Bolt yield and torque 1

[PressuresOn]

Hi,

I am in a confusing predicament with bolt torques and bolt yield assessments. We are reviewing a new software for gasket assessment of seating stresses etc. It is quite good in that it enables us to insert different bolt yields from test certificates and then assess the effective seating stress for the gasket based on different torque values.

However, a senior engineer in another department has rubbished the software as he insists that the ability of a bolt to stretch under torque is purely dependant upon the modulus of elasticity and not the physical yield as shown on the material test certificate.

In summary, I need to assess if my thinking is wrong. I have, up until this moment, worked on the basis that a bolt with a higher yield will enable a higher torque to be delivered and thereby deliver a greater seating load. However, if i consider the other engineers assessment then yield should not have any part to play.

Any assistance or help that folks could offer would be of great help.

Many thanks

 

[trottiey]

You are correct in that a bolt with a higher yield will enable a higher torque and seating load before yielding, all else being equal. But a pressure vessel design should be limited to the allowable stresses given by code, regardless of how much stronger a given batch of material is. So the bolt yield from the test certificate should not matter.

 

[r6155]

Torque is an indirect/vague/uncertain method.

Use stretching method to stress the bolts ( yield stress and modulus of elasticity are included in the formulae )

Regards

r6155

 

[TGS4]

But a pressure vessel design should be limited to the allowable stresses given by code, regardless of how much stronger a given batch of material is.

That is true for the pressure boundary material, but NOT in the context of bolts. Refer to ASME Section VIII, Division 1, Appendix S, as well as ASME PCC-1. Typical target assembly bolt stress will almost always exceed the allowable.

PressuresOn - I suspect that your "software" is performing some calculations that would set the bolt torque based on a fraction of yield at the operating condition - I typically use 90% of yield. In that case, the maximum amount of torque that you could apply to the bolt is indeed a function of the yield. (Whether you use mill cert yield or minimum-specified yield is a topic in and of itself).

The other engineer at your work is correct that the ability of the bolt to stretch (otherwise known as Young's Modulus) is independent of yield. However, the elastic interaction of the entire bolted flange joint assembly is only one part of the equation. The maximum amount of assembly bolt torque is another, and that is likely where the yield strength comes in.


Be very very careful of using software that is a black box - one where you don't know what's going on in the inside. If you don't understand the calculation process, then you could have a garbage in - garbage out situation. It also makes it extremely easy for someone to "rubbish" the software. If the software documentation doesn't explicitly spell out what calculations they are doing, get the software vendor to explain it in painful detail. Ask lots of "why" questions.

 

[PressuresOn]

Thank you for the replies.

All of the components are desinged using code stresses and code formulae. This is the matter of practical torque application and elevated stresses at hydrotest. We do utilise bolt tensioning where applicable but in this instance we have a client who insists on torque application. So in summary, I need to move away from what would be the ideal or perfect solution and simply ask for assistance on;

1) Calculating bolt extension for different torques and yields
2) Ascertaining how yield and modulus of elasticity play off each other when determining bolt extension
3) Getting a basic understanding of why it is possible to apply a higher torque in a bolt if it has an increased yield value above code minimums

Sorry if it seems a lot but I am really struggling with this concept and it is starting to undermine my confidence.

Thank you

 

[PressuresOn]

Hi,

Sorry TGS4. I was writing the previous reply before I spotted your post.

The program does employ a formula for calculating bolt torque based on yield value. We calculate and assess all components to the design code. However, we then perform other auxillary calculations that are not code requirements but need to be performed so that we attain the best chance of gasket seating on site. In this instance, we employ the yield directly from the cert and this has lead to a much improved life of the gaskets in operation over the last 4 or 5 years.

We cannot change the design of the equipment as it is standardised and client controlled but we can be imaginative and innovative to ensure we get the best results. I understand that the Youngs Modulus is a factor in the ability of the bolt to stretch. I guess I simply wanted to try and ascertain why our methodology works when there is no clear formulae that I can employ to prove it.

Thank you

 

[TGS4]

Please refer to ASME PCC-1. High assembly bolt stress (whether obtained from torque, extension, or some other methods) is essential to the safe operation of a bolted flange joint. The only catch is to ensure that, during operation, you don't cause your bolts to yield. Once you do, you lose elastic behaviour, and all of the calculations, and the assumptions thereof go out the window.

The other aspect, that I failed to mention, is that you have to ensure that your high bolt load does not crush your gasket - if you have a gasket that is susceptible to crushing. For a typical spiral wound gasket with a solid metal centering ring (and even solid metal inner ring) it is very difficult to crush them. Other gaskets (CMGC or jacketed gaskets for example) may crush, and so you have a balancing act of making sure that you have sufficient bolt load to ensure a seal, but not too much to crush the gasket so that it does not recover upon unloading.

Bolted flange joints are very complicated - so don't get discouraged. Some of the best minds have been working on this topic for decades, and we still struggle with understanding. If you like long and complicated math, I recommend reading EN-1591. It's a good approach to the elastic interaction that goes on in a bolted flange joint.

Oh - and don't forget about external loads (forces and bending moments).

 

[PressuresOn]

Thank you for the reply.

The gasket programme ensures that our effective gasket stress remains somewhere in the safe region - between the recommended and maximum stress.

All of the bending moments and forces are calculated and included in our submissions to the Notified Bodies.

Thank you for your help.

 

[trottiey]

I mostly agree with TGS4, and I'm aware of Appendix S. My answer was specific to design work. I recognize that actual assembly conditions may vary, and I try to stay out of the way when that happens. Note that Appendix S "will be important only when some unusual feature exists."

I maintain that any bolt that actually yields during assembly should be replaced prior to service. That yield point places an extreme limit on bolt extension. Below that limit, bolt extension decreases almost linearly with torque. I hope that helps with your three questions above.

 

[lganga]

You can read Pressure Vessel Design Manual 2nd edition ¨Bolt torque for sealing flanges"Dennis Moss.
I calculate them as 60 to 70% of yield bolt stress
Also you can get information from the fabricator of hydraulic torque devices.

 

[unclesyd]

I hope this program is flange specific in that it takes in all parameters of flange.

Nearly all our bolting is calculated based on getting a certain bolt stress, usually 50,000 to 60,000 psi.