Lock nuts / Jam nuts – MEW 311

[Nicholas FarrParticipant @nicholasfarr14254]

Hi, well all the industrial machines by various manufacturers, that I've had my spanners on, that have had a normal nut and a thin one, I can't recall any with the thin nut under the normal nut. Theories are very good but not all of them always work in the real world.

Regards Nick.

[Howard LewisParticipant @howardlewis46836]

Like Nicholas, I cannot remember having encountered a thin nut beneath the full nut, always. like me, thin on top

Howard.

[Robert Atkinson 2Participant @robertatkinson2]

Whatever way fitted a pair of plain nuts provide only minimal locking against vibration. There are much better technical solutions. The only justification to use two nuts is historical accuracy. That being so, fit them as they were on the original design. If you need locking use a tab washer, split pin or modern solution.

Robert G8RPI.

[Frank GorseParticipant @frankgorse]

Michael,

I’m not a Welsh speaker but I know that there is no letter ‘k’ in Welsh,so it would have to be ‘llocnut’. Or of course you could simply use Lloctite instead.

[SillyOldDufferModerator @sillyoldduffer]

NASA's Fastener Design Manual makes sense to me. (Thanks to Ches Green.) It says:

• The 'experts' cannot agree on which nut should be on the bottom.
• This type of assembly is too unpredictable to be reliable.
• It's a bad way of locking a fastener and proper locknuts should be used instead.

I suspect the Original Post explains why jamb nuts never work for me! The instructions require a thin nut to be tightened to a specified torque and then held so it can't turn whilst a thicker top nut is torqued down hard to a different value. Tightening the top nut has to be done precisely so the lower nut's threads transfer the load to their opposite flank, thus working in the opposite sense as the top nut to resist unwinding. Yuk! And it's not just a matter of applying a couple of spanners and hoping because the required torque, whatever it is, must depend on the material!

NASA's full text :

Dave

[Michael GilliganParticipant @michaelgilligan61133]

Touché Frank

MichaelG.

[DrDaveParticipant @drdave]

I have been following this thread (no pun intended) with interest because I was also in the thin nut on top frame of mind. Now that I am back at work after the Christmas break, I have been able to look up "lock nuts" in our technical libraries. The first reference that I found was in the ESDU (Engineering Sciences Data Unit) library, one of the go-to sources of data for Aerospace Stress Engineers. ESDU 14002 "Lock nuts and other thread locking devices" notes that "it is preferable to assemble the thinner nut first because the threads of the second nut are required to carry a greater load. However, this is not essential". In addition to this, the Data Sheet gives the theory behind this statement, showing the load transfer between the various parts of the bolt and the two nuts during the various stages of tightening.

OK, it still looks wrong to my eye, but the thin lock nut should go on first, as noted by the OP and the references that he has linked to.

Dave

[Ramon WilsonParticipant @ramonwilson3]

I've just read this thread for the first time having seen the subject discussed ad infinitum in the past.

Having delved deep into it in the past I understand (I'd like to think) the logic of the thin nut first but – purely from a model engineering perspective and as a maker of stationary engines I'd say I'm firmly in the camp of thin nut last purely from an aesthetic point of view.

Whilst making the Corliss engine I perused may images in the volumes on stationary engines I have but could find little evidence of the former but many of the latter.

That said however I still found the need to put thin nut first because of potential 'you have the lock nuts wrong' kind of comment.

To me this still looks 'wrong'. The one thing that surprised me given the amount of technical evidence to be found to confirm the principle it was the real lack of practical evidence in the many images perused. I don't think I found, for instance, a single engine with lock nuts on the foundation bolts – just heavier thicker nuts set on what appear to be just plain plate washers.

This is one of those discussions (second only to the metric versus imperial topic surely) that brings forth good argument from both camps but for me it comes down to what looks right on the model just built. 

Keep smiling – Tug

 

Edited By Ramon Wilson on 05/01/2022 22:00:51

[Former MemberParticipant @formermember37818]

[This posting has been removed]

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Friday Rack on 07/01/2022 10:17:11:

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.

.

Spot-on logical analysis

That is why the Railway Bridge reinforcement that I showed looks like it does.

MichaelG.

.

Ref. p1 of this thread … and more here:

https://www.model-engineer.co.uk/forums/postings.asp?th=147730&p=1

Edited By Michael Gilligan on 07/01/2022 10:50:27

Edited By JasonB on 07/01/2022 12:54:24

[Nigel McBurney 1Participant @nigelmcburney1]

The practice of thin nut on top goes back many years,and was used on many of the engines we model or restore,in many designs the top cap just held the bearing halves in place and there was a gap between the bed plate and the bearing cap so the thick nut was NOT fully tightened,if tightened too much the iron casting snapped,and I have seen many of these welded or brazed,the thin nut was then tightened against the thick nut. The question is then why not just use two thick nuts, I can think of one example, the popular Amanco o/c farm engine did have two thick nuts on the main bearing cap studs ,though in this instance the cap nuts could be done up tighter as the bearing caps had shims under the cap ,Amanco were very good,production engineers so the reason for their bearing cap design was possibly ,two full size nuts meant only one part number,no assembly problems as it did not matter which nut went on first and ham fisted owners could pull the nuts up tight, and if pulled down tight why two nuts well it may be that customers expected to see two nuts. On the other hand Blackstone continuous lamp and hot bulb engines built up to the early 1920s only had one nut on each main bearing cap and its not that the lock nut got lost ,the studs were not long enough to take a lock nut. On big ends , again some caps pulled up tight others just pulled the bearing shells together though the pulled up tight caps were more popular,here again it was full nut and half nut plus a split pin,though by the 1920s threads changed from coarse to fine, and single castellated nuts with split pins came into common use,and as Ramon mentions bolts holding engines down just had single nuts,I once owned a 21/2 ton hot bulb portable and the bolts holding the engine bed to the girder chassis were whit and only single nuts despite the rough ride portable engines got when being pulled along rough roads, the only fastener that I found loose on that massive chassis was an iron rivet.I have read lots of discussions over the years on the lock nut position, i have never seen a lock nut under a full nut though I put this down to to never having been involved in industries requiring such fittings, I gather that the applications are highly stressed,and use very high tensile materials, which are far different to a century earlier where the small nut on top was adquate for the application,and materials of the time.so if you are modelling or restoring old machinery engines etc put the thin nut on top.

[Neil WyattModerator @neilwyatt]

BEGINNERS WORKSHOP

These articles by Geometer (Ian Bradley) were written about half a century ago. While they contain much good advice, they also contain references to things that are out of date or describe practices or materials that we would not use today either because much better ways are available of for safety reasons. These articles are offered for their historic interest and because they may inspire more modern approaches as well as reminding us how our hobby was practiced in the past.

[Nigel Graham 2Participant @nigelgraham2]

Neil –

Clearly where necessary we need compromise for safety, function or practicality on a model – I am not aware BR Standard Class locos had copper boilers and brass cabs – but very often need care for fidelity. We don't use 'Nyloc' nuts on a traction-engine's motion-work, for example, but replicate its original details.

We also know a lot of the early model-engineering designs and methods raise eyebrows nowadays for fidelity and safety alike – strange how those engines were ever built and did work!

We would though use modern methods and fastenings where hidden, or on a non-model like a riding-truck or piece of workshop equipment.

'

So how to get it right if original works drawings or unaltered prototypes are unavailable?

It's wisest to look not in old hobby publications but old industrial text-books, etc, ones fairly contemporary to our replicating or restoring projects, for things like this!

As I did – and it gave three solutions to going nutty:

1) The theoretically correct thin nut first (with explanation),

2) The thin nut on top, described as common practice only because it is simpler (not needing thin spanners, which the book hints were not readily available),

and

3) Two full nuts – but my reference-book admits that looks ugly and clumsy and not always possible for space reasons, so some manufacturers used two nuts of about two-thirds full thickness. (Presumably that still allowed two ordinary spanners).

Did Amanco use pairs of two-thirds height, rather than standard full-, nuts as their original fasteners?

Edited By Nigel Graham 2 on 07/01/2022 12:18:15

[Ramon WilsonParticipant @ramonwilson3]

Posted by Nigel Graham 2 on 07/01/2022 12:00:52:

It's often wisest to look in professional engineering text-books contemporary to what we are are replicating or restoring, for things like this!

As I did – and it gave three solutions:

1) The theoretically correct thin nut first (with explanation),

2) The thin nut on top, described as common practice only because it is simpler (not needing thin spanners, which the book hints were not readily available),

and

3) Two full nuts – but my reference-book admits that looks ugly and clumsy and not always possible for space reasons, so some manufacturers used two nuts of about two-thirds full thickness. (Presumably that still allowed two ordinary spanners).

Did Amanco use pairs of two-thirds height, rather than standard full-, nuts as their original fasteners?

As said Nigel, I have text books which explain this in detail but seven differing books on stationary steam engines with many many images don't support that it was a practice that was consistently used – little visual evidence of the 'correct' practice in any of the images in these volumes all by George Watkins

Stationary Steam Engines of GB Vol 1, 2 and 3

The Textile Mill Engine Vol 1 and 2

The Steam Engine in Industry Vol 1 and 2

As also said, for me the technical aspects are one thing, the model making aspects are another – BA bolts, nuts, even with the smaller head are not usually of true scale appearance compared to those in these images for instance.
Yep, it's a right ol' chestnut is this one

Tug

[SillyOldDufferModerator @sillyoldduffer]

Posted by Nigel Graham 2 on 07/01/2022 12:00:52:

It's often wisest to look in professional engineering text-books contemporary to what we are are replicating or restoring, for things like this!

As I did – and it gave three solutions:

1) The theoretically correct thin nut first (with explanation),

2) The thin nut on top, described as common practice only because it is simpler (not needing thin spanners, which the book hints were not readily available),

and

3) Two full nuts – but my reference-book admits that looks ugly and clumsy and not always possible for space reasons, so some manufacturers used two nuts of about two-thirds full thickness. (Presumably that still allowed two ordinary spanners).

…

Agreed, but did anyone notice the point I lifted earlier from NASA's modern advice on the subject? It's that locking with two nuts isn't a good idea! Either way round. Difficult to get the torque right and vibration tends to shake the nuts loose. Looks good and is practical enough on a slow running marine or fixed steam engine, but extremely unwise on an IC engine or anything else that vibrates quickly.

For that reason, lots of more positive locking methods were developed over the last 150 years: split pins, wires, nylon inserts, serrated flanges, deforming threads, adhesives, tabs and castle nuts etc. Two jambed nuts are simple and cheap, but are otherwise best avoided.

I remember reading somewhere that those spiky spring-washer things are more a statement of intent than an effective lock!

Dave

[Michael GilliganParticipant @michaelgilligan61133]

Posted by SillyOldDuffer on 07/01/2022 12:29:22:

[…]

I remember reading somewhere that those spiky spring-washer things are more a statement of intent than an effective lock!

Dave

.

Having previously worked for eleven years in a vibration test facility … I endorse that opinion.

MichaelG.

[DrDaveParticipant @drdave]

Ah! That's got the hornets going again… To address the earlier point (which appears to have been removed) if you put the thin nut on first, it will strip when you dissemble the joint. If you consider what happens, this is clearly not the case.

I.e. install the (thin) lock nut first and tighten to, say, 25% torque. This gives roughly 25% preload. Add the main nut & torque fully. This unloads the thin nut (and actually reverses the load that it sees, providing the locking function) and gives full preload to the assembly.

If the main nut is removed, it goes back to a thin nut applying the 25% preload: it cannot retain the full preload because that was introduced by the (now removed) full nut.

We do seem to have moved into two spheres here: the "technically correct" thin nut on first, and the "historically or aesthetically correct" thin nut last!

One point that might have been missed, asI found from the ESDU Data Sheet that I referenced, is that this all applies to bolts that are predominantly loaded in shear. For those in tension, a lock nut is not required (have you seen many on a car engine?).

And lastly, as others have pointed out, lock nuts are not an acceptable locking mechanism for safety critical applications anyhow. So is the order really that important?

[Michael GilliganParticipant @michaelgilligan61133]

Posted by DrDave on 07/01/2022 13:43:59:

[…]

One point that might have been missed, asI found from the ESDU Data Sheet that I referenced, is that this all applies to bolts that are predominantly loaded in shear. For those in tension, a lock nut is not required …

.

That’s “interesting” … considering that it is generally bad practice to load a bolt in shear

MichaelG.

.

P.S. __ The vey long nuts on my bridge example are clearly required [and presumably ‘engineered’] to apply the requisite load … and it is most likely that the thin lock-nuts are there to discourage tampering by person or persons unknown. 

Edited By Michael Gilligan on 07/01/2022 14:01:09

[DrDaveParticipant @drdave]

Why interesting, Michael? A modern steel building is held together by shear bolts, and the average aeroplane is peppered with them. I agree that applying shear through the threads is a no-no, but they bolt shank is fine.

[AStroudParticipant @astroud]

If used in steel buildings the bolts are likely to be high strength friction grip where the shear is taken by the friction between the 2 clamped surfaces.

[Ramon WilsonParticipant @ramonwilson3]

Yep, like I said t'is a ripe ol' chestnut.

[SillyOldDufferModerator @sillyoldduffer]

DrDave says:

• ' A modern steel building is held together by shear bolts' and

AStroud says

• 'If used in steel buildings the bolts are likely to be high strength friction grip where the shear is taken by the friction between the 2 clamped surfaces.'

I think they're both right because it depends on the circumstances. Deciding what matters is surprisingly complicated when you don't already know the answer.

One of the things I like about engineering is it's full of hidden depths. Even simple nuts and bolts get the old grey cells going. It's a good thing most jobs aren't critical and I can just whack a fastener of about the right size on and do it up tightish!

Dave

[duncan webster 1Participant @duncanwebster1]

Posted by DrDave on 07/01/2022 13:43:59:

Ah!…..

install the (thin) lock nut first and tighten to, say, 25% torque. This gives roughly 25% preload. Add the main nut & torque fully. This unloads the thin nut (and actually reverses the load that it sees, providing the locking function) and gives full preload to the assembly.

…….

As I said before this all depends on the length (stretchyness) of the bolt and the clearance in the thin nut. Most unlikely to give the required preload.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by DrDave on 07/01/2022 14:01:47:

Why interesting, Michael? A modern steel building is held together by shear bolts, and the average aeroplane is peppered with them. I agree that applying shear through the threads is a no-no, but they bolt shank is fine.

.

"interesting" was deliberately within quotation marks [for the sake of a little sarcasm]

Yes … proper 'fitted bolts' are O.K. in shear, provided that they are effectively 'dowel pins'

But any bolts used in clearance holes should be considered a clamping device, and the side-loads should be taken by friction between the clamped surfaces.

MichaelG.

[DiodeDickParticipant @diodedick]

In proper engineering, "standard" bolts and studs go through clearance holes and do not resist shear, unless something moves to take up the clearance. Resisting shear is meant to be done by friction (produced by the clamping force) at the interface. The thought of a portal frame structure squirming under varying load as the connections shift to and fro is quite scary. And yes I do know about flexure under load and wind conditions, but that flexure is accommodated by stress, and consequent strain, in the members, not in movement at the joints. The only bolts that are intended to resist shear are "fitted" bolts that are the full diameter and in reamed holes.

[Author]

Posts