Bolt Fatigue at Elevated Temperatures

[PSSC]

I have a situation that brings up several questions for me.

It is basically a clevis with the cyclic load being applied to the pin.
But the added complication is that the assembly is at a constant 1700F

I had planned on using a rod (as the pin) with each end threaded so that no threads are in the shear planes.
I had considered requireing a preload and peening the threads to keep the joint solid so that the pin was not transferring the load.
But I can not guarantee that the bolt will not loosen and the force would then be transferred to the pin/bolt, so I need to make sure that if that does happen the pin/bolt can handle the load
It has been quite awhile since I did fatigue calcs in college.
But this goes even beyond that.
The basis of fatigue calcs is Ultimate Strength.
But at continuous high temps (1700F) I have always used creep strengths.

So question one would be which would you use at continous 1700F, Ultimate Strength at that temp or Creep?

Question two would be am I even approaching this in the best way?

I would not be surprised if there is a better solution out there, and would be glad to listen to any ideas.

Thanks

 

[TVP]

1700 F (925 C) is an extemely high temperature when it comes to designing with threaded fasteners, preloaded assemblies, etc. For example, ASME codes only allow 2 materials to be used at a maximum temperature of 1650 F: UNS N06002 (known commercially as Hastelloy X®, Nickelvac® HX, Nicrofer® 4722, Altemp® HX, Inconel® HX, Pyromet® Alloy 680) and UNS R30556 (HAYNES® 556® Alloy). The allowable stresses are not very high in this temperature range. Take a look at this previous thread for some additional information:

thread330-269903

 

[metengr]

Actually, your design approach would be to use an allowable stress value, similar to what the ASME Boiler and Pressure Vessel Code uses for materials, where both ultimate tensile and creep to cause x% maximum creep deformation are considered, and the lower value of the two becomes the controlling factor for safe design.

Your exposure to elevated temperature service at 1700 deg F will be a difficult design problem because of the limited choice of bolting materials that can survive at this temperature.

The only bolt material I have found that could possibly be used is N08810 per ASME B&PV Code application (pressure vessel application), which is permitted only up to 1650 deg F. The allowable stress for this material at 1650 deg F is 0.98 Ksi.

The information above only serves as a guide to give you some understanding in how to deal with a complex problem. Complex in the sense that you have time dependent properties to deal with to avoid either excessive creep deformation or failure from stress rupture. Keep in mind the environment will play a significant role because of corrosion concerns at this service temperature.

 

[TVP]

Just to be clear, I made a mistake on my previous post: there are 3 grades, not just 2, with max use temperature of 1650 F, with the 3rd being UNS N08810, as metengr correctly pointed out.

 

[PSSC]

Thank you both for your replies.

I did make a mistake on one part of the situation, the design temp is 1600F.

I did read the other thread.
I should have said that this part will not be on a pressure part.
This is for a retainer to hold a tube in place inside a fired heater.
The fired heater would have been built under API 560.

I had originally planned on using NO8330/RA330 for the pin.

@metengr, do you mean that I should find the fatigue strength per the usual procedure, then compare that to the creep strength and use the lower of the two?
If so, I assume you mean to use the Tensile strength at the design temperature and not the Tensile at room temp.
Is this correct?

Thanks to both of you.

 

[PSSC]

I think I may have answered my own question.
Reviewing my material from school, I reread the fatigue section in Shigley and Mischke Mechanical Engineering Design.
There is a "Temperature Factor kd" used.
The listed values for the coeffienct don't go to 1600F, but it does make a good case for using their procedure at elevated temps.

 

[metengr]

@metengr, do you mean that I should find the fatigue strength per the usual procedure, then compare that to the creep strength and use the lower of the two?

No. I really don't think you need to go this route. If you use allowable stress values as mentioned above, this has already been done in the form of allowable stress values. I was only trying to provide an alternate approach by using published allowable stress values instead of attempting to do this work on your own.

 

[cloa]

http://www.ami.ac.uk/courses/topics/0124_seom/index.html

This suggests your conditions are probably too difficult to analyse.