pretorque required for a friction joint (shear load) when exposed to temperature change 1

[jedstress]

Hi all,

I am currently working on a problem which involves exposing a bolted joint to an increase in temperature (delta 250 C).

In the past I have dealt with typical pretorqued bolt joints (ambient temperature) where the required pretorque was approximated by working out the actual required joint clamp load based upon the external applied axial and shear force (friction coefficient usage). The min required pretorque was approximated based upon this minimum required clamp load to avoid joint slippage and separation (using machinery handbook text etc). As long as the external forces did not exceed the pretorque induced joint preload, the joint was assumed to have no slippage/separation.

However, I am struggling to work out how to account for the impact of thermal expansion in the above scenario. in the ambient temperature scenario, the external force has very small impact on the bolt preload (due to relief of joint compression...). But, in the case for a joint where the coefficient of thermal expansions for the plates and the bolt differ, and assuming that a bolt load due to thermal effect has been calculated, does this value need to be added to the bolt preload due to pretorque? Maybe I'm overthinking it and maybe that preload plus thermal load needs to be assessed against the bolt yield allowable and thats it.

Im trying to make sure that bolt strength isn't exceeded whilst also making sure that the bolted friction joint doesnt slip. For slippage to occur, I assume that the thermal expansion of the bolt would have to exceed the joint expansion? And to avoid that scenario, the bolt would have to be pretorqued enough such that the preload is greater than the required joint clamp load (friction joint) after deducting the relieving thermal load?

My apologies if I have totally misunderstood how to account for joint thermal loads. I would much appreciate it if someone could point me in the right direction regarding calculating the required pretorque. Please note that I am not too concerned with evaluating precise values for pretorque; I don't mind a simplified conservative approach.

Many thanks in advance

 

[desertfox]

Hi Jedstress
Basically you need to size the bolt for the required preload and ensure that the temperature rise of the joint doesn’t either yield the bolt or increase the compressive stress on the clamped parts so that the nut/bolt head embeds into said parts.

This was a failure of two dissimilar bolted materials subjected to a temperature difference of -270 degrees Centigrade

Thread404-477333.

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

 

[jedstress]

Hi desertfox,

Thanks for the response.
The two flanges forming the bolt joint will be made from some carbon steel (one will be S355, other is unknown right now but a carbon steel). The bolts will most likely be grade S. So the coefficient of thermal expansion amongst the bolted joint hopefully isn't too different (unlike the issue they had with failed bolts in the thread you mentioned). If possible, have you got some literature which explains how to calculate bolt load due to temperature change? I have come across some websites but wasn't too sure about the validity of their methods.

Regarding the post you mentioned above, I never read the entire thread but it seem like the bolts were failing in shear. The post author mentioned that the bolts were supposedly ductile and failed in tension yet there was no evidence of necking. I am no expert but just seem like maybe the bolts failed due to shear resulting from thermal contraction. From a design point of view, shouldn't the author incorporate bolt joint clearance holes to allow for such movement? The large shear forces arising from thermal contraction would have easily overcome the bolt preload and friction and just slipped. I will read the entire thread to see what the final outcome was.

The bolt joint I am looking at will have clearance hole in one of the flange to allow for such thermal movement thus negating any additional shear. In my case, an FE model has also been created to measure the thermal displacement hence allowing me to determine the minimum bolt hole clearance.

 

[desertfox]

Hi jedstress

It was thought that the failure was by shear originally however as the thread developed it turns out the the bolts were essentially brittle and tensile loading on the bolts was responsible for the failures due to the loss of all bolt preload and the external force of the solenoids loading all the bolts in tension.

I think I can help further and will post later.

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

 

[desertfox]

Hi Jedstress
This link might help

https://becht.com/becht-blog/entry/differential-thermal-expansion-a-challenge-to-flange-joints/

[pre][/pre]

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

 

[jedstress]

Hi desertfox,

Thanks for the link. For my own knowledge, wouldn't The bolt area and flange local area have an impact on the thermal induced bolt stress i.e. the stiffness of the various parts? The link uses only deltaT, thermal coefficient difference and bolts youngs modulus to calculate the bolt stress. My apologies in advance if I am missing the obvious.

 

[jedstress]

Hi Desertfox,

Slightly off-topic question, have you ever used BS13155:2020 for the assessment of lifting beams? I've added a separate post regarding this British Standard and was wondering if you had ever implemented it. Hopefully I will get some replies on that thread soon. You have been really helpful on this post hence why I asked the question regarding BS13155.

 

[desertfox]

Hi Jedstress

In response to the bolt stretch, think of it like this ; if the materials of flange and bolt are the same then the length increase in the bolt will be the same as the increase in thickness of the flanges, areas don’t come into it. Now Let’s say that the bolt material now as a smaller coefficient of expansion than the flanges, now the flanges cannot reach there free expansion point because they will be restrained by the bolt, also the bolt will exceed its free expansion length because the flanges will drive it further, this means the restricted compressive displacement in the flanges will equal the axial displacement (growth) in the bolt because equilibrium will need to be maintained.

I haven’t used the BS you quoted but have designed a lot of lifting rigs, I will take a look at your post

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

 

[jedstress]

Desertfox, many thanks for clarifying the thermal bolt load query. You have definitely helped me better understand the task within a very short space of time. So thanks once again for putting up with me .
Yes I would appreciate it if you could look into my other thread too.

 

[desertfox]

Hi Jedstress

You are very welcome

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