Error in Geometer reprint article on lock nuts
|Ian Newman 1||18/12/2021 21:25:30|
|20 forum posts|
I'm sure I will not be the first to point this out, and I have probably posted this in the wrong place, but....
In MEW issue 311 (Jan 2022) there is a reprint of the notes by Geometer describing methods of locking nuts on studs (p22). The reprint comes with a header giving the caveat that some information in the reprint may be obsolete or incorrect in the light of "modern" methods and techniques, but there is one error that I believe should be corrected.
The illustration "C" shows the use of a "jam nut" placed on top of a full nut, which is generally considered incorrect usage.
There are any number of textbook and internet site references that explain the locking mechanism and hence the correct location of the half depth jam nut, which should be under the full nut. A brief intuitive description is given below, in the description, assume the stud is pointing "upwards" and the nut is being tightened "downwards":
When a bolted joint is critically specified, the required clamping force in the joint is defined. This force requires a certain tensile force in the bolt (or stud). This tensile force is difficult to measure in a bolt so an approximation is used by measuring the torque applied to the bolt or nut when tightening.
In using a jam nut, the half nut is fitted first and tightened until half the desired tensile force is applied through the stud. In this state, the "lower" stud thread flanks are pressed firmly against the "upper" flanks of the nut thread.
Now the full nut is tightened down on top of the half nut. As it is tightened the tension in the stud projecting from the top of the lower half nut increases until it is equal to the tension due to the half nut. At this point the load is taken off the lower half nut and the full tensile load in the stud is carried by the full nut. In this condition, the flanks of the stud thread are lifted off of the flanks of the lower nut thread and the half nut threads are no longer in contact with the stud threads.
Tightening the top, full, nut past this point increases the tension in the stud, but also pull the stud up through the lower half nut so the "upper" stud tread flanks make contact with the lower flanks of the lower half nut.
The top full nut is now tightened to give the design tensile force in the stud. When this is achieved, the stud is stressed to deliver the specified clamping force, the top nut experiences the full downward force due to the tension in the stud, and the lower half nut experiences an upward force (equal to half the tensile force in the stud) pulling it in to contact with the upper nut.
In reality, as these tensile forces cannot be measured and are only estimated through torque measurements, this is a very hit and miss method and should be avoided in any truly critical situation.
All the best,
P.S. some sample sites supporting the above explanation are referenced below:
|Speedy Builder5||18/12/2021 21:49:05|
|2615 forum posts|
That's all OK until you dismantle the joint, whereupon unless careful measures are taken, the full tension of the joint may be taken on the half nut as the full nut is removed.
|Michael Gilligan||18/12/2021 21:53:11|
20190 forum posts
Ian … You may enjoy reading the thread that I started just over two years ago
[ with pictures of modern practice ]
|Andrew Johnston||18/12/2021 21:57:59|
6603 forum posts
So how does a counter nut work, as it goes on top of the full nut?
|Ian Newman 1||18/12/2021 22:08:59|
|20 forum posts||
The disassembly is the reverse of assembly: The top nut is undone and half the maximum tensile force is taken through the half nut.
All the best,
|Ian Newman 1||18/12/2021 22:25:56|
|20 forum posts||
Simple answer, it doesn't work. Check the recommendations of bolt manufacturers, international and national standards, etc.
The idea of putting the thin nut on top is not supported by the arithmetic when you actually analyse what happens in a bolted joint with a double nut.
It is like the idea that you can increase the strength of a bolted joint by increasing the threaded thickness of a nut: A bolt always fails in tension before the threads fail in shear (i.e. the bolt snaps before the thread strips - unless the thread is damaged, in which case all bets are off....). Therefore increasing the threaded depth has no effect on the failure point.
Do the sums - the numbers don't lie
All the best,
|Ian Newman 1||18/12/2021 22:27:41|
|20 forum posts||
Some lovely illustrations in your thread! Thanks for pointing it out to me.
All the best,
|36 forum posts|
The usual justification for putting the thin nut on top is purely practical - otherwise a very thin spanner may be required. This is not valid if the thin nut is tightened first, etc...
re - measuring the tension in the bolt: Triumph motorcycle big end bolts (5/16" or 3/8" dia.?) were tightened to achieve a specific elongation, and hence clamping force, of the bolt. I may have more detail on that, but is a bit late for crawling about in the attic.
The much bigger sizes used bolt heating is used to lengthen the hollow bolt and the subsequent cooling tightens it. In steam chests at close to 565deg C the stud-bolts are hotter than the heaters can achieve. There is a variety of "patent" nuts for high pressure applications, including hydraulic stretching, out there. Someone with recent experience in petro-chem might be able to expand (sorry) on this.
All of this is technically interesting, but of little relevance in model engineering.
|duncan webster||18/12/2021 23:38:46|
|3989 forum posts|
I see the reasoning behind putting the thin nut on first, but I think it is flawed. Lets say the bolt head is at the bottom of a vertical bolt. The theory is that I put the thin nut on and tighten it to 25-50% torque, this means that the upwards face of the female thread is abutting the downfacing male thread. Then put the thick nut on and tighten to 100% whilst holding the thin nut against rotation. This puts extra stretch into the bolt so that the upwards face of the top nut thread abuts the downfacing male thread, and vice versa on the thin bottom nut. This is critically dependant on the length of the bolt and the clearance in the thin nut/bolt threads. If the bolt is long and therefore relatively stretchy, the nuts will jam together at less than full torque, and so the bolt will not be properly tightened. If the bolt is short and stiff it could be that the stretch is sufficient to make the thin nut threads be in fresh air.
There was a very good article in ME not many years ago, and one by Tom Walshaw a good few years ago
|derek hall 1||19/12/2021 05:40:07|
|224 forum posts|
I agree with Duncan.
The late great Tubal Cain (T Walshaw) discussed this in the ME and proved that the thin nut goes on top.
I have worked in engineering all my life and have never seen any practical installation of a thin nut first then big nut, that's not to say somewhere there is an example.
This topic always seems to come up every few years! But I will go with the vast experience of Tom Walshaw and say big nut first then thin locknut on top!
|Keith Wyles||19/12/2021 08:57:47|
|94 forum posts|
When I have used lock nuts it has often been to prevent a nut undoing when it isn't fully tightened. Equally, only having full nuts, both were the same size.
|Michael Gilligan||19/12/2021 09:21:34|
20190 forum posts
Nipping back a couple of years, to the thread that I linked last night … I have an hypothesis:
The long nut is being used because it provides a long ‘bearing’ and keeps the load axial
… a shorter nut in that position would cant over; introducing stress concentration in areas of the thread, and localised loading at the nut/washer interface.
Edited By Michael Gilligan on 19/12/2021 09:28:42
|Gary Wooding||19/12/2021 10:28:01|
|983 forum posts|
It seems to me that the sole reason for torquing a nut is to apply a particular tension to the bolt. Under tension, the bolt stretches; uniformly up to the nut when it rapidly reduces as it gets further into the nut. The first few threads in the nut also distort to nearly match the shape of the stretched bolt. If a thin locknut is then tightened down onto the original nut, then the tension in the bolt, inside the original nut, is increased accordingly, but it certainly doesn't reduce the original tension in the bolt. How can it? So I can't see any objection to using an external thin locknut.
|Rod Renshaw||19/12/2021 10:41:47|
|376 forum posts|
I can't claim any personal expertise in this area but I am with the "thin nut on top" school because of Tubal Cain's explanation which included the fact that the bolt stretches and the nut compresses as the assembly is tightened. This factor, which renders most of the simple explanations void, is strangely missing from many of the counter arguments in previous posts.
Tom Walshaw ( Tubal Cain) worked in the heavy engineering industry most of his life until he became an academic and professor of applied engineering in later life. His analysis is good enough for me.
Sorry, just noticed Gary's post which I had not seen when I was typing mine, I agree with him .
Edited By Rod Renshaw on 19/12/2021 10:44:39
|Andrew Johnston||19/12/2021 11:09:15|
6603 forum posts
I was intending to start a discussion, but given the OP's sporadic posting history it doesn't seem worthwhile. I'll stay in the lock nut on top camp. I suspect that the nut and locknut arrangement has been largely superceded by single component solutions, which are cheaper and easier to install.
|Nigel Graham 2||19/12/2021 11:42:44|
|2137 forum posts|
I turned to, and cite below, one of my old handbooks written for professional machinery designers and engineering-degree students way back last Century, historically closer to what much model-engineering and model-engineering workshops replicate. Notwithstanding any sophisticated modern developments in aerospace engineering and the like - though very important in their spheres they are not necessarily relevant to our work.
It shows three practices (also assuming a vertical stud pointing upwards, for simplicity):
d = stud diameter. Half-nut depth = 1/2 d. Full nut depth = d.
Italics denote quotes.
A) Half-nut below the full nut.
B) Half-nut above the full nut.
C) Two equal-sized nuts each of reduced depth = 2/3 to 3/4 d.
A) is described as the correct practice,
B) is faulty but common for practicality - and only because ... spanners are rarely thin enough... !
C) allows practicality especially where there is insufficient clearance for two full nuts as was sometimes used - although that also ... offends the eye... we are told. Indeed it does!
The rationale, copied verbatim, my using bold for the original's italics:
... a little consideration will satisfy the student that is is the top nut which takes practically the whole load, so of course the thick nut should be there as the true lock nut...
So you must decide for yourself for your own project, really, since as like as not the trade-built prototype may have been "wrong" too - or more likely we see preserved examples re-assembled "wrongly" after overhauls - but that's not really an excuse when the fastenings are fully exposed, as on a marine-engine, locomotive or traction-engine.
Make the few thin spanners needed, from gauge-plate? Or mill down some low-cost commercial spanners?
If you making non-model equipment, e.g. workshop equipment or miniature-railway rolling-stock of purely functional form, by all means use alternatives such as insert-nuts.
One of the diagrams also adds a root-diameter extension to the stud, approx. d/3 high, to take a split-plin for added security. The reduced diameter also helps you locate the nut on the stud, especially in a confined space. The extended stud is shown with a somewhat flatter end than the others, and its thread not protruding above the nut.
Useful knowing such proportions in fine-scale replicating.
Machine Design, Construction and Drawing. Henry J. Spooner, C.E.; pub. Longmans, Green & Co, London, 1913.
It may be 100 years ago but the general theory and practice of bolts, studs and nuts have not changed in that time, save for detailed developments such as plastic-insert nuts and specials for particular trades or very exotic applications.
(When your model-engineering is concerned with replicating very old machinery, or using very old machine-tools, it is worth seeking out these old text-books written for the industry. Although we often make compromises on materials etc. for building methds, function and safety, such books are a valuable repository of contemporary detail practices, component proportions and calculations, etc.)
Edited By Nigel Graham 2 on 19/12/2021 11:44:27
1430 forum posts
I have always worked with the practice that if the thin nut is on top it’s because the full torque to stretch the bolt has been applied to the large nut on the bottom then the thin nut is fitted and torqued up against the full nut to prevent it from moving and reducing the torque applied to the bolt, the thin nut is acting as a “locknut”. Worked 22 years on aircraft and then moved into general engineering and only found thin locknuts on top of the standard size nuts, some applications had wire locking as well. Dave W
|Phil P||19/12/2021 12:35:10|
|802 forum posts|
I too am sticking with the thin nut on top.
I don't care how efficient the arrangement is in theory, but if the full size engine I am building a model off uses it, then I will as well just so it that it appears true to prototype.
|Ches Green UK||19/12/2021 12:59:06|
|57 forum posts|
Here's a link to a useful NASA 1990 'Fastener Design Manual .... https://ntrs.nasa.gov/api/citations/19900009424/downloads/19900009424.pdf I'm not sure where I first saw this referenced...it may have been on here. There is a 2012 version available on Amazon.
As mentioned in this thread, the Bolt Science website is excellent.
Vibration (mechanical and thermal) is the enemy of fasteners. As is poor installation. Bolt Science has a good video on the Junkers Test Machine for establishing a fastener's resistance to vibration.... https://www.boltscience.com/pages/junkertestvideo.htm
Vibration testing is part of the normal Environmental Test Procedure for aircraft parts, as it is for many other mechanical/electrical devices. I've watched an SMA connector unscrew itself (using a strobe to highlight the movement) as we swept up through the vibration frequency spectrum expected on the aircraft, and then 'tighten' itself back up as we swept down the spectrum. At the end of the test the connector was finger tight. Obviously the connectors would be fitted to a specific torque using a torque spanner.... something had gone wrong with this particular connector ...an investigation followed. IIRC, we found there may have been dust contamination in the connector screw that gave a false torque reading initially.
|626 forum posts|
I'm also a member of the thin nut on top brigade.....
For a laugh,
when I was in development of wind turbines we reg used bolts over 50mm in dia...
all were hand made with x ray testing etc etc....we used geared nut driver heads/torque multiplier's.....
often over hundred of lbs/ft....
no need for any kinda lock nut....we just relied on bolt stretch.....
ps... some of the bolts were 1.5m long....
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.