Specified tightening torque vs ambient temperature

GA. My sentiments exactly.............and of course this is why, in expensive equipment, we tighten important bolts to a given "stretch," rather than a torque loading.

For those who may not know.

This stretch of the bolt may be measured electronically, by an dial indicator, micrometer, "go" "no-go" gauge or other appropriate method. The bolts may be tightened individually or as a group using hydraulic jacks on each stud (once the nuts have been put on the studs), all interconnected and using an hydraulic power pack.

As mentioned in the previous post, since all the components are at same temperature, no adjustment in the torque are usually necessary. And if the materials are similar, so will teh coeff of ther,mal expansions and if properly designed, no additional stress will be encountered at the joints.

The point is the proper design here, say for pipelines, the length of the pipe lines, which may run in 10s to 100s of meters, if not given sufficient flexibility will cause compressive stresses on the ends (joints) in summer and tensile in winters. Though this may not be a problem for you here.

The other portion is the material of fastener- these are usually not meant for sub-zero, especially unalloyed steels. Check whether this is meant for subzero operation.

Third is the thread lubricant, that should be rated for your temperature, else it will be more of an impediment than the lubricant.

Lastly, though theoritically the torque is highly inaccurate, but the industry depends on the torque tightening. The tensioning of the bolts is neither economically feasible nor practica in >90% of the cases. The necessary factor of safety is usually built in the design against the usual 10-15% pre-stress variation. In fact in our cases except a few very critical joints, all others are tightened through torque only.

Lastly, though theoritically the torque is highly inaccurate, but the industry depends on the torque tightening. The tensioning of the bolts is neither economically feasible nor practica in >90% of the cases.

That my friend depends on the industry in which you work, when you work in heavy engineering.......particularly on large slow rpm marine engines.......and I might add, it is also common place in most of your larger high speed diesels, as well as most medium speed diesels..........it is also very very unlikely that even "superman" would be able to achieve the required torque loadings required............and not least of all my friend, if you do not strictly abide by the manufacturers instructions, you could end up with a muiti-million dollar engine on your "slop chit."

The necessary factor of safety is usually built in the design against the usual 10-15% pre-stress variation

Just another point, on these large engines, 10-15% variance in torque loadings could cause you to continue to blow head gaskets, because of inconsistent tightening of individual cylinder head nuts............or more serious, and expensive consequences with big end bolts..........it could even cost someone their life.

I urge you not to be so flippant when making these statements.......someone may take you seriously.

We are OEMs making Turbines (steam upto 1000MW) and Gas turbines upto approx 350MW (single units). And our products do include different low and medium speed machines too (ranging from approx 20RPM with 3MW power consumed to a few hundred RPMs to of course GTs with 10KRPM)

The main point I made there is - you can not stretch all bolts- it is not feasible- I will really like to find a way to strecth the fasteners connecting the buckets to the shrouds where the approach itself is a bit difficult. And then these have to be with special mechanical locks to prevent loosening in these high speed machines. Even a metallic chip has once washed off a few compressor stages. In these we nefer try to take a risk with the stud and nut (which becomes mandatory in stretch on blind holes).

The critical ones- casing parting plane, Rotor and similar bolts are stretched due to the necessity. And in these it is not only the stretch load but also the strecth amount (length) is recorded.

But as I mentioned, it is not possible everywhere and that is the place where the factor of safety is built into our design. So though the posts here keep on insisting that the torque tightening is highly inaccurate- the end result is this is the indutry practice even at OEMs - except the critical joints.

These joints are clearly depicted in our written down procedures, drawings and the O&M manuals (all the three places).

Infact with all due regards I will like to one to seriously mull over the

% of fasteners Stretched (Precise prestress)

% of fasteners Torque tightened (semi precise).

% Of fasteners tightened by normal impact/non impact gun (least precise)- here you do set the torque value but too many degrees of freedom.

And that goes not only for the OEMs but also the users.

To make a wild guess it will be 10% vs 20-30% Vs 60-70% - And that includes the leaders - Let me exclude ourselves, i have visited quite a few world over- as well as I have watched the machines erected (capital goods) in our work shop, and that is the conclusion I can make.

Of course, some applications preclude the use of "more accurate" tightening processes. However, and surprisingly to some, this is moot...

You can still get highly accurate results with torque wrenches (calibrated or not), impact guns, bolt heating (resistance or induction), or even, get this: Bubba and a couple of his buddies hanging off of the end of a long "cheater bar".

For critical joints, using any of these options it's important to do what should always be done: Measure the result of the tightening step. Unfortunately, and almost beyond belief, this is rarely done.

If you do something (ie "torque" a bolt) and then don't measure it, how can you know what you've achieved? You can't. The best thing that you can do is guess, hope and pray that your effort has resulted in that which is required for the integrity and safety of the joint. Whenever a joint is "torqued' and the process stops, putting your faith in the unknown is exactly what you're doing.

How does one measure bolt load? One measures elongation.

Bolt elongation has a direct relationship to bolt load (as long as the fastener hasn't been tightened to beyond "yield"). Thus, by knowing the necessary elongation of a fastener at the required clamp load, one can easily "tune" the tightening process and eliminate all of the unknown variables.

• If the bolt has been "torqued", additional torque is applied (or released) until the fastener's elongation is within spec.
  
• If hammer and a flogging spanner was used, the hammer is swung again until the required elongation has been achieved.
  
• If a cheater bar is used, another crew member is hung off of its end until the nut is turned enough so that the fastener has stretched enough.

There are a number of ways to measure stretch. Some of these methods include the use of dial indicators, specialised ultrasonic elongation equipment or even the use of optical sensors which change colour upon correct elongation. Some provide much greater accuracy than others. In any case, any verification method is much safer than relying on "guessing and hoping"

Hi jjny,

Good answer from 'boltintegrity'!

You have specified in your question what a difference of 40 °C will make? IE: - 15/25 °C.

If you are talking of combustion engines, they run several tens or hundreds of times the 'ambient temperature' was when working on it.

The 'ideal' situation, assuming you can get or have 'spare' bolts, is to tighten until the bolt goes 'plastic'. That is usually done by 'feel' rather than a torque wrench though. BTW, When I say 'plastic', that is the state where the bolt will not get any tighter without breaking. It is already as tight as it can go. I admit this does sound 'hit & miss', but when working on lots of small (car) engines a torque setting was rarely known or used by me.

Given a production run or an engine (if that is what you refer to) which is inaccessible, in that it is too far away (in another country), or it is literally hard to get at, such as the 'engine' on a wind turbine, then you must use a torque. What I referred to was before any modern aluminum engines so the engine was not at great 'risk' of distortion.

If you refer to for instance the torque of pipe bolts in a Refinery or similar, then again the torque would need to be correct. Oftentimes 'in the field' you will have to strip something down and repair it and will not have the use of a torque wrench.

These are just thoughts from someone who has stripped and repaired several hundred if not thousands of engines and other machinery. You just do not always have the tools and have to 'make do'. I often worked on very heavy and very large Machines, where the machines were so old it was not possible to know a torque setting, and these machines were always repaired most of the time, using brand new bolts and or whatever tightening type of devices were designed for that particular machine.

I agree this is not 'accurate' in anyway! But sometimes you just have to work with the tools you have and in my case they often did not include wrenches large enough so I made my own sometimes................. It is called 'making do', but in real life situations this is more common than worrying about any torque settings. B-U-T, that is not to say a torque setting should not be used if you know the setting and have a wrench! I remember there was talk on CR4 of how to adjust to the correct torque even if you do not have a "Torque wrench".

Below is a CR4 thread and other sites which explain how to 'torque' a 'bolt' with no torque wrench. See # 2.

1. Torque wrench - Wikipedia, the free encyclopediaA torque wrench is used where the tightness of screws and bolts is crucial. ... When no torque is applied to the lever arm the indicator rests parallel to ... an electronic torque wrench but the tightening angle from the snug point or ...
   en.wikipedia.org/wiki/Torque_wrench - Cached - Similar -
2. CR4 - Thread: Tightening a Bolt - Without a Torque Wrench17 posts - 12 authors - Last post: 17 Feb 2009dear,can any help me to tight a bolt up to 200n/m without a torque wrench.I want to calculate the force per degree suppose abolt is tighten ...
   cr4.globalspec.com/.../Tightening-a-Bolt-Without-a-Torque-Wrench - Cached - Similar -
3. Methods of Tightening Threaded FastenersThis affects the conversion of the tightening torque into bolt preload and should ... More unusually, quote that a calibrated torque wrench is to be used to check the ... whereas the gauge pin does not move since it experiences no load. ...
   www.boltscience.com/pages/tighten.htm - Cached - Similar -
4. BoltsThis minimizes the chance the bolt will break because there is no slack ... Take the time to learn, invest in a torque wrench, and do it right. ... Don't use a wrench to tighten bolts or nuts that are meant to be finger tightened only. ...
   www.virginiawind.com/tips/bolts.asp - Cached - Similar -
5. Best Way - Torque Tool Use Guidelines | eHow.comBest Way : Torque Tool Use Guidelines. If you've ever stripped a bolt while ... If working with different metals, particularly aluminum engine casings, ...
   www.ehow.com/way_5602885_torque-tool-use-guidelines.html - United States - Cached

I hope my 'musings' are at least 'entertaining' and at best useful!

Take care and Happy New Year!

CR4 Admin: removed link to virginiawind.com -- site is about semiprecious gemstones

Thank you very much, people. We are assembling a diesel generator after engine overhaul. Conditions here are not ideal but with your suggestions and responses you have confirmed that my hunch to proceed was right.

Just that the scope of work was not well-defined from the outset. We have had to feel our way and are now on the assembling part (if all the parts get here).

Once again, thank you.

I don't wish to beat a dead unicorn but, I think that it's very important to provide my few farthings on the comments regarding the importance of using "calibrated" torque wrenches (or the "next best thing").

Many authorities specify that torque wrenches must be calibrated before use on critical joints. Furthermore, if a hydraulic wrench is to be used, the pressure gauge must also carry a current calibration seal. The "better" torque wrenches usually have no problem maintaining the required +,- 2.5% accuracy.

Let's step back, though, and look at what this really means:

A calibrated torque wrench results in its actual torque output (from the application of manual, electrical, pneumatic, hydraulic or other force) being within plus or minus a documented percentage of its indicated torque. Important, right? Of course it is, especially when you need to ensure that you've got proper bolt load lest the joint fail in service. However, the use of calibrated torque wrenches provides a dangerous false sense-of-security ...

Joint reliability doesn't depend on "torque", it depends on clamp load!

A "properly torqued" bolt (tightened by a calibrated torque wrench, even) can, in fact, be completely loose. A similar bolt, tightened to the same torque, with the same calibrated tool, may be tightened to well beyond a safe level. How can this be? Further comments are here and here.

Think about it: You demand that the torque wrenches have an accuracy of +, - 2.5 % yet the inaccuracy of the process can be +, - 20, 30, 40, 80 or even over 100%! Hmmm... looks like the trees are in the way of the forest