Is this true..?

Many moons ago an old engineer told me that when fixing with screws/bolts into a blind tapped hole there is no need for the screw/bolt to be any longer than 2 x diameter. So for instance if you used M6 screws then the threaded part of the screw need be no longer than 12mm, he said any longer gives no benefit.

I was just sorting through some screws for the loco chassis and this sprung to mind.

Ron

True

Sort of! The theory is that as the bolt is tightened it tends to stretch slightly, which slightly increases the pitch of the thread. So only two or three threads are actually carrying the load. Hence increasing the depth of thread does not increase the load carrying capacity. From a practical viewpoint times two is reasonable. It's tricky to tap shallow blind holes, so best avoided if possible.

Andrew

I'm guessing it refers to the thread length where the accumulated shear strength of the thread interferences equals or exceeds the tensile strength of the core - so that if you overtighten you'll break off the head rather than strip the thread?

If so it can only be a rule of thumb and may vary with material and pitch/depth of thread.

During my monkey-days with a motorbike in the 70s, I noticed that I always seemed to be stripping BSF bolts, but breaking the heads off BSWs... 

Edited By Mick B1 on 29/06/2018 10:57:04

I suppose nuts would be made a lot thicker if more thread engagement gave a significant increase in holding capacity.

+1 to Andrew's 'sort of'!

As a rule of thumb, it's not unreasonable. But there are other factors that make a difference, in particular the fit of the thread. A loose fit won't be as strong as a close fit.

Where strength matters, 'Old Engineer' is sent to make tea while Stress Engineer is consulted. Stress Engineer can do the sums. Taking into account the strength of the materials carrying the external and internal thread, he can calculate the minimum number of threads at 100% engagement needed to develop maximum strength. It's not many, see half-sized nuts. If max load is exceeded he knows if the bolt will break before the internal thread strips , or vice versa. More important, he knows that cutting perfect threads is difficult, expensive, and slows down assembly. Even though they are not full strength, it is usually more convenient to make threads to be a loose fit, perhaps achieving less than 50% of maximum in the case of a cheap general purpose fastener, like those holding my lathe on its stand. Stress Engineer will balance the need for strength against cost.

Considerably more care is taken specifying and making threads for safety critical applications. Aircraft are interesting because they also need to minimise weight, leading to components being more stressed than in ordinary applications. Many aircraft have crashed because the wrong bolt was substituted during maintenance.

I'd guess that on a typical model steam-engine, and certainly most of my crude efforts, a 2x diameter rule is perfectly acceptable. Far from sending for Stress Engineer, I rarely bother even to look up the reference tables that simplify his efforts for ordinary mortals, and instead deliberately loosen thread fits by using slightly oversize tap drills, or reduced diameter rod. I do that because it increases the life of the tap or die while reducing the chance of breaking a tap in the hole. Pragmatism rules.

Dave

Edited By SillyOldDuffer on 29/06/2018 11:28:52

It all depends on the function of the parts being attached to one another, but greater thread bearing would lessen the likelihood of unscrewing over time where vibration might be an issue. With longer screws or bolts, one is more likely to notice a part becoming loose while still attached, rather than suddenly becoming loose and falling apart.

I like Jason's nuts logic. But don't forget to allow for the materials in use. Ie aluminium is going to want more turns. We had a head of the Design Office who used to quote 3 turns for the Al chassis. It was fine for the covers, mostly, but things like heavy vibration for aircraft design proving showed this up seriously.

Like many other things,it seems to be a case of "Horses for courses"

I read somewhere that a double decker bus could be supported from a 1/4" bolt True of false?

Has anyone ever tried it ?

Haven't heard that but a 1/2" bolt was said to be good to straight lift 5 T

Look at the length and dia of the threaded part of eye bolts for example vs weight of lift

It was Colin Chapman (of Lotus fame) who famously said that you could hang a double decker bus from a 1/4 inch bolt. But then he was well known for engineering on the limit....

Brave Man !!!

If you could actually get an 8 BA grade 8.8 bolt it should support 440 lbs before it snapped, which is more than twice my weight by a fair bit. Would I travel in a lift held up by an 8 BA bolt, don't be silly!

Nuts are about 1 x diameter thick, usually. Some cylinder head nuts etc about 1.5. I was taught 1.5 x diam. is considered max strength but 2 x is a bit of a safety margin.

Also, was taught that tapping into aluminium, use a coarse thread as fine will pull out easier. Some textbook theory is at variance with this but over many years I have seen BMW motorbikes with countless stripped fine threads in cast aluminium on drain plugs, filler plugs, cylinder studs and the like. But Harley-Davidson, who use coarse threads in aluminium, seem to last 80 years and more without stripping out.

Thanks guys, interesting.

Hopper you jogged my memory with 1.5 x D, I can remember now the old guy telling me that 1.5 times was the minimum you should work too but he always rounded it up 2 x D with his jobs.

Always taught dia x 1.5 min anything else is a waste of effort.

Colin

Screwed fasteners are just one example of the general case with all tension joints is that most of the load is carried close to the end of the joint. The outer end for joints between materials of similar young modulus and also when the rod, screw or other tension member is less stiff than the material of its socket or anchorage. If the anchor material is of lower stiffness then the load is carried mostly at the far end.

The typical variations you encounter in screwed joints have more to do with engineering and constructional competence than real physics.

Prof J.E.Gordon has a nice chapter on this sort of thing in his book Structures, or Why Things Don't Fall Down. Including a nice drawing of a clearly hard working thrush in the process of extracting a worm from the lawn with a stress distribution diagram illustrating that a long worm needs no more force to pull out than a short one. Well not unless the short one is very, very short!

Despite being originally published in the mid 1970's that book and its companion The New Science of Strong Materials, or Why You Don't Fall Through the Floor are excellent reads for anyone wanting bit more knowledge of the science involved in materials and structures. Seem to hit the sweet spot of enough information without obligatory maths to give a good idea of what is actually going on and when things aren't what you'd think they ought to be.

Clive

Plus one for 'Structures' by Prof. J.E.Gordon.

A good read throughout.

One gem I was pleased to find was what he says about the English longbow being made from MEDITERRANEAN yew. Yet that same Mediterranean yew doesn't work in the Mediterranean region because the climate is too hot. My father, a skilled woodworker, always asserted that our English yew was too splitty and windy-grained to be any good for making a bow. He would have been pleased to read the book.

Best regards,

Swarf, Mostly!

Back to the advice given to the OP ~ Geo. Thomas always struck me as a pragmatic model engineer and in the M.E. he wrote one article about the engagement of threads in studs and the thickness of nuts. As usual, he supported his claims with experimentation.. For example, he found "....for our purposes it is enough to know that an 8 BA (mild steel) thread will not strip if the length of engagement is rather more than half a diameter........(but) the figure of 1.25 to 1.5D which I have used is ample for its purpose".

Confirming the comments made by others above i.e. 1.5D in mild steel is enough.

That article and many other invaluable ones are reproduced in his "The Model Engineers Workshop Manual".

Stress Engineer knows that the cross section of a short length of thread is greater than the core diameter of the screw.

You don't need a great number of turns for the thread to be stronger than the thread diameter.

As Jason observes, why should a blind hole require a greater length of thread than a nut expected to take the same load?

Notable exceptions are threading into soft materials or with greatly truncated (under-diameter) threads.