Reuse of Fasteners

<...The bolt supplier and manufacturer recommend that these bolts should be replaced....>

Now, why would this be? Um, errr..........

1) The grade 8.8 is a bit low on our side of operation to be called HT

2) On our gear boxes, and other components the bolts are used, and quite a few are removed and re-assembled almost once a year. We have not recommended replacement and I am not sure the customers will enttertain even if we do

3) The bolts are upto M52 size abnd under fatigue. If the joint is properly designed, I do not see fatigue crack growth in normal condition, in fasteners,

4) A visual inspection of thread is usually enought. However if you have doubts, a LPI can be used to supplement, as is done in a few of our areas, where the failure have high impact. In you case, if you do the LPI for a duration, you may have some data about the probabilities.

5) You can have some other avenues too, say you have a set of fasteners with higher grade say 12.9 in first location, torqued to the value of 8.8.

These can be re used again and again, being much lesser stressed. and location 2 can assemble the 8.8 normal?

If you are the assembly plant, and only these are used for once assembly, no further stress than the tightening, I don't know why the manufacturer has recommended not to re-use?

look at this

http://www.fastenal.com/content/feds/pdf/Article%20-%20Reuse%20of%20Fasteners.pdf

I presume that the torquing spec for the two operations was the same.

During first torquing the thread surface does change and during second torquing the friction will most probably be higher. For the same torque the pre-load will correspondingly be lower and the risk to get loose higher. This is the reason it was recommended to scrap the bolts and use new. In fact BOTH surfaces are modified so that the new bolt will slide on an already modified female thread area with a slightly higher friction and a slightly lower pre-load at same torquing level. The use of a higher grade does not change the friction evolution. Your idea is not to be thrown away, you should be pragmatic: measure the pre load at first and successive torquing with same torque and same components and get an information about its evolution (losses). If those are repeatable you can give a specification for the second torquing so that you obtain SAME pre load. Do not forget requested is the pre-load. You can also measure the torque-angle function and compare results when you torque to same angle first and following times. The result will be similar. This way you can assure quality and cost reduction at same time.

"...never re-use fasteners..." is a common spec. Alas, unless the fastener's load (preload or process-applied) has gone beyond yield, or if the fastener has been subjected to high-temperature creep conditions, it's not necessary. This is similar to the instructions on shampoo bottles stating that you must apply and rinse twice

One of the comments suggested that you should "check the preload". An excellent suggestion! Unfortunately, checking it via torque is almost utterly useless. Torque is only a measurement of the resistance felt when turning a nut. In practical terms, torque has limited relationship to preload. In fact, applying the same torque with a calibrated torque wrench to two identical fasteners may result in one of them being too loose while the other is too tight (and possibly yielded). Because of this, even the torque-angle method is dubious. The only way to "check the preload" is to do it after the fastener has been tightened.

The preload of fasteners on critical applications is verified by measuring elongation. This is often done ultrasonically. One of the larger electrical conglomerates (the one with the meatball logo) uses load-indicating bolts on their busbar assemblies. These change colour as the proper preload is applied. The original colour returns if the bolt loosens. Since colour change is somewhat subjective, this technology is not nearly as precise as measuring elongation. However, as simple "go, no-go" devices (particularly in such an application as yours) they're considered to be ideal. In any event, this is a much more sound way of bolting critical fasteners than the "guess and hope" method of torquing without verification.

If anybody is interested, I'll post relevant links to the technologies.

May I ask you to be so kind to read once more what I wrote. I mentioned the "torque-angle" function which gives a lot of information if tightening is understood and result correctly interpreted. Unfortunately the very few are those who try to go into the depth. Methods as sound are of course very good and precise, they have but a draw back: the contact surface must assure a good sonic coupling between emitter-receiver and bolt. This is not always possible and many times not economical. In bolting technology there are credos and instead of analysing and choosing the optimal technology which could not always be the most precise people narrow their horizon to the limited practical experience they have.

Hi...

A lot of work has already been done on bolting. I suggest you read all proper information.

Thanks,

Anil Tiwari / Delhi

if i may:

since the torque pattern has been determined, is it possible to just with hold these fasteners, since removal to facilitate further installation is required, applying the final torque values , sequence upon final assembly ? or possible just tighten the fastener to " snug ".?

possible to " hold " the "section " in place w/ tie wraps until 2nd phase of assy completed?

gl

2 cents

Thank you for all the comments made, it has given me food for thought. Having reused the same cylinder head nuts on my motor cycle a number of times made me wonder.

Hi...

Fasteners are used in 2basic ways:

1. As a shear fastener wherein wherein the load acts perpendicular to centerline of bolt tending to cut the bolt. In this case grade 8.8 bolts are a waste. You can even do with a grade 4.5 fastener. In this case the fasteners may be reused as many times as possible as long a the bolt shank is not worn-off or has developed cracks.

2. As a friction Grip fastener the load is transferred from one member to another through the grip area without passing through the bolt. In order to generate a proper friction grip you must tighten the bolt sufficiently or to about 70% of the bolt materials proof load i.e. a little below the yield point. For this purpose a calibrated bolt tension meter is to be used. Otherwise I could show you how to fabricate one.

Friction Grip bolts cannot be reused as the material is already stretched to almost its elastic limit in the first instance. Grade 8.8 is the lowest category of friction grip fastener and these friction grip joints are known to perform very well under conditions of fatigue.

Anil Tiwari / New Delhi

You forgot the very usual situation when the load acts parallel to the bolt axis and tries to separate the fastened parts.

Hi...

You must avoid using a fastner in this way. It is unsafe. They tend to unscrew under vibratory conditions.

Anil Tiwari / New Delhi

Bolts DO NOT unscrew when tensioned! They unscrew when there is a shear movement between the parts even very small. This is the reason why the universally recognized lab machine to test loosening is based on a transverse movement. Of course if the joint have not been correctly designed it can get loose also under tension but only if tension > pre-load. under such conditions the fatigue life of bolt is sooo short that it has mostly no time to get loose.

............if this was the case, one would look at the shear stress of the material.........and then of course one can always look at FOS.

Hi..

Steels of different chemical compositions exhibit vastly different tensile properties but under shear there is not much of a variation. As a thumb rule you may assume 5mt per inch as a SWL under condition of single shear.

Anil Tiwari / New Delhi

My first concern would be, "how do I know the bolts were torqued to the correct value originally?" (human error). How do I know the bolts I'm reusing haven't been over-torqued? In a non-critical application, it may be ok to reuse, BUT where a failure can result in loss of life (as in an airplane), bolts shall never be reused. In the U.S. Navy, components subject to high shock, where a failure would be catastrophic, fasteners are not reused if removed for any reason.

"None of the applied torque on the various size fastners exceed 65% of the fastner's proof load stress which is well below the yield point." Are you sure?

Hi..

HSFG or High Strength Friction Grip bolting has been in vogue for many years now. I first used it for joining large span bridge girders in 1981. The high tension bolts were of 24mm dia & had to be soo tightened that each bolt would ultimately compress the joint with a tensional force of 21-metric tonnes. The smallest joint had a minimum of 32 or more bolts. We had to ensure coefficient of friction of 0.45 between the mating parts. All the work was carefully done and the tensions in each bolt was monitered with bolt tension meters & torque wrenches in the initial spans. Later-on the workmen got sufficiently trained and were able to achieve the desired bolt tension by the part turn method. Hundreds & thousands of bolts were involved.

All the bridges are working fine at this coastal site. I am told that a HSFG joint ultimately becomes superior to the parent section and no failures have been heard-off. These joints have replaced revits world wide and have a much longer life compared to weld or revits under reversible stress conditions. They even out last the parent structure.

There is a lot of litrature available on these in India.

Anil Tiwari / New Delhi

"Always follow the manufacturers instructions"............I tell this to all of my students.

There are cases where fasteners (nuts and bolts) are subjected to severe stresses, e.g. big end bolts...........in smaller engines, if these bolts are removed, they should be replaced with new nuts and bolts.

However in larger engines, including some of the larger high speed engines and on just about all medium and slow speed engines, when big end bolts are removed, their free length is accurately checked to ensure that the elastic limit (or yield point) of the bolt has not been exceeded.............if the elastic limit has been exceed, the free length of the bolt will be greater (all dimensions will be found in the workshop manual).

It maybe also worth noting that in these engines, fallible torque wrenches are not used,..............the loading on the bolts are carried out by measuring the actual stretch of the bolt. This is achieved by the use of "go", "no-go" gauges, micrometers, dial indicators, etc.................for reasons that should be obvious.

I guess in your situation the bolts are not manufactured with the degree of accuracy or consistency to allow you you to do this..............in view of this I can only say "Always follow the manufacturers advice"

Hello Mobi,

Great post! I have something to add:

When measuring elongation of fasteners (either mechanically or ultrasonically), it's very important not to assume that the free length of every fastener in a lot is identical. When verifying preload, a difference of a few .00x" or so can be quite misleading. Especially so in fasteners with shorter effective lengths and hence less elongation for the same load.

Regardless of the "quality" of the fasteners, it's prudent to record the free length of all fasteners. Thus, after they've been tightened, one can be assured of reading a meaningful delta-L measurement.

Exactly. Unless you've mark and measure every bolt used before tightening, you can't get any reliable data. Also since most bolts are cold formed, they won't even have a good flat parallel surface for accurate measurement.

"Always follow the manufacturers instructions"

Indeed. A very good suggestion if one needs to have a scape goat at hand in the event of failure: "Blame the OEM!" However, if one's intent is to prevent failure because one can't afford the down-time while waiting for repairs or for the court proceedings to be settled, this recommendation should be taken with a grain of salt. Or, rather, a huge train-load full of salt.

It's woefully disturbing to see so many manufacturers specifying "torque" without including some type of verification step in their assembly procedures. At the very least, they should be providing a disclaimer. There seem to be only a few who understand that providing a simple torque spec puts them at great risk for warranty claims (perhaps even extending to consequential damage costs!). Enlightened manufacturers (most often those who have been forced to pay such claims) routinely include the applicable elongation values with the recommended torques of crucial fasteners. Some even include the actual bolt stress values! These manufacturers will often not provide warranty coverage unless proof exists that each critical fastener carries the proper preload (usually a table of free lengths and subsequent elongations).

On the other hand, perhaps the other manufacturers are actually quite wise (although a bit devious!): If a joint failure occurs due to bolt relaxation or breakage, they could claim that their torque specification was based on certain assumed friction factors and that the customer should clearly have known that there is no way of guaranteeing those factors in actual conditions.

Measure the bolts before installation . Measure them again after disassembly. Same length? Yield point not reached, ready for reuse.

Also consider that in some aircraft installations, torque is measured by bolt stretch.

Hello Jerry NH,

I understand exactly what you mean when you write: "Also consider that in some aircraft installations, torque is measured by bolt stretch".

For the benefit of those who are not quite as astute as you, may I suggest that it be re-phrased to read:

"Also consider that in some aircraft installations, torque bolt load (or bolt stress) is measured by bolt stretch"

"Torque" cannot be measured (sic) by bolt stretch. Torque is measured by torque wrenches (for all intents and purposes). The effect of torque can be determined by measuring bolt stretch (and hence the torque can be adjusted to provide the intended bolt load/stress). It's a subtle but very important difference which many people still don't quite grasp. This may help

bolt load (or bolt stress) is measured by bolt stretch"

.............or strain.