within acceptable limits.

It is my guess that there will be a lot of work to be had by life extension experts in 5 years after the first crop of P91 pipelines start to exhibit failures.

 

[metengr]

Yes. Except the Larson Miller parameter you may be referring to is called a "tempering parameter", and is used to determine the hardness of a metal based on tempering temperature and time during heat treatment. This has been applied for predicting the hardness of welds during post weld heat treatment.

The parameter was developed based on a plot of hardness versus tempering temperature and time. The hardness change of a metal during tempering follows a typical Arrehnius equation (reaction rate) or plot because of changes in carbide size related to time at temperature exposure. The tempering parameter was developed under the same concept as the Larson Miller parameter for predicting creep life in metals based on time and temperature exposure, except stress is plotted instead of hardness.

 

[davefitz]

For hardness, it seems to be called the Hollomon-Jaffee Parameter, P

P=T(18+log(t))*E-3
T= deg K, t= hrs

 

[davefitz]

http://www.industrialheating.com/CD...rstory/BNPCoverStoryItem/0,2830,99587,00.html

The attached link provides a good primer on the use of both the Larson Miller Parameter and the Hollomon Jaffee parameter.

 

[Metalguy]

If only things were so simple. There is at least one type of situation that seems to defy the time/temp tradeoff for PWHT, at least where resistance to elevated temp service (~500+ deg F) corrosion fatigue/SCC is concerned. The steels involved were SA-533, type B class 1 and 2. ASME Sect. III code used to allow temps. below 1100 deg F if longer times were involved to get the same hardness. But many pressure vessels given this PWHT developed HAZ cracks, while those PWHT'd at 1100 or higher did not.

Lab. testing using service conditions verified that specimens PWHT'd at 1100 or higher did not crack, while those held for longer times at slightly lower temps. did.

I'm not sure why-I'll have to find some time and go reread the EPRI report.

 

[metengr]

Metalguy;
You are right, this is not so simple in the real world. In my past life I was involved with a nuclear steam generator (SG) corrosion fatigue cracking (as some would say environmental assisted cracking - EAC) problem in the heat affected zone of the carbon steel shell to cone transition weld. It turns out the manufacturer of the steam generator used the longer time at lower temperature option to meet the ASME code requirements for PWHT to avoid damaging the internal tubes. We had shown from a sample removed that the actual hardness of the HAZ was not reduced below the threshold for EAC and resulted in a preferred site for crack initiation in service. If the PWHT could have been performed at the higher temperature, this would have reduced susceptibility to EAC in service.

The lower temperature extended time approach for PWHT did not reduce the hardness of the HAZ to the level that one would have anticipated assuming the time/temperature interaction. The reason for this I believe is that the extended time durations referenced in the Code are still too short at lower temperatures to have a direct affect on carbide morphology for some materials. The major benefit of extended time at lower temperature only served to reduce residual stresses (which is still very benefical).

 

[Metalguy]

Aha, so it's simply a HAZ hardness/microstructure thing. I am presently involved in Italy building 6 BIG nuke steam generators-~700 MWe each, with 12,500+ tubes ea. I specified a min. PWHT temp. of 1125 deg F to be sure we actually got above 1100. Seems that the total *possible* error stackup is ~21 deg F., so I've got a worst-case cushion of 4 deg!