Rod bolt stretch VS rod bolt torque

[DSMer]

I was recently assembling the bottom end of a Mitsubishi 6g72 ( 3.0L DOHC v6, twin turbo) for my Dodge Stealth. I had my machine shop install APR rod bolts into the factory rods, and resize them. While I was installing them I used a combination stretch/ torque method. ARP’s specs were 32lb/ft with their moly lube or .0068” stretch. What I did was zero the stretch gauge, torque the bolt and then measure the stretch. All but one bolt required the recommended 32lb/ft or just sightly more, say 34 lb/ft to get the .0068 stretch. Of course the last one was an odd ball. After I torqued it to 32, I measure the stretch and found it to be .0085”! I un-torque it and it retuned to the length it started with. I tried it again and it stretched the same and retuned to the same. So then I when with straight stretch. After I stretched it I measured the torque and found it was about 25lb/ft.

So my question is, do I go by torque or stretch? I don’t want this to end up like a few of the stories I have read were people went by one or the other and the motor ended up spinning just the rod bearing that had the oddly torqued bolt(s). I know stretch is the more accurate way to tighten fasteners, but I have read one to many stories of things going awry in situations similar to mine. Is this one bolt weaker? Are the threads just smoother so there’s less friction while stretching it? What would you do?

 

[MikeHalloran]

I'd destroy that bolt and get another one. The extra stretch suggests a hidden flaw.



Mike Halloran
Pembroke Pines, FL, USA

 

[Greenlight]

Go by the stretch and not the torque. The torque reading can be affected by many things, such as the "smoothness"/surface finish of the threads on the bolt and the nut/threads in the rod, the type and amount of lubricant used on the threads, the surface finish/hardness on the washer (if a washer is used), etc.

The bolt stretch is affected by the material properties and bolt dimensions.

Seven lb/ft is not unreasonable.

 

[patprimmer]

I would back off all the bolts, check for uniform finish on the bolt bearing surfaces, maybe swap the positions around and see if the variation follows the bolt or stays with the position.

If it moves with the bolt, I would replace the bolt, just in case it is related to a flaw rather than friction differences. A failed bolt is much more expensive and inconvenient than a replacement bolt.

If it stays with the bolt hole and cap, I would presume it is related to less friction on the bolt in that hole.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Warpspeed]

I am with Mike, throw that bolt away and get another, something is very seriously wrong.

 

[DSMer]

I am glad I asked! It didn’t seem right, as the only thing that could really change would be the surface finish of the contacting parts. I used an ample amount of the ARP moly lube on all of the bolts. I think I will try pat's idea, as it does make sense.

Thanks,
Ryan

 

[carnage1]

The best final number is still stretch the bolt is a defined length spring so strech direcly relates to the clamping fource.

 

[tbuelna]

Bolt stretch measurement (or more correctly "strain") is the best indicator of bolt preload, since it is independent of variables involved in the 'torque' method, such as friction.

Another reliable method for preloading bolts is the angular measurement. Where the bolt is snugged up, and then turned an addition number of degrees. This method will also provide an accurate and repeatable preload, since it is a function of the thread pitch.

 

[swall]

If these bolts have a turned down "stretch" section, I would mic them to see if the one oddball was undersize.

 

[patprimmer]

I don't think anyone is questioning whether stretch is a better method than torque, and degrees turn after snug is a mix of stretch of the bolt and compression of the joint, so is also a very good method.

The real question as I see it is why does only one fastener in a set stretch to the specified length at a considerably lower torque than the rest.

I think the question is, is this a flaw in the bolt, or a variance in friction. That presumption inspired my answer to move the bolt and see if the characteristic is dependant on the bolt or the position it is installed in.

My method will not determine if it is variation in friction that is bolt rather than cap related, like a loose thread.

It could determine if it was nut related if the variance follows the nut when the nut is moved.

I guess to test for bolt friction variance, test the drag on the loaded thread by measuring the torque necessary to compress a suitable strength spring a specified amount.

However, when it comes to the relative cost of a rod bolt, vs a rod bolt failure, if in doubt, throw it out.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Warpspeed]

I agree with all that, but is it really possible that one brand new nut has a vastly reduced friction to all the other nuts in the same box?

More likely a defective nut would add more friction and decrease stretch.

I cannot help feeling that a single soft bolt discovered in a new set of otherwise perfect bolts should be discarded without any hesitation.

 

[patprimmer]

It is entirely possible (even if not probable)that all but one bolt had it's threads slightly damaged on inserting into the rods, or one was damaged, then cleaned up with a die that is a bit tight and leaves a loose thread.

Also, all but one nut could be from a different batch and have slightly different thread diameter, or the machine shop might have lost one original and replaced it with one that looked the same, or, or, or.

However, I do say several times, that if in doubt, don't use it. I think we agree on that point.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[waross]

I'd be inclined to stretch it to destruction to;
A> Satisfy my curiosity about internal flaws, and
B> Make sure it wasn't inadvertently used later.
If it strips before it breaks, I've Still satisfied part "B".