Differential thermal expansion between stainless steel flange And low alloy steel bolts 6

[Ehiman1]

Hello, I have a heat exchanger flange in stainless steel with bolts in low alloy steel at high temperature (400 C) and camprofile gasket in stainless steel.
The code is ASME VIII div1.
We checked the flange with appendix 2 and then with ASME PCC-1 appendix O to check the joint against leakage problems.
Is it possible without FEA calculation to evaluate the impact of differential thermal expansion between bolts and flange as well?

Thank you

 

[TGS4]

What will be the temperature of the bolts? I think that such an assessment may be possible without FEA.

 

[Christine74]

What specifically is your concern? Since the bolts are exposed to the atmosphere they will operate at a lower temperature than the flange, plus austenitic stainless steel has a higher thermal expansion coefficient than the low-alloy bolts so the joint will get tighter, not looser, due to the differential thermal expansion.

Is your flange not an austenitic grade of stainless?



-Christine

 

[Ehiman1]

Unfortunately we don’t know the bolts temperature without a thermal analysis and this is the first doubt.

My concern is not about the bolts because SA-193 B7 has an important yield/tensile strength, so they can absorb an additional load due to thermal expansion.
To me the problem is that this additional thermal load could create an higher stress in flange and potentially an higher flange thickness (SS 304).

I thought to calculate the bolt stress due to thermal expansion with worst case possible (inside operating temperature vs ambient temperature), add this load to design bolt load and verify the flange with this new force.

It is very conservative but it could be helpful, what do you think?

 

[SnTMan]

This is very common combination of materials/construction. If your flanges are adequately designed (as it appears they are) I'd not worry about them. Unless the bolting is tightened very near yield, I'd not worry about them either.

You could do a simple analysis based only on lengths and temperatures or a (quite) complex one to include stiffnesses.

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[Christine74]

B7 stud bolts are commonly torqued to 70-80% of yield which is around 80,000 psi, but flanges designed per Appendix 2 are only designed for bolt loads of 25,000 psi. Yet this rarely (if ever) causes damage to the flanges.

I don't expect any issues here but if you're concerned you can always insulate the bolts.

-Christine.

 

[davefitz]

One can use belleville washers ( spring loaded washer) to account for differential expansion betweenm the ferritic bolts and the austenitic flanges.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[MJCronin]

I strongly agree with some of the comments above...

I also believe that long industry experience with this specific combination of flange bolting is important here

SnTman: This is very common combination of materials/construction. If your flanges are adequately designed (as it appears they are) I'd not worry about them. Unless the bolting is tightened very near yield, I'd not worry about them either.

AGREED

Christine74: B7 stud bolts are commonly torqued to 70-80% of yield which is around 80,000 psi, but flanges designed per Appendix 2 are only designed for bolt loads of 25,000 psi. Yet this rarely (if ever) causes damage to the flanges. I don't expect any issues here but if you're concerned you can always insulate the bolts.

AGREED - I have seen the B7s torqued to 80ksi on stainless flanges and in satisfactory service for many years. Insulation of bolts causes long term maintenanceaccess issues ... IMHO

davefitz:One can use belleville washers ( spring loaded washer) to account for differential expansion between the ferritic bolts and the austenitic flanges.

DISAGREE: I believe that, as a general policy, this will cause more problems than this is worth. There will be clearance issues and new maintenance assembly requirements. Extra long stud bolts, flat washers will cause problems. There is a near infinitesimal difference in actual expansion between the SS material and the B7. Stick with the tried and true combination ..

Again, all of the above in based on my experience and opinion






MJCronin
Sr. Process Engineer

 

[MJCronin]

This topic has been discussed many times on "eng-tips" in the past ...

https://www.eng-tips.com/viewthread.cfm?qid=269577

https://www.eng-tips.com/viewthread.cfm?qid=240354

https://www.eng-tips.com/viewthread.cfm?qid=70790

https://www.eng-tips.com/viewthread.cfm?qid=350704

The final acceptability of B7 bolting materials for stainless flanges also depends on the bolt coating selected by the engineer and the flanged joint environment

MJCronin
Sr. Process Engineer

 

[desertfox]

Hi Ehiman1

Yes an analysis of the thermal expansion is possible without fea. I would tackle it by assuming that the bolts don’t see a rise in temperature although in practice they will. So you need to start with the preload of the bolts ie the 70% or 80% of bolt strength, then calculate the expansion of the joint, next step is determine the external load due to the joint expansion as though the bolt preload was fully restraining said expansion. The external load from the joint expansion should be shared out between the bolts and the joint materials based on the stiffness of the bolts and flange materials.
Getting the stiffness value of the bolts is fairly straightforward and the more difficult bit is the flange and gasket stiffnesses, however a rule of thumb is to take a surface area of the flanges to be about 2.5 times bolt diameter and workout the stiffness of the joint using that area and the flange thicknesses already known.

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

 

[EdStainless]

The only thing that I have ever seen someone do in this case is to use larger, heavier washers than is typical.
Just to assure good load distribution.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[3DDave]

Performing a non-FEA analysis assumes the temperature is constant throughout both the flange and the bolts and that the temperature of each is known.

Neither assumption is usually true - the stress due solely to coefficient of thermal expansion is likely on the order of 1 psi/degree Fahrenheit.

 

[BJI]

The temperature rate of change is most important here, particularly with stainless steel. The interaction between axial expansion and flange rotation (relative radial expansion) can cause high increase in bolt load during start-up transients. More caution would be warranted when going to higher bolt loads in these cases.

You could follow the procedures in WRC 510 to calculate temperatures, and also the mechanical interaction with temperature.

 

[MJCronin]

A description of WRC 510 bulletin and instructions on how to purchase can be found here:

https://www.normadoc.com/english/wrc-510-2544062.html

DESCRIPTION OF DOCUMENT:

Thermal events and transient thermal effects are known to play a major role in pressurized flanged joint leakage. When a leak occurs the engineering challenge is to understand and properly diagnose the role temperature and thermal transients in that failure and thereby specify measures necessary avoid future leaks. Also, since most pressure vessel codes require the consideration of thermal effects without providing the methodology, perhaps the greater engineering challenge is to the flanged joint designer. This is not only to determine temperature effects on flanged joint designs to comply with code requirements but also an evaluation of the design to assure leak free operation considering anticipated thermal events. This Bulletin provides a set of analytical tools and guidelines for addressing these challenges.

Just a quick note to the Third World about purchasing technical documents:

When you "pdf" a technical document and share it with everyone else on the internet, you effectively "kill the golden goose" (look it up) for any future Pressure Vessel Research and Education.

The only reason this seminal bulletin WRC 510 exists, is because an institution funded someone who decided to devote an important part of their career to do the work !

WRC Bulletins must be purchased for the whole system to work .... as you probably already know

When thieves refuse to pay, the system dies ....

My opinion only ...



MJCronin
Sr. Process Engineer

 

[GD2]

Ehiman 1,
Do you have an actual leak or simply wants to do a leak verification check for the HX?
If an actual leak, which end is the leak - Fixed end or Floating end?

GDD
Canada