Bolt strength for stupid people

Quote:

Originally Posted by jonse

Lots of misinformation in this thread.

...

Lots of misinformation in this post as Dougal has pointed out. I don't know why anyone is "thanking" a post like this.

Bush65 has lots of great (and correct) threads on bolted joints if you search.

e.g.

(I can't find the posts confirming that friction between bolted joints is the main force, however you get the gist from the posts below)

Quote:

Originally Posted by Bush65

For anyone who wants a reasonably accessible reference covering the basics for bolts, design, tightening, etc. download the Ajax Fastener Handbook. One link here.

Bolts and bolted joints is a much bigger subject than can be crammed into a small handbook - Handbook of Bolts and Bolted Joints: John Bickford, Sayed Nassar: 9780824799779: Amazon.com: Books.

Edit: also some technical info on the Ajax site.
http://www.ajaxfast.com.au/downloads...JOINTPAPER.pdf
http://www.ajaxfast.com.au/downloads...evstension.pdf

Quote:

To seal a joint which opens up during dynamic load cycles will require a gasket with enough flexibility to retain a seal.

As I have repeatedly stated the problem of introducing a gasket between the joint faces is that it greatly reduces the fatigue strength of the bolt. This reduces the load capacity of the axle. I have only done some quick mental arithmetic, but expect it to be "in the ball park", which indicates that portal axles bolted to stock axle tubes reduce the load capacity of the front axle to about 60 or 70% of a stock defender axle. This capacity reduction assumes that the bolts are pre-loaded sufficiently to prevent the joint opening. If the pre-load is not sufficient then the load capacity needs to be reduced even further so that the pre-load will prevent the joint opening.

Besides the greater moment and torsional shear loads during braking due to the increased offset and weight, portals impose additional load cycles upon the bolts that the stock axle does not have - torque reactions from stock half-shafts are transferred through the diff housing to the inner section of the axle tubes to the radius arms, but the portal box transferes the torque reaction through the swivel ball and bolted joint, into the outer ends of the axle tubes to the radius arms. When the bolts are already overloaded the last thing they need is thousands of additional load cycles reducing their life.

IMHO, oil leaking is a problem, but it is not the important problem here and no attempt to prevent oil leak should compromise the strength of these bolts. If they break on the Simpson crossing (which is a lot more severe from anything that I gather has been encounted so far), then a gasket won't prevent the diff or portal box from filling with sand while the truck is dragged back down a dune.

The svivel ball has a spigot that locates in a recess inside the flange on the end of the axle tube. There is a chamfer in the corner of the recess.

An O-ring could work if you obtained one of the correct section and inside diameter to fit into the space between the small fillet radius at the swivel ball spigot and the chamfer in the axle tube recess. Then you still have a metal to metal joint.

Quote:

Originally Posted by Bush65

In most cases with bolted joints, the thing which is important is the pre-tension in the bolt.

Tightening torque is one method for achieving a required pre-tension, but because the tightening torque is resisted by friction between the mating threads, and between the underside of the bolt head (or nut) and the joint/washer, only a fraction of the torque contributes to pre-tension.

The friction increases as pre-tension increases and also varies greatly from one bolt to another.

Research has shown that the accuracy of achieving desired pre-tension is in the order of +/- 25% when torque control is used. In many cases this is not acceptable and other methods are used, such as part turn method. For critical applications, one of the best methods is direct measurement of bolt stretch.

Australian Standard AS4100, Steel Structures Code, Clause 15.2.5.3 does not permit torque control to be used for tightening bolts in structural joints. With steel structures (e.g. buildings), bolts may be either snug tightened of fully tensioned depending upon the design requirements.

Machinery is usually intended to be disassembled and reassembled, so bolted joints are usually designed for bolts to be tightened to approximately 65% of their proof load. In some instances when loads are high and larger bolts can not be used the design will require higher pre-tension up to 100% of the proof load - these bolts should not be re-used.

The fatigue strength of bolts subjected to cyclic tensile loads, is increased when the pre-tension is considerably higher than the maximum applied tensile load.

In the case of the flywheel connection bolts, the torque load is carried by the dowel in shear and the friction between the flywheel and the mounting face on the crankshaft. Note the bolts are not loaded in shear - they can only be loaded in shear if the dowel and friction joint fail first (in which case the bolts will also fail).

To transmit the necessary torque through the bolted connection, the required friction is achieved through the pretension in the bolts clamping the mounting surfaces together.

There are a number of factors (including: shank diameter, thread pitch, thickness of plies, size of bolt head and depth of bolt engagement into the crankshaft) that determine the part turn angle to use for the required pre-tension. Also a number of factors for the relation between tightening torque and pre-tension.

So it is difficult to determine resulting pre-tension from those published figures, but having confirmed that Isuzu published these, and mistakes in interpretation were not made, then I would use the part turn angle method.

I know from experience and investigating many cases of bolt failure, that insufficient bolt pre-load leads to more bolt failures than over tightening. Failures due to over tightening usually causes bolts to fail while they are being tightened (in which case the bolt can be replaced immediately). - BTW, the last revision to the Australian Standard for nuts was changed to be inline with ISO and revised some nut thicknesses to ensure that bolts would break, before threads would strip, so that failure due to over tightening would be obvious to the person tightening the bolt/nut.

Taking into consideration the normal practice of specifying tensions of 65% proof load for machinery, I like to err on the side of over tensioning.