Van's Air Force
Western Canada Wing
I was in a class some years ago that was totally devoted to proper torque in fastening systems. The instructor went to great pains all morning to show us how every day we were violating some basic ground rules of proper use of threaded fasteners. By the end of the morning, at the lunch break, I think we’d all about decided that it was hopeless, and that we’d never be able to successfully torque a bolt.
The afternoon was spent teaching us the correct methods to avoid the common failures we had discussed all morning.
In the exact situation you described, where a friction component was causing inaccurate torque readings, he suggested that we use a very accurate measurement of the existing drag on the fastener, and add that amount to the torque spec for that particular system. Since nut plates have a higher friction when new than when re-used, I might suggest running the bolt into the nutplate once, and then removing and measuring the torque the second time, and then adding that amount to the value originally intended. (For example, 6 in.lb drag + 35 in lb torque = 41 in lb corrected installation torque.) Measuring the torque on a new nutplate would give a higher reading which would then be incorrect for the reinstalling of the bolt. Each time in the future you remove and replace the bolt(s), you will want to use the corrected torque rather than the standard torque, so make a note of the corrected torque.
Of course, we all know that you only turn the nut and hold the bolt, right? In a perfect world, maybe. Anytime the bolt is turned and the nut held, the torque applied will be incorrect. The drag of the bolt in the hole will contaminate your reading, making it read okay when the actual torque is too low. What about tightening bolts onto soft materials (like Aluminum)? Did you use washers to spread the load and prevent galling? Did you use the correct thread lube? In some cases it’s none, with the plain, cad?plated surfaces bearing on each other. In other cases it can be some pretty bizarre (and not commonly available Sunday morning at the airport) lubricants. (We already have “Mouse Milk;” can “Yak Butter” be far behind?) Any substitution will give bad readings.
What we are attempting to do is not measure torque, but rather to approximate the amount of preload, or stretch, we apply to the bolt. Torque is only a secondary means of reaching our goal. Any of the factors (like incorrect thread lube for the torque spec given, incorrect technique on the part of the torque wrench operator, mis?calibrated tools, etc.) will give bad readings, which means improperly loaded fasteners.
For instance, did you know that your torque wrench—freshly calibrated by someone traceable to the National Bureau of Standards—is not accurate in the first 1/6th and last 1/6th of its range? That’s right: a 100 ft.lb torque wrench is accurate only over the middle 2/3 of its range. So it should not be used for anything outside of 16?84 ft?lbs! Size the wrench to the job, and don’t fool yourself into thinking that you can just figure the conversion from in lb to ft?lb, set that 100 ft?lb torque wrench to 3 foot pounds, and have 36 inch?pounds. No telling what you’ll have, that far away from the calibrated zone.
Having said all that, many joints in aircraft are designed so that the primary load is shear and not tension. In these applications it is only necessary to see to it that the bolt is still in place (nut has not come off) and the exact torque is of no other concern. It is difficult and awkward to measure small torques accurately. Most people over torque the airframe bolts on their planes. I used to employ a car “mechanic” (who learned his trade in the Navy) who used the “uniform torque standard” (he wrung the little ones off and left the big ones loose) and was relieved of his bolt tightening responsibilities. I never let him near my airplane.
(Editor’s Note: Bob has made a good
summary of bolt torquing practices, and raised some very important points.
However, I want to add some further detail regarding bolts mounted in double shear, as I think Bob's information could be misinterpreted.
On light aircraft it's not uncommon to find bolts used as pins, such as when a bolt, castle nut, and cotter pin are used as the pivot pin of an elevator hinge. In such cases, Bob's statement that it's "only necessary to see to it that the bolt is still in place (nut has not come off)" is correct. However, often when a bolt is double-shear mounted the bolt is essentially a clamp running through the joint, whose purpose is to prevent relative motion of the various parts of the
joint. Relative motion in a joint of that type will lead to fretting and eventual
fatigue failure of the joint. In such cases, proper preload is
as important as in any other application. For excellent
discussions of bolts and threaded fasteners in general, read Carroll Smith’s
“Nuts, Bolts, Fasteners, and Plumbing Handbook,” and John Schwaner’s “Sky
Ranch Engineering Manual.”)
A member of the
Web Standards Project
Support free speech.
Van's Air Force Western Canada Wing is not affiliated in any way with Van's Aircraft Incorporated. Western Canada RVator is not a publication of Van's Aircraft or any other corporation. All products reviewed or mentioned are not necessarily recommended for use by RV builders, but are described for information only. All builder's tips are presented only as a source of information and a forum for exchange and the sharing of ideas and construction methods. No responsibility is assumed, expressed, or implied as to the suitability, accuracy, safety, or approval thereof. Any party using the suggestions, ideas, or examples does so at his or her own risk and discretion and without recourse against anyone. The members of Van's Air Force Western Canada Wing, the editor of the Western Canada RVator, and all authors and contributors are not responsible for any product or builder's tips misuse, incorrect construction, or design failure, nor any other peril.
All material on this web site is copyright Van's Air Force Western Canada Wing, or copyright the attributed author, unless otherwise noted.