Oval-Shaped versus Nylon Self Locking Nuts

[930LDR]

The hardware kit provided by Pelican includes 50 M8 Oval-Shaped Locking Nuts and 100 M8 Nylon Self Locking Nuts.

Can someone provide some guidance on when to use one versus the other?

Thanks in advance.

[BURN-BROS]

Hi Rob,

Starting with the 3.2 and up (and maybe late 3.0 liter engines), Porsche used prevailing torque nuts (oval) for the perimeter studs and chain boxs to case.

Earlier engines used the Nylocs.

[930LDR]

Thanks Aaron - Wasn't sure if the oval-shaped supplied by Pelican were in fact revailing torque. Guess I'll use those on the oil pump and perimeter studs. Nylon self locking elsewhere.

[BURN-BROS]

You're welcome.

[tadd]

So long as the hardness class is the same, why not use the elliptical nuts? They will take heat/cool cycles better than nylon which will eventually cold flow and degrade.

Just a question...

tadd

[BURN-BROS]

The mag cases can weep thru the threads. The aluminum sealing washer and the nylon effectively eliminate the potential problem.

[Walt Fricke]

Regular nuts torqued against steel washers on 8mm studs to 18 or 19 foot pounds ought not to come loose anyway. I just use regular nuts for the case perimeter and the chain boxes, and have never had a problem.

The nylocks come into their own on the valve covers, which get a crush washer but are speced for a much lower torque, as you are not clamping vital parts together there. So they make sense. I have used these over and over, with what one supposes is decreasing nylon extra friction each time, and have never had one come loose. Can't say about their retaining the ability to keep the oil from dampening things a bit, but never any identifiable real leaks (the kind that actually drip).

But obviously if you bought a kit full of nuts, makes good sense to put them where intended.

[304065]

Originally Posted by tadd

So long as the hardness class is the same, why not use the elliptical nuts? They will take heat/cool cycles better than nylon which will eventually cold flow and degrade.

Just a question...

tadd

Here's something I don't understand.

Torque is a proxy for preload in a fastener-- it is an approximation of the force necessary to induce the appropriate stretch into the assembly so it doesn't loosen up and at the same time keep the fastener below the yield point (except in the case of torque-to-yield fasteners).

OK, we know that surface condition (rust, grease etc.) can alter friction values, thus altering the torque setting to obtain an equivalent stretch.

So riddle me this: when the factory went to prevailing torque lock nuts did they alter the torque setting and note that in the factory workshop manual?

Prevailing torque nuts, because of the distorted thread, will behave the same way as rust-- your wrench ends up clicking at a lower preload than in the case of a pristine, lubricated fastener. The result is insufficient preload, a bad thing.

Can anyone help? Or does this fall into the "difference is so minor that the factory didn't give a Schvitz about it?"

[PFM]

304065

That is the same question I had, Does this change the torque value required?

I will wait on this one.

[Henry Schmidt]

"difference is so minor that the factory didn't give a Schvitz about it?"

[tadd]

I assume the throw money at it answer is the torque-to-yield wrenches that monitor the gradient ?

So let me say a few stupid things...
1. How important is the 'correct' clamping force on the outer case? There are pins for location and sealing is from 'goo of choice'.
2. If one is not a critical number (like a clamping fit on the non-pinned CV joints), then does the under clamped perimeter bolts matter if eliptical nuts are used?
3. Given that the outer studs are exposed, wouldn't it take only one or two micrometer measurements to get one in shape?

All I know is "pins are for location and bolts are for clamping" cause it was beaten into me .

tadd

[Turbo_pro]

Originally Posted by tadd

I assume the throw money at it answer is the torque-to-yield wrenches that monitor the gradient ?

So let me say a few stupid things...
1. How important is the 'correct' clamping force on the outer case? There are pins for location and sealing is from 'goo of choice'.
2. If one is not a critical number (like a clamping fit on the non-pinned CV joints), then does the under clamped perimeter bolts matter if eliptical nuts are used?
3. Given that the outer studs are exposed, wouldn't it take only one or two micrometer measurements to get one in shape?

All I know is "pins are for location and bolts are for clamping" cause it was beaten into me .

tadd

Wow Tadd, that's what we needed to clear up that issue. "The science is settled"

[chris_seven]

The variation in coefficient of friction between the face of the nut and the face being clamped along with thread galling issues and inconsistent lubrication can often mean that for a given torque a signicant variation in preload can occur and values of up to 30% are often reported.

For fatigue applications where the use of preload is critical some understanding of the issues involved may be important.

It has become common practice to use a figure known as a 'K factor' which is also referred to as a'Nut Factor' in correcting for these potential errors.

The 'nut factor' is often an emperical figure that is applied using simple 'look up tables' and for example Riverhawk in the USA have a ready reckoner on their website.

Bolt Load Calculations/Formulas - Stress, Stretch, Torque, Turn-Of-Nut - Riverhawk Company

Henkel also have a table which shows the allowances needed for various grades of Loctite.

It real terms, however, it is better for critical applications to determine specific nut factors experimentally and there are various methodologies used and a 'Google' search will show a number of approaches.

Aero Engine and F1 companies tend to use their own standardised method and I have a copy of the approach used by Cosworth which I could provide on a PDF.

There is also an excellent NASA Document (NSTS 08307) discussing preload in bolts (21 pages) and this is also available as a PDF.

[Flieger]

21 pages on how best to tighten a nut on a bolt? I love engineering.

[euro911sc]

Originally Posted by Flieger

21 pages on how best to tighten a nut on a bolt? I love engineering.

when that bolt is holding your ass to the space ship I don't mind 21 pages Hell, make it an even 30 with pictures

[JohnJL]

I dont know how this thought got in my head, but I was under the impression that the oval nuts are correct inside any oil passages including inside the portions of the rocker assemblies that are included in the oil flow. For some reason I thought this was the case because the nylock nuts occasionally push out a sliver of nylon on assembly (I have certainly seen this on a number of occasions) and that is very much not something you want wandering around in oil passages, squirters, etc.

[JohnJL]

Originally Posted by chris_seven

I have a copy of the approach used by Cosworth which I could provide on a PDF.

I'd love that please...jlandry at johnjlandry.com

[Walt Fricke]

Regular nuts don't come loose when properly torqued, and are all one needs for the head to cam carrier junction. At least, I've not had one come loose on me, nor when used with the case peripheral studs and similar places.

The only where I think nylocks are valuable would be the valve covers, and that's because torque sufficient to prevent backing off will squish the aluminum/magnesium washers which are part of the seal.

[lucittm]

Another body on the dogpile...
I think there was a tendency to over tighten the nylock nuts if oil leaks because the nylock does loosen its "clamping" force on the thread surface over time. The prevailing force (oval, or interrupted concentricity) nuts will come to a given torque measurement with slightly less clamping force on the mating surfaces due to the increased drag on the threads compared to nylon and may prevent over tightening of the gasket. Is that right?

Mark