Nuts and bolts (technical)

As with most things, "it depends"....

Tightening to a specified torque is probalby the best way to torque when you don't know the length of the bolt or the stiffness of the stack of what you are fastening. There can be variations in bolt load due to friction in the threads, contamination of the threads or other factors, but if you are don't know what the specifics of the joint are, then you are just guessing trying to do torque angle tightening. Or, think of it this way, if you have a long bolt you need to have more angle of turn to get the right stretch into the bolt. For that reason, unless you really know a lot about the joint or have done a lot of work to determine the TASA you are better off tighening to a specifc torque.

What happens is that if you have a long bolt and a "soft" stack of items that you are fastening, the amount of tensile load in the bolt will be very different than if you have a short bolt or a very stiff stack that you are fastening if you are using angle of turn. In something like a con rod bolt, the person designing the system as the benefit of additional analysis and he knows his joint very well, so he can determine that for the length of bolt that he has, along with the stiffness of what he is fastening, an angle of turn will result in a specific amount of stretch in the bolt and that is how he wants it done. Similarily, ARP knows their fasteners well and in a con rod they are clamping a pretty well known stack of steel, so they can specify a stretch and hit their tensile load very accurately.

This is why bolting different wheels or adding spacers changes the stack stiffness, so we are better off tightening to a torque, to get the right tenslie load in the bolt.

If you can ACCURATELY measure the stretch in a bolt and know the grip length, you can accurately determine the amount of tensile load in the bolt and could back out an amount of stretch that would properly put the right amount of tension in the bolt, but using a torque wrench is plenty good enough.

Trying to use TASA at home without a lot of analysis and testing is a really bad idea. You need to research the joint and determine the parameters of the joint, and that takes specific tools and is more work than it is worth.

 

Cant find any arguement with anything Solo stated.

Early on when Ferrari changed to the Ti conrods I did a great amount of testing regarding different thread lubricants using an electronic calibrating torque / angle torque wrench. The differences obtained in stretch, calculated clamping force and torque at a specific angle was really surprising between the dozen compounds I tested just on the Titainium.

One has to know the thermo expansion amounts, yeild point of the fastner, the stack resistance as Solo brought up....too many variables to easily pick a TASA number out of the hat. That said, trial and error has its costs!

Dave

 

A friend of mine worked as a mechanic on an ocean going fishing boat. I remember him discussing this issue and telling me that the head bolts on the deisel engine were not torqued using a wrench or angle, rather they attached a hydrolic device that loaded the studs to a specific load that should give a specific elongation. If all this was correct they would then fit the fastner (not sure if it was a nut or some type of collet) and then unload the hydrolic system.
This method not only applied the correct clamping force but checked the specs of the stud and discounted any friction from lubricant or dirt.


Russell

 

Well this is part of my job so although I’m sure there are plenty of people with more experience it should be more or less like this…

Torque only values give a stress in the bolt which is around 90% of the yield. You can’t go any higher with torque only because there are factors (variation in material strength, friction and torque) which mean that it is not very accurate and if you go over the yield then obviously you’ll just turn the bolt off. There are tables of torque values available for standard bolt threads and friction values. The only time you would use less than these values would be if you are tightening into or onto something soft like aluminium.

Torque plus angle gets you into yield so the bolt is applying as much load as it can. These loads are 30-40% higher than the torque only values. With a bit of experimentation it would be possible to find your own torque and angle specification for any situation, at least with the lower and medium quality bolts (8.8 and 10.9 in Europe). The high quality bolts (12.9) are brittle and leave no room for error if you get it wrong. Most important fixings (head bolts, conrod bolts) are 10.9 quality and these can take well over 10% permanent yield before necking and breaking. So a good solution is to find an angle that gives you say less than 1% permanent elongation (of the clamped length) every time. This is big enough to measure with a vernier gauge and will allow you to use the bolt several (less than 10) times. If it is an important bolt then it would obviously be an idea to do the tests with one bolt and then use a new bolt for the final job.

The bolts and threads should always be well lubricated to get the highest load and the smallest effect of friction.

If it's the wheels on your car and you are racing, I'd still use the manufacturer's recommendations

 

If you have a lug or other bolt repeatadly coming loose I suggest there is a problem besides the proper torque. Any time I have had this happen on a race car there was always something I found to fix the problem like matting surfaces that were not flat etc. On one of my formula cars I replaced a nut that held the front spindle in to fix the problem. The other side never came loose and I checked the spindle and bearing and inspected the nut. I could see nothing wrong with the nut but after measuring the spindle and bearing stack after each session I could find nothing obvious. I replaced the nut and the problem went away?

I check the wheel torques every session before the car goes on the track! Even if I have not touched the wheels. I also torque them before I put the car in the trailer and still check them before I go out the first session. I have been racing since 1984 and have never lost a wheel.

 

I love fasteners
Image Unavailable, Please Login
Image Unavailable, Please Login

 

Not having an engineering degree its hard to agrue specifics, but experience has shown that most assemblies are more often overbuilt than under. As long as bolts are not overtorqued they usually dont pull out, snap or strech, and as long as they are reasonably tight they dont generally loosen up, provided the fasteners or material threads are clean and in good shape. It has seemed that corrosion is more of a factor more often than overall assembly practises. Most of the bolts that have given me grief are those that twisted off during disassembly due to corrosion and had to be extracted.

I recall back when torquing wheel bolts was in its infancy. IIRC it was the euro cars like Porsche and BMW that started it, they got all fussy about impact wrenches claiming they could warp the rotor, or overtorque the studs to the point they would be weakened. Most all the shops laughed at the idea, most mechanics just went by feel and most didnt think it so critical that rocket science needed to be employed. The only wheels ive seen come off were ones that were undertightened, or not rechecked shortly afterwards.

Just my two cents.

 

The big push to torque wheels in the auto industry started in the early 70s because GM cars were warping their cheap brake rotors like potato chips from overtorquing. Even the Porsche alloy lug nuts took abuse better than those cheap GM rotors. The Germans almost all had bolts that passed through the rotors so they were just clamped between the wheel and hub and could be torqued until the bolts broke and would not stress the rotor into warping.

 

I have delt with ARP a considerable amount for fasteners for Ferrari and they do not just draw numbers out of a hat. Considerable knowledge and testing goes into it but they can only comment on the fasteners themselves. If it is an application where clamping loads can cause distortion such as a cylinder or bearing housing where conentricity comes into play you need to be smart enough to know that and know how to cope with it. Likewise the strength of the material you may be screwing it into. Dave and I both have used their fasteners on F50 motors and are aware of some serious pitfalls in those motors if you are not careful.

Deviation from doing things exactly the way the factory outlines is always a minefield and should be approached with caution and in the case with Ferrari even doing it their way is all too often a minefield and is why those of us that have been doing it a long time and have learned the hard and expensive way too many times are so ready to spread the word of caution. Sadly all to often the only result of those cautionary comments are negative and derisive comments from the peanut gallery. In the case of ARP, if you are up to speed on what to look out for their products are a real benefit when it comes to correcting some of the serious shortcomings for some of the product offerings from Maranello. The F50 is a good example. I would never dream of putting one of those back together with all Ferrari parts. Some of it is unmitigated garbage.