Heating and cooling rates for PWHT

Date 07-07-2008 11:44

the heating rate should be depended on the thickness of the plate, RT should be carried out after PWHT,PWHT can release the joint's stress.

By Bill M

Date 07-07-2008 12:59

Since you mentioned "SA516 Gr70" I am assuming ASME work. If it is Section VIII code work, see UCS-56 for more info on requirements for rates of heating and cooling.

By js55

Date 07-07-2008 13:42

Bill is right, 516 is pressure vessel grade material, but it is often used in other applications simply because the stuff is terrific. Its tough, ductile and clean. And this is the reason as well that RT after PWHT, if I may disagree, is not necessary. The stress relief cracking tendency of this stuff is pretty much nil. Even with long trm step stress relief for big vessels.

By js55

Date 07-07-2008 13:50

As for what happens in PWHT with that material there are primarily two phenomena. Stress relief that takes place through what is essentially a 'creep' mechanism involving weld regime stresses. This is responsible for the precipitous drop in tensile strength, and increase in toughness and ductility in the first minutes/hours of the stress relief. Then, (though it has been happening all along but now becomes predominant) is carbide precipitation. The more gentle slope of the stress releif curve is generated from this phenomena wherein more and more carbon becomes locked up in carbides and removed from solution. This precipitation stage is also what is responsible for some materials getting stronger and less ductile with additional time at temp. For example, often with V or Nb alloyed materials.

By Geese_howard (*)

Date 07-22-2008 04:01

Just to expand a little application codes examples discussed generally in this forum, british defense standards for ships propulsion shafts repairs, limits the heating rate to 50ºC/Hr or 5000/(shaft diameter in mm), whatever is less, and cooling rates to 62,5ºC/Hr or 6250/(shaft diameter in mm) whatever is less, both parameters only demanded over 300ºC, natural cooling is permited after 300ºC. (There is a whole world of welding works and regulations outside ASME IX BPV applications :) )

Obviously, the lower the rates, lower the stresses induced by thermal gradients, so lower heating rates and specially lower cooling rates are advisable, but
(This is a HUGE "but" i personally have about this), too slow cooling rates after PWHT at 550ºC or higher, may induce temper embrittlement in many alloyed quenched and tempered structural steels, or Cr-Mo piping steels, i think codes do not take this into consideration enough, by example setting minimum cooling rates, lets say using temper embrittlement susceptibility factors, such as Watanabe J-factor, or bruscato factor to determine the rates. Maybe somebody here knows of a code that take this into consideration, it might be interesting to have a look at that if exists.

By js55

Date 07-22-2008 13:34

If I may be allowed to ramble with some progressive food for thought. Those heating and cooling rates are exceedingly conservative for most alloys. And the most recent data is purposely demonstrating, for the possibility of code adjustment, that heat conduction through thickness is actually quite rapid. And since thermal homogenization is the thinking behind slow cooling rates the progressive thought is that homogenization is actually taking place at more rapid cooling rates than current thinking allows (once the thermal gradients are homogenized there is no more need to hold at that temp-or in other words heating and cooling rates should be engineered based upon actual thermal conductivity and not possibly antiquated, though admitedly effective, emperical judgement), and there are some who are arguing for more liberal heating and cooling rates even on thicker materials.
The slow cooling rate based upon thickness concept is old, and code bodies have justifiably been reluctant to change it since there seems, at least to my limited knowledge, to be few failures associated with rapid cooling rates in through thickness stresses. And they will, and should, not change without definitive data.
I have a tendency to look at it this way. Such slow cooling rates can require an up ramp during stress relief of 10/12/14 or more hours. We commonly had 24/26 hour cooks for Grade 91 with holds of only a couple hours.
But if through thickness stresses are such a deep concern let me ask theoretically how is that the much more extreme thermal gradients (much faster rates and much higher temps) of the welding regime cause so few problems in HAZ's and adjacent UBM. Not to mention that in stress relief much of the assembly is moving to an extent at the same time. Whereas with the welding regime most of the assembly is not moving at all. My though would be that there are much higher stresses at the HAZ UBM transition than there are in PWHT rapid heating rates. Granted, the HAZ/UBM transition is in the X axis and not the Z axis of through thickness gradients.

By Geese_howard (*)

Date 07-23-2008 00:42

I mostly agree with you, but there is some cases in which some "specials" problems may arise with too high heating or cooling rates, by example where materials with very different thermal expansion coefficients are joined, in metals of low thermal conductivity, and in cooling after PWHT of high alloy steels where retained austenite can be present even after the first PWHT (as in CA6NM) and transformation to martensite can be uneven in the thickness leading to stresses related to the volume change from phase transformation. The conservative thinking of the codes may also be related to the need to be more general in stablishing PWHT parameters that work for every case. But at least in carbon and low alloy steel , i think cooling , and specially heating rates might be higher without any harm.

By sunn625 (*)

Date 07-10-2008 09:07

maybe you can adopt VSR. It is unnecessary to install furnace and control temperature. My info:sunnn@semboo.com
It can relieve your welding stress.improve fatigue life and anti-deformation

By RonG

Date 07-10-2008 17:49

Boon, from my point of view all NDE should be completed prior to PWHT for the simple reason should repairs be required you have to perform the PWHT again after repairs.

After PWHT we only sometimes require MT inspection.

Yes heating /cooling rates are very important. Most every PWHT spec I have seen will not let you exceed 200 F'/hour up or down ramp above 600 F'. We seldom can go that fast up ramp due to configuration and mass and some of our specs require a slow down ramp but they are mostly higher Chrome alloy's.

The ramp rate heat source and gradient control be taken in to consideration. A large mass requires a lot of BTU's but some folks confuse BTU's with Temperature. Of course if you PWHT the entire work piece that point is moot. We work with a lot of finished machined equipment and are confined to some tight gradient requirements.

By Boon

Date 07-13-2008 11:19

When PWHT specify do not exceed 200 F'/hour for heating, what does that mean? Am I right to interpret that the temperature must not be higher than 200 F before one hour is up?
Is it a choice when writing WPS or must it depends on applicable code, to have PWHT on complete product or just the weld area? Should this be mentioned on WPS?
What is BTU?

By RonG

Date 07-14-2008 03:21 Edited 07-14-2008 03:24

Correct, or you can look at it as 50'F every 15 Min etc.

In simple terms you have a start temp which in most cases can be 600'F from there you star in to a controlled up ramp not exceed 200'F/Hr.

when you reach the PHWT temp called "Soak temp" the "Soak time" in most cases is 1 hour per inch of maximum cross section.

Then a controlled down ramp not exceed 200'F/hr to usually 600'F

"Is it a choice when writing WPS or must it depends on applicable code, to have PWHT on complete product or just the weld area? Should this be mentioned on WPS?"

That information generally comes from a PQR (Procedure Qualification Record). With out the PQR the WPS invalid. This is kind of like the old "Which came first, the Chicken or the Egg?". You can write the WPS first but you must qualify it thus it becomes a PQR from which you generate addition WPS's. Which ever, A WPS must reference a PQR. Got it? Good explain it to me please.

BTU is "British Thermal Unit" measurement of heat.

Hope this helps.

By Boon

Date 07-14-2008 16:04

WPS (preliminary) is written before procedure test is carried out and the PQR from the test is the reference for the WPS or additional WPSs.

But I believe both WPS and PQR refers only to the weld and test coupon and not explain whether should the complete product be subjected to PWHT.

Eg. A test coupon is likely to be within 16" x 16" whereas a welded pipe section can be be more than 30" in diameter and 48" in length.
For test, the PWHT is likely to be for the complete test coupon size. Should we interpret this to mean the complete production piece whatever the size must be subjected to PWHT?

By RonG

Date 07-14-2008 16:33

Naw just the area of the weld. A Post Weld Heat Treat for most purposes' is to relieve residual stress's created by the shrinking of the weld it is often referred to as "Stress Relief"

That is one of the reasons Gradient control is important to prevent creating more stress in different areas.

What is you back ground and what kind of work are you doing? if you don't mind my asking?

If you do or do not PWHT the coupon B/4 you bend or pull it, that information is called and "Essential Variable" and become a part of the PQR just like Volts/Amps, base material and Filler.

A PQR usually needs to be tested for properties like tensil strength and others like hardness. PQRs can can be used provide a great deal of Metallurgical information but for the most part all you need to get a WPS is the Essential Variable.

I am not a Welding Engineer by any stretch of the imagination just done a lot of PQR's, WPS's and PWHT's.. By done I mean physically carried out the work and recording the variables.

There are a great many minds here (much better than me) if you give a little more info about your self you maybe amazed at the level of expertise you can tap here.

By Boon

Date 07-15-2008 00:23

It depends on the fabrication code whether the complete product requires PWHT.
Under SA-672 section 5.3.1 it states "Classes 20, 21, 22 and 23 pipe shall be uniformly heated within the post-weld heat-treatment temperature range indicated in Table 2 for a minimum of 1 h/in. of thickness or 1 h, whichever is greater".

Can we interpret the "pipe shall be uniformly heated" to be referring only to the longitudinal weld on the pipe?

Part of my responsibility is to prepare WPS's, PQR's (through third party) and ensure our welders are certified for fabrication works.
The only problem is I have not done any with PWHT.

By RonG

Date 07-16-2008 01:08

If I am understanding your question. NO. "uniform heating" means +/- 25' F in most cases.

I don't what to get to deep in to the dynamics of PWHTing pipe. We don't dont do pipe, but suffice to say if you wrap electric heating pads around a pipe with insulation the heat will be uniform peripherally with a longitudinal gradient in both directions.

I really do not think any one expects to to PWHT the whole string.

Your PQR coupons would be done (PWHT) in an oven at the lab where it will be tested.

I do not believe you will need to PWHT the WQR coupons. Never seen that done. WQR's only to show the Welders ability to produce a sound weld not mechanical properties of the coupon.