Cookies may be used if you click links or banners to other commercial websites from this site. No cookies are used directly from this website.

For more details, read my Privacy Policy.


Bolted Joints

Two components bolted together rely upon the friction between their respective surfaces. The level of friction depends upon the co-efficient of friction of the combination of surfaces, and the clamping force of the bolt or stud. Except in special cases, bolts, screws and studs should only be loaded in tension. This tensile load is what we are interested in maintaining.

To keep the joint tight without any special locking devices, we rely upon friction under the bolt head/nut plus the friction in the thread. The effect of the thread friction is partly governed by the threadform. On page one, I showed that many threads have 60 degress between flanks (55 deg. for Whitworth). The larger the angle, the greater the impact the thread friction has in keeping the joint tight. Thread with very small angles (either square threads, or buttress threads) are used for machine gearing systems where low friction is necessary.

Tightening up the joint in the first place, is a more complex challenge to appreciate. Having an accurately calibrated torque wrench won't solve this problem. If you apply a given amount of torque when tightening up a bolt, you will not know how much stress the bolt is experiencing. One can do a partial calculation based upon the pitch of the thread. That will reveal how much tensile force is applied to the bolt from a given amount of torque from a wrench, but without allowing for friction. Unfortunately, friction is considerable at the best of times! Divide the force by the cross-sectional area of the thread core, and this gives nominal stress.

But how to estimate the level of friction? Typical figures would be about 50% of the tightening torque absorbed by friction under the bolt head, and about 40% by the thread friction. That would leave 10% for tensile stress on the bolt, or a 10% tightening efficiency. Even more unfortunate, is that the friction level can vary widely. So the efficiency can vary between 5% and 15%. Clearly, 15% would give 3 times the tensile stress compared to 5%, so some guidance is needed to try and establish what torque setting is required for a given set of fasteners.

Efficiency figures much above 10% are likely to be achieved with high grade fasteners, cadmium plated, using a good thread lubricant. Commercial fasteners, unplated and lubricated with engine oil, might achieve approx. 8-10%. In practice, care must be taken to ensure that all threads are in good condition, clean, and lubricated in the same way for each fastener in the joint. If it is possible to measure the stretch of a bolt, then that can provide a more precise measurement of stress.

Shearing While Tightening

More friction in the thread will reduce the tensile stress in the bolt, for a given torque wrench setting. However, it will also increase the torque applied to the bolt while tightening is in progress. This twisting applies a shear stress to the bolt, which if high enough, will cause the bolt to fail. Therefore, good lubrication is recommended, especially if the bolt material/plating has a high co-efficient of friction. However, whatever torque settings are specified (for your big-end bolts, cylinder-head nuts, etc.), should be considered as appropriate as long as the original specification fasteners are used - and lubricated with engine oil for engine fasteners.


Studs should not be screwed in right up to the end of their thread. This would place undue stress at the thread end, which is a potential weak point to start with (depending upon stud design, but the AC's stud designs are rather outdated!). Studs should also not be tightened hard into their holes. There are suitable thread adhesives available. Stud holes in softer materials, such as aluminium alloy, should be coarse. Even better, is to have thread inserts, so if you have stripped threads, then this problem can be turned to advantage.

I've heard of people making cylinder-head studs that were case-hardened! A recipe for disaster if ever there was one! The high tensile alloy steels need for joints such as cylinder-heads, are not suitable for surface hardening, and do not require it anyway. Be very careful when having new studs made. They must be of the correct material, correctly heat treated, and if plated, this must also be meticulously done.


For important highly stressed joints, washers are best avoided if this is possible. They cause torque-wrench readings to be unreliable. If they are not hardened washers, then the nut/bolt head may sink in and thus reduce some of the pre-stress in the bolt/stud. Tab washers are the worst in this respect, and should not be employed on high stress joints.

Nuts (choosing)

Old practice (in the UK at least) was to specify a lower grade of steel for the nut, relative to the bolt or stud. So, mild steel nuts grade K (later designated grade A) were used with bolts up to HT grade D (later designated grade R). Grade L nuts (later grade P) were used with grade E and F bolts (later grade T and U). The reasoning for this was partly that the stresses in the nut (apart from the thread) are lower than in the bolt. Also, the threads in the nut would yield first, permitting even load distribution along thread engagement.

Modern practice is to match the strength grade of the nut to the bolt, such as a grade 8 (metric) nut to a grade 8.8 bolt. The modern reasoning is that a nut thread might begin to strip when tightened, and this may go un-noticed. With higher strength nuts, the weakest point during tighening is likely to be the stud, which if it breaks, will certainly be noticed!

<< Page 6 ****** Page 8 >>

Page 1 - Screw Threads

Page 2 - Bolts, Screws and Studs

Page 3 - Nuts

Page 4 - Nuts (Self-Locking)

Page 5 - Washers

Page 6 - Materials and Strength Grades

Page 8 - Wood Screws and Nails

Page 9 - Fasteners for the AC 2 Litre Saloon



Website started 29th December 2006