Rotoflex bolt orientation

[billwill]

It is imperative that the shear force does NOT go across the THREADS of the bolt but ONLY across the shank/solid part non-threaded part.

Where the bolt is fitted with the thread exposed to the joint/joining/mating face between the doughnut and the flange then the bolt WILL undoubtedly shear.

Understand this simple point - shank of the bolt through/across the joint/joining/mating face - and you can't go wrong

Good point...

But this would be more a matter of the correct shank length on the bolt, would it not, rather than which way it goes in. Presumably the shank length should be long enough that only a small amount of thread is inside the metal hole/tube that the nut is against.

[Grizzly]

I read that you don't get an accurate torque reading if the nut is against the coupling due to the shape of the rotoflex tube (not being round on the later type).

[elj221c]

The bolt is essentially a 'fitted' or 'fit' bolt so that there is no movement of the bolt in the 'spiders' or the brake disc. Movement there will bring rapid bolt failure.

I repeat.....

[Quart Meg Miles]

There is no Mechanical lock between the Drive Flange and Couplings so the bolts are the only thing stopping the joint from turning, the tightness of the Bolt has nothing to do with the point the shear force is applied.

Friction is enough to lock them together and if the bolts aren't tight enough then they move and you can hear the bolts click as they limit it.

There are other important joints which also rely on friction between bolted surfaces e.g. brake calipers.

Perhaps we aren't talking about the same thing: I have assumed "stopping the joint from turning" means the whole Rotoflex assembly rather than the individual bolt fixing, and I have taken "the point the shear force is applied" to mean the joint between the materials being clamped together, which shouldn't move.

[rgh0]

The issue is not one of shear stress on the bolt but one of cyclic bending. / stress change.

As as been stated the clamping force imposed by a correctly tightened bolt clamps the rotoflex to drive shaft and friction transmits the torque.

As the shaft rotates the bolt sees bending loads on every revolution as the donuts twists to accommodate the shaft alignment from diff to wheel which is changing all the time. This cyclic loading is what leads to a fatigue failure. If the bending stresses occur at a thread which they do if a threaded section of the bolt goes across the interface between donut and shaft then the stress and time required for failure will be less. If the bolt is not tensioned correctly the failure can occur very rapidly.

Incidentally I would never use the current donuts with the design of bolt holes made from two roughly formed plates.

cheers
Rohan

[Grizzly]

Ahh ok, Thanks for explaining.

So if the bolt is tight in the hole it simply doesn't flex so doesn't cause stress fracturing? So it is still correct to fit the longer shank bolts?

[rgh0]

If the bolt is tight in the hole it will still see cyclic bending and tension loads. However If the bolt pre-load is sufficient those loads do not result in cyclic stresses of any magnitude and as a result you do not get a short term fatigue failure, regardless of whether the threads are across the join or not.

However good engineering practice does not put the threads across the join as the thread form at this location makes the bolt much more likely to fail rapidly if tensioned incorrectly and slightly more susceptible to long term failure even if tensioned correctly.

cheers
Rohan

[Grizzly]

Well, after mentioning this to a friend who is a retired Aero Stress man i think i have the jist of it now.

Seems It's all about how the threads are cut and how they introduce a sharp angle at the bottom of the tread making a weak point (cracks love to spread from a corner corner it would seem), the idea is to make the Flange nut side as tight as possible to the shank of the bolt which in turn remove the threads from the cyclic effect.

He seemed to think because the cyclic effect is increased in the Coupling end because its not a good fit on the bolt it would be very unwise to fit the nut to that end.

[elansprint71]

Any thread which has been "cut" will have stress-risers along the root of the helix. For this reason bolts for critical applications, e.g. in the aero industry will be rolled, rather than cut. If you understand the principle of forging vs casting, you'll get the idea. The "weakest" points will be any un-engaged threads i.e. those not in a nut, or other female thread, which will be subject to the highest level of stress concentration and therefore subject to plastic deformation.

Interesting to see folks going round in circles and, as Dr Johnson once said "a little knowledge is a dangerous thing".

Not only should we use bolts with the correct shank (aka grip-length in some parts of the world) but we should use correct grade. A cheap mild-steel Chinese bolt might be perfectly adequate in certain applications but snap like a carrot in others. Stick with what was designed-in and you will be OK. The detail codes are on the head of quality products.

For further reading I would recommend Carroll Smith's excellent "Nuts, Bolts, Fasteners and Plumbing Handbook". Rather mis-leadingly the "plumbing" referred to is not how to fix the drain from your bath-tub but refers to the high-pressure plumbing such as that used for fuel and hydraulic system. An excellent read for those evenings when it is too cold to go out and drop the differential on your head.

Ask Father Christmas to buy you a copy, it surfaces from time to time on ebay.

[Grizzly]

Yes and that is the reason no cut bolt thread should cross a joint it would seem. Threads are a moot subject with Stress men, he took one look at the Sue Miller 'Correct' bolts and called them cheap crap so yes Aero and Auto are two very different things.

Frankly there is a good reason i don't get into this sort of thing, my Dad was Assistant chief engineer at BAE before retiring a while back and if i bring up such things i get a 2 hour explanation which i zone out half way through (you know how the old timers get )

Just too add, i am looking at this from the perspective of using the later Couplings with the plate type holes, i suspect the cast and drilled type are far more stable but they tend to wind up more then the later ones so catch 22.

[Grizzly]

Interesting to see folks going round in circles and, as Dr Johnson once said "a little knowledge is a dangerous thing".

Welcome to Internet Forums.......... I spent hours searching this forum for information about some thing as simple as which way round the Drive bolts go and why? for the simple reason i had two conflicting stories (I usually use Mr Bucklands manual as verbatim, i know there are a few mistakes in the Lotus Manual but seems Mr Buckland was mistaken and people have taken it at face value because it is one of the few mistakes he has made) thing is i Paint classic cars so its not often i have to predict the breaking point of a Bolt but i have had a Bolt and a coupling go on my +2 so didn't want to get it wrong through Manual Misguidance......So you live an learn.

Btw, yes i came across that book mentioned before. Maybe worth buying a copy.