I am not too sure that I like the idea of Aluminium, even the new grades from Kaiser don't seem to have a greatly enhanced fatigue life compared to 7075 , which I spent some time fatigue testing in the late Seventies and have quite good data regarding both Kthreshold Stresses and fatigue crack propogation rates.

Also modulus is a bit low and deformation in twist and buckling would be a concern.

We have just finished the details of the new design.

http://i197.photobucket.com/albums/a...ps42b77aac.jpg

Very nice design Chris.
Are the bolts of the interferance fit type?
My old 427 Ford had those type of bolts (had to be replaced every time)...it stopped all chances of the bolt getting loose.
If I had one to look at...as I remember...it had a sort of triangular look to the thread area when observed from the thread end.
Bob

Chris, stupid question maybe, but why?
Is it just because you can (which is a reason I can understand :) ) or will the rods be better/cheaper than Carillo/Pauter/Pankl/R&R etc.

Two reasons;

The first is as you say, because we can.

The second is simply money. It is much less expensive to make rods to your own design than pay the current prices being asked in the marketplace.

It a question of who adds value and how much they charge.

It is the same argument with expensive fasteners.

The selling price of ARP fasteners in the UK is almost twice the price that they are sold for in the USA and even buying them and importing yourself the cost is still much higher than in the US.

By approaching specialist manufacturers in the UK such as MellishEngineering or Tachart and buying 250 pieces prices are lower even than US prices.

By specifying suitable materials and heat treatments there are good savings to be made.

Yes, that is an annoying fact...

If you decide to make a 3.6 rod that is significantly cheaper than a Carillo here in Europe, let me know! :)

Bob,

Bolts aren't interference, alignment is achieved with the dowels.

I have had another look at small details to try to reduce weight.

Now 530g without bolts and the FEA still looks good.

http://i197.photobucket.com/albums/a...ps0e36313c.jpg

Chris.
I wasn't referring to the alignment of the cap to rod.
The interference fit is the bolt thread to cap in your design.
The threads were very shallow...and they dragged a bunch during assembly (torque) but I never had one come loose...and I believe they were used in the Mark 7's for LeMans (with the wider rod crank journals).
I know there are "stretch" type bolts now...so perhaps these are a better solution..but in your design, I don't see how the stretch could be measured...so I guess torque is used.It still looks like a great design BTW.
Bob

You can measure the stretch on Chris' rod bolts no problem. The rod cap is open on the bottom end and you can access the dimple on the tip of the ARP 2000 bolt with your stretch gauge pin no problem. My Pauter rods are set up the same way.

Thank you Kevin...explains a bunch.
I should have known that Chris would have allowed for this.
Bob

I understand a fare amount about bolted joint design, clamp load, the yield strength and yield behavior of materials etc. but am curious to understand more...

in particular, why are fasteners in Porsche engines, taken beyond the yield point? (I think I have heard that at several joints this is the case).

Going beyond yield does not add much clamp load and if the clamp load is relaxed the fastener unloads along the elastic loading curve. So what are the benefits and reasons for going beyond yield?

I don't believe that the bolts in early 911 Con Rods are stretch bolts in the now accepted sense.

In Volume 1 of the Workshop Manual (1965) Porsche state that con rod bolts should not be re-used as they are 'malleable'

I think that this means that to develop the correct preload the bolt would yield and therefore could only be used once with saftey.

Clearly con rod bolts must be preloaded to a level that is greater than the cyclic service loads as this elimnates any fatigue loading and means the bolt can be selected on a simple strength basis. If the bolt is not sufficiently preloaded then it would have to be designed on a fatigue basis and it is likely that size and weight would increase.

I just don't think that the quality and process control around on the very early sixties gave a good enough product.

This comment regarding rod bolts disappeared from all later versions and additions to the 911 Workshop Manual and has never re-appeared.

I also don't believe that the concept of the stretch bolt was around in 1965.

The first modern stretch bolt I had experience of was in a FIAT Uno fitted with a FIRE Engine (Fully Integrated Robotised Engine).

This engine was as its name suggests built by Robots. The head was fitted - as is now common practice- using a multi-headed Nut runner.

A single drive cable drives a gear box with 8, 10, 12 spindles all which drive an individual nut from a common drive.

The control of torque was made with torque limiting devices atatched to each shaft as individual PID Control was impossible due the the common drive system.

Torque limiters in the 80's were not very precise and some problems occurred.

By loading bolts to just past yield the scatter in the preload that occurs with elastically loaded bolts that are simply torque tightened was reduced significantly.

The slope of the load/displacement curve reduces significantly at the yield point and errors in the applied torque result in less variation in preload.

Stretch bolts are not black magic they are just a convenient method of using automated assembly techniques in a consistent manner.

If we can carefully measure torque or elastic stretch accurately there is no reason to use a 'stretch 'bolt when rebuilding engines assembled with this type of fastener, a conventional elastic bolt will do the job just as well.

I've heard cost and weight savings being reasons Porsche used stretch-to-yield bolts, but this ^^^ seems like the most plausible.