Torque Specs

[BDA]

I’ve recently had some of my long held assumptions and practices challenged.

For example, I was told that a beam type torque wrench was better than a click type. That shocked me but it made sense when it was pointed out that if there is an imperfection in the threads, it could cause the wrench to click prematurely. Fasteners are more likely to fail if they are too loose. Another factor is that if a beam wrench is held at a value, the fastener can continue to turn staying at that value (probably not said as well as it could be but hopefully you get the idea).

Then, I had always assumed that torque specs were assumed to be for dry fasteners. Someone much more knowledgeable than I said he assumed the specs are for lubed fasteners. I got ARP head studs recently and they come with its own lube. They are designed for higher than spec torque so that makes sense. I don’t think I’ve ever seen it specified in a manual. Does anybody know whether torque specs are for dry or lubed fasteners?

While I’m on the subject of torquing fasteners, it’s a good practice to retorque your head nuts or bolts after driving your car for about 1000 miles. I was surprised when an engine builder I knew back when I was racing retorqued a head by just setting his torque wrench to the torque spec for the head nuts and testing the torque for them. The problem is that they coefficient of static friction is higher than dynamic friction so his torque setting was never enough to actually move the nut. It’s better to loosen the fastener a quarter turn and then retighten it to the proper torque.

[TurboFource]

The most accurate thing is to measure bolt stretch as you eliminate the issues of lubrication etc….
but it is not so easy to do in most situations….

[BDA]

Yes. That is the underlying idea. Easy for a rod bolt. Not so much for a head stud! 

[jbcollier]

The only good torque wrench is one that is calibrated regularly.  And, yes, I have seen "bad" beam type torque wrenches.  Friend of mine rebuilt a V8 and the rods started knocking after a 1K or so.  Turns out his beam torque wrench was 20% off.

When was the last time you had your torque wrenches calibrated?

Make sure you store them properly and ONLY use them to torque fasteners, nothing else!

Exactly what to do and when depends on the engine.  A base Renault engine requires re-torquing only once.  A high compression Renault engine (807-12/13/20) requires re-torquing every 15K or so.  Most modern engines use stretch bolts and you just torque them the once and no re-torquing.

On "most" older engines, here's what I do after an overhaul.  Torque to spec.  Warm up the engine till the thermostat opens.  Shut the engine off and change the oil and filter.  Let the engine cool overnight.  Re-torque the head (in sequence, back off each fastener a 1/2T and then torque to spec) and reset the valve clearances (pushrod engines).  Drive 100 miles or so.  Change the oil and filter again.  At 600 miles, repeat.  Somewhere between 600 and 1200 miles, re-torque the head and reset the valves again.  Now follow the regular maintenance schedule.

[EuropaTC]

Then, I had always assumed that torque specs were assumed to be for dry fasteners. Someone much more knowledgeable than I said he assumed the specs are for lubed fasteners. I got ARP head studs recently and they come with its own lube. They are designed for higher than spec torque so that makes sense. I don’t think I’ve ever seen it specified in a manual. Does anybody know whether torque specs are for dry or lubed fasteners?

When I was working at ICI our Pipe Design engineer decided to investigate this. He set up shop with new bolts and a variety of lubricants and spent a couple of days messing about trying different methods of measurement, etc. I have a vague memory of strain gauges but it's hazy although the guy in question was one of those very thorough guys so it's probably accurate.

Anyway, the lubricant made an enormous difference, oil, WD40, moly disulphide grease, etc, they all gave different results for apparently the same torque reading. The conclusion was that the bolt should be dry/clean. Rust/dirt made the difference you'd expect so if it was a critical item then it was new bolts.

Brian

[BDA]

I guess I muddied up my post a bit talking about torque wrenches. JB, of course, gives important information. I would only add that technique plays a big role. Taking a torque reading requires you to keep the wrench moving smoothly throughout. As for the type of torque wrench, I won’t be retiring my click type wrench. I have a better understanding of their limitations but I don’t build motors for a living and of all the motors I’ve built, and people I know have built that were built with a click type wrench, I’ve never seen one with a fastener failure so while a click type wrench may not produce the right torque, I seems good enough (which I admit is a poor standard).

In case anybody is interested, I also asked a real professional engine builder I know what kind of wrench he used - he uses a Snap On Torque O Meter for his engines and a click wrench for less critical things like wheels. I’m interested in the new (to me) electronic torque wrenches because they might have the advantages of a click wrench but not have the deficiencies (but I’m not likely to buy one for the reasons above and because my professional engine builder friend also says that most of the problems with click wrenches comes from operator error).

My real question was meant to be whether torque specs should be assumed to be lubed or dry. So thanks also to Brian for his input. I agree that it makes a significant difference since some of what the torque wrench is measuring is friction so a lubed fastener will be stretched more than a dry fastener at the same torque.

Does anybody else have insights about whether workshop manuals mean by their torque values?

[BDA]

I'm surprised that given the difference in the effect of a torque value on lubed and dry fasteners that manuals don't explicitly specify which they mean and I'm also surprised this topic doesn't generate more interest.

For those and other reasons, I'm coming to the conclusion that the precise torque value is not THAT important for most vehicles that do not fly which is not to say they should be ignored.

Lubed, dry, click wrench, beam wrench, digital wrench - do the best you can whatever way you do it.

[TurboFource]

I think it is implied that it is “dry” spec unless otherwise noted….

[Pfreen]

That is what I have always assumed.  Dry, unless noted.
The ARP con rod, main and head bolts have always required their grease along with a torque spec.
I have worked on large engines that required special grease along with a torque spec.  I have also used hydraulic stud stretchers which preload the stud, and then the nut is basically hand tightened, guaranteeing the correct stretch.
However,  I have never seen in my experience automotive specifications with lubed bolts.   
Maybe new engines do.  I don't know,

[BDA]

I've always assumed dry too but I recently had a discussion with a professional race motor builder who assumes lubed. Just saying...

I don't expect to be building any more motors but if I do, I'll use dry fasteners. Right or wrong, good or bad, it seems to have worked for me till now. I'll bet if I had been lubricating my engine fasteners all that time, they would have worked too.

[Pfreen]

The torque difference is over 30%, so 30% more stretch.  I would not think that is good on a critical item.  I think best to use ARP bolts on those items.

[BDA]

You can't argue with quality and certainly ARP makes great stuff! I put a set of their head studs in my motor and was very impressed with their quality and attention to detail (eg. there is an internal hex in each stud so you don't need a stud wrench).

Fun fact: More fastener failures happen because of under tightening than over tightening. In case that surprises you, see this: https://smrp.org/News/Solutions-Monthly-Newsletter/Member-Written-Article-Submissions/Bolt-Failures-Why-Learn-to-Recognize-Mechanical-Failure-Modes

[TurboFource]

The torque or stretch has to exceed the maximum load that the fastener will experience so the fastener does not "work" and fatigue resulting in failure....if that makes sense...

[BDA]

I don't know specifically with regard to the max load as you put it. I think you're saying that the preload has to exceed the load the fastener experiences or it will fatigue. That seems to be what the article I linked to was saying. One of the more common fastener failures is due to fatigue because it was not given the proper preload. To quote:

The most common failure mode of bolts is fatigue failure.  Bolts can fail by due to several reasons including (1) the bolt had not been sufficiently preloaded/tension during installation, (2) had lost its preload during service, (3) the applied cyclic loading had been excessive – far beyond the strength of the bolt and/or (4) the bolt had been understrength (quality issue/wrong grade).

My engineering classes were a LONG time ago but I went back to one of my old text books to look up about bolts. There was a section on fatigue loading, there was discussions about preload, and I was quickly lost (in my defense, it was over 40 years ago that I took that class!). It would be pointless for me to try to interpret what my text book said about our situations but I say this (and it applies to all materials) that a fastener that has been stretched past its elastic range (the fastener will have yielded) will not have the "spring" to clamp the pieces together as well. If the force continues after yield, the material will continue to stretch till it brakes.

There is a thing called a Torque To Yield bolt that is a new thing for me. From what I've been able to find out, they are supposed to be tightened past the yield point. And again, from what I've read, the use of these bolts is for use on modern engines with aluminum heads and are supposed to in some way deal with the fact that aluminum has a coefficient of thermal expansion twice that of steel or iron (or something else that would be important. what I've read was not specific). I know nothing beyond that. I don't believe that pertains to our motors (even those like the twin cam with an aluminum head and iron block). Those bolts are use once, then throw away and as far as I know, Torque to Yield bolts are not used on our motors.

As far as I know (and I don't claim to have up to date information) Torque to Yield bolts are a special case and the normal situation is that the bolt should not be stressed beyond it's yield point.

[TurboFource]

That is what I was trying to say