Nut screws washer and bolts

[Kiwi BlokeParticipant @kiwibloke62605]

[Kiwi BlokeParticipant @kiwibloke62605]

No, not the apocryphal headline following a madman's assault in a laundry and subsequent flight, but a request for enlightenment.

Simple question, but I suspect the answer isn't. Why are shakeproof washers used under nuts, but are rarely seen under bolt (or screw) heads?

More puzzles. 'Keps' nuts or nuts with external star washers clearly 'lock' the nut to the substrate, but, once loosening starts, achieve virtually nothing. Spring washers also 'lock' the nut to the substrate, but also load the thread, increasing friction, should loosening occur. This seems a better idea, so why use K-nuts? And why not add some anti-loosening device to the bolt head too?

I'm asking this because we tend to take familiar things for granted, and tend to accept things as 'normal practice', without thinking. I've just been working on some machinery which used many K-nuts and 'cone nuts' (an all-metal nut, similar in principle to a nylock nut), but no spring washers. None of the mating screw heads had any washers. Also, on other machinery, some plain nuts on the end of bolts were used with no washers, others had plain washers, and some had spring washers. There seemed to be no pattern to this. Nylock, Philidas, stiffnuts, etc. I understand – I think.

Where can I find the answers?

[pgk pgkParticipant @pgkpgk17461]

My theory is that washers were invented as a cosmetic statement to spawn a whole new fashion industry 'cos someone was too mean to buy a longer spanner. Then along came the nyloc, but washer buffs (like i-phone owners) stay set in their ways of paying for jobs.

pgk

[Robert Atkinson 2Participant @robertatkinson2]

Lock washers are not placed under the head of a bolt in a nut and bolt assembly because the shank of the bolt provides additional friction.
Spilt / spring lock washers do add tension as well as friction but put an off-set load on the nut and thread which can be a fatigue issue in high load applications. Star washers are virtually useless in high load applications. They flatten and are just high friction. Both types damage protective coatings /surfaces. There are "Wave" type spring washers that are better with small fasteners. and don't damage the surface.

Critical nut and bolt designs lock the nut to the bolt, either directly or both to the substrate. Really critical ones like aircraft controls have thread locking and a secondary means of preventing the joint coming apart.

Google aircraft thread locking.

Robert.

[David Standing 1Participant @davidstanding1]

Posted by Kiwi Bloke 1 on 23/10/2018 11:14:04:

Where can I find the answers?

I don't even ask the question, unless there is a good engineering answer why not, if I want a nut and bolt that won't shake loose, I just go for nyloc every time.

[MichaelRParticipant @michaelr]

Have a look at Nord Lock Washers Here I have used these on my IC engines. Mike.

[David TParticipant @davidt96864]

Posted by MichaelR on 23/10/2018 15:13:59:

Have a look at Nord Lock Washers Here I have used these on my IC engines. Mike.

We’ve trialled those at work in lieu of a Nylock nut, but without huge success. I suspect that may be due to our application though as it is a little…….. perculiar. I find Nylocks to be quite reliable though.

[Kiwi BlokeParticipant @kiwibloke62605]

'Lock washers are not placed under the head of a bolt in a nut and bolt assembly because the shank of the bolt provides additional friction.'

I find that hard to believe if the bolt is in a clearance-diameter hole. Apart from that, I'm heartened to see that Robert Atkinson 2 shares the same ideas that I do.

Should serrated flange nuts be regarded as similar to nuts + star washers? They're popular these days.

I still suspect that common assembly practice owes rather a lot to dogma and questionable beliefs…

[Ian S CParticipant @iansc]

On a nut and bolt assembly the lock washer is usually placed under the nut, the bolt is held in place, and the nut turned to tight and the washer bites in, but if for some reason the nut can't be turned, the washer is placed under the bolt head, the nut is held in place while the bolt is turned to tight. You could probably start something like the argument about lock nuts, and which goes on top and bottom, thick or thin—–Don't go there please.

Ian S C

[HOWARDTParticipant @howardt]

Having worked on life time critical designs in the past and seeing all the calculations required to just prove a fixing method, I think the best is to do it the way you are comfortable with. There are numerous documents out there that prove by calculation and test how things should be fitted. Working in a model environment we don't have any safety critical fixings, except perhaps for those things attached to pressure vessels. I have attached a link which some may find interesting.

**LINK**

[Nigel BennettParticipant @nigelbennett69913]

Another +1 for Nord-locks. We had a chappie in some years ago at work to demo them. He had a little vibration rig, to which he affixed a correctly torqued-up M8 nut and bolt. It loosened within seconds. So did a pair of locknuts. And a bent-beam nut, and a Nyloc. The nut & bolt fitted with Nord-Locks just sat there and took it. Since then we've used them on all sorts of critical applications with great success, provided that you fit them correctly – IN PAIRS with the coarser serrations facing each other. (We had a complaint once from a client that screws kept coming loose – despite the instructions, the twits had used just one Nord-Lock per nut and thrown the "spare" ones away…)

Possibly David T's application wasn't suited to the way they work – the hardness (or otherwise) of the mating materials can upset them. Did you try approaching Nord-Lock themselves for advice, David?

[mark costello 1Participant @markcostello1]

Wonder if a split common lock washer on both ends of the bolt would offer any advantage?

[Brian WoodParticipant @brianwood45127]

The discussion thus far completely ignores the situation with studs where there are of course no bolt heads. In the past, lock nuts onto washers were used with 'legs' one of which could be bent onto the component with the opposite leg bent back over the nut.

Or for wheel bearings the situation differs again with a wide plain washer, bent one way onto the nut that sets the preload and back the other way to lock it against the lock nut of the assembly. Then of course there are aircraft wired nuts, it is not all about spring washers and nylocs

Brian

[Alan VosParticipant @alanvos39612]

I want to know who decided the roof rails on a VW Passat should be attached without any vibration proofing. Just nuts, bolts,and plain washers. We stopped on the A26 in France to try and find what was rattling round annoyingly in the roof load. Then somebody pulled on the roof rail. AARRGGHH. FIxed using a Leatherman and gaffer tape. Fortunately, a hire car on the company account, specfically ordered to take a roof load, so no bill to me for the proper repair..

[Perko7Participant @perko7]

I was told by my grandfather that the use of plain washers was to reduce friction between the nut and the substrate to allow more accurate torquing of the nut/bolt assembly. I don't know about star washers, i refuse to use them. I do occasionally use lock washers on larger nuts where there is some vibration as the washer will bite into the nut and substrate and reduce the potential for loosening. I also use 'Nyloc' or similar generic locknuts in smaller sizes as i find them more convenient, although they lose their effectiveness if removed and re-fitted more than twice.

My preference for critical items is castellated or slotted nuts with split pins, or wired nuts. All the big-end bolts and main bearing bolts on my old 1949 Ford Prefect had wired nuts. They never came apart and they did not need a lot of torque to keep them in place.

[Perko7Participant @perko7]

I was told by my grandfather that the use of plain washers was to reduce friction between the nut and the substrate to allow more accurate torquing of the nut/bolt assembly. I don't know about star washers, i refuse to use them. I do occasionally use lock washers on larger nuts where there is some vibration as the washer will bite into the nut and substrate and reduce the potential for loosening. I also use 'Nyloc' or similar generic locknuts in smaller sizes as i find them more convenient, although they lose their effectiveness if removed and re-fitted more than twice.

My preference for critical items is castellated or slotted nuts with split pins, or wired nuts. All the big-end bolts and main bearing bolts on my old 1949 Ford Prefect had wired nuts. They never came apart and they did not need a lot of torque to keep them in place.

[Sam Longley 1Participant @samlongley1]

Question:-

If one is using a pair of nuts as a locking nut & one of the nuts is a half nut which nut goes on first?

Does the half nut go on first with the full nut on last or is it the other way round?

I have a small model which will have that & this thread has me wondering. I did read about it years ago ( I think that it was in RCME mag) but cannot recall the correct answer.

Edited By Sam Longley 1 on 25/10/2018 09:46:52

[Kiwi BlokeParticipant @kiwibloke62605]

Half nut on first.

[FarmboyParticipant @farmboy]

'orses for courses

Surely much depends on the materials you're bolting together?

I was also told it was friction related, but that only applies to plain washers. They might also spread the load over a larger area than the face of the nut on softer materials.

Almost all the nuts I encountered on agricultural machinery seem to have used spring washers, and I was told they should be replaced with new whenever the were undone . . . but I can't remember that ever happening.

Mike.

[Ian S CParticipant @iansc]

Kiwi Bloke 1, sorry no. the full nut on first and torqued down, the half nut is then put on to just tight. In aero engine use, mostly cylinder base nuts on air cooled engines the full nut goes on and torqued (special ring spanner required to fit torque wrench} then the lock nut, these are a special nut pressed from thin steel(about ,020" thick) is put on and pulled up tight, usually with a ring spanner as you can't get an ordinary socket on the nut.

There was a long series in Model Engineer in the 1990s, and I think the same occurred many years before, The conclusion was as above, 1/2 nut on top.

Don't use a split pin on a stud, it won't lock it, it just becomes a bolt.

Ian S C

[Tim StevensParticipant @timstevens64731]

My views are based on the articles on nuts etc in books such as Machinery's Handbook, along with years of experience. In the first place, star washers (= shakeproof) are only fit for one application – the earth (ground) connection to the painted surface of a fridge or cooker. Spring washers are little better – see comments above about side loads etc. Also, spring washers can cut into the nut and component when removed, destroying flatness and providing nasty spikes for mechanic's fingers.

But I start from the position that a properly flat surface at right angles to the thread at both ends is key – get that right and tighten the nut properly (a full-page topic on its own) and you don't need a shakeproof or a spring washer. Fine threads can also improve reliability against backing off with vibration – just look at a modern big-end bolt.

Nyloc nuts can be a much better option, except in situations like inside a crankcase where the temperature can melt the nylon and it then goes through the (often cooler) oil pump. Loctite does the same job cheaper, but is also limited by temperature problems.

A flat washer can help avoid problems with surfaces that can pick-up under heavy loads, but a much better, neater solution all round is the Belleville Washer (= cone washer). This is the only way I know to provide a degree of extra load in the event of bolt stretch or material compression (caused by over tightening or over-revving, for instance), while not being affected by high temperatures and not damaging the assembly or the washer when re-used. And they can provide a good guide to the right torque if the washer thickness and diameter are specified carefully.

Damage to fingers is also a severe problem with locking wire and with split pins. RR engineers have a way with split pins which reduces this problem, but I rarely see it used.

I hope this helps some people and adds to the knowledge of some, too.

Cheers, Tim

[Howard LewisParticipant @howardlewis46836]

In my Industrial experience, most lock washers were pretty ineffective. Prevailing torque nuts, such as Nyloc, were pretty good, but only really reliable for single use. "Nylon" inserts in threads were pretty good, as were anaerobics.

Tightening to yield creates so much friction in the thread that the fastener does not slacken (!/2 UNF in W range, tightened to yield, produces a tensile load of 9 tons, and a permanent extension of 0.002 – 0.004". I've never heard of one of those cylinder head bolts coming loose!)

For the ultimate in retention, wire locking, a tab washer, or a split pin through a castellated nut, are very effective, if properly done. Rolls Royce tab washers were Stainless DDQ!

Howard

[HopperParticipant @hopper]

Posted by Sam Longley 1 on 25/10/2018 09:38:30:

Question:-

If one is using a pair of nuts as a locking nut & one of the nuts is a half nut which nut goes on first?

Does the half nut go on first with the full nut on last or is it the other way round?

I have a small model which will have that & this thread has me wondering. I did read about it years ago ( I think that it was in RCME mag) but cannot recall the correct answer.

Edited By Sam Longley 1 on 25/10/2018 09:46:52

This is a bit of a vexed question. In recent years there has been an ISO standard published, backed by science, that says the half nut should go on first and the full nut on top. This is because as you tighten the top nut, it takes all the strain, the half-nut beneath it becoming in effect a washer between the top nut and the casting below.

HOWEVER, the common practice up until very recent times was to put the full nut on first with the half-nut on top to "lock" it. I still have my tech. drawing books from when I studied marine engineering in the 1970s and it shows in several examples the half nut on top in the old manner. This book was a prep guide for the MOT marine engineer's examinations for chief engineers in charge of sea-going vessels so they took such details very seriously. (EG you could fail the exam if your drawing did not include a split pin where required.) Other old books I have concur with this old method.

I read somewhere that the reason the thick nut went on first and the thin nut second was simply because a standard spanner back in the day was wider than a standard nut so could not be fitted on to the thin nut once the thicker nut had been screwed on above it. Thus there was no way to hold one nut still while tightening the other. Makes sense but the source is apocryphal.

So I guess the situation is if you want best modern practice with the most chance of stopping the nuts coming undone, put the thin nut on first and the thick nut second, and use a thin, or ground down, spanner to hold the thin nut while nipping up the thick one.

But if you want your 19th or early 20th century model engine to look authentic, put the thick nut on first wiith the thin nut second.

[Kiwi BlokeParticipant @kiwibloke62605]

Ian, it was your suggestion earlier that the lock-nut order debate shouldn't be re-ignited, but then you did…

As Hopper says, it is (now) generally accepted that the half nut goes on first. The reasoning is only a few clicks away, should you choose to search out the science, rather than regurgitate dogma. The locking nuts to which you refer, sometimes known as Palnuts, are obviously much more elastic than conventional half-nuts, and compress and deform when tightened, rather than causing significant elongation of the stud. The deformation under compression 'bites' into the stud thread, thereby directly increasing nut-to-stud friction. They therefore work very differently from the conventional nuts being asked about. Stud elongation, resulting from tightening the upper (conventional) nut, unloads the lower nut, transferring stress to the upper nut, hence the upper nut should be the stronger one. When a Palnut is used, the lower nut remains the fully-loaded one.

Apart from any other considerations, if, as you suggest, you were to use a conventional nut on top, torqued 'just tight', it would tend to loosen more easily than a fully-torqued, non-locked, single nut, thereby defeating its intended purpose.

This whole subject is far more complicated than it seems at first sight, and it shouldn't be assumed that industrial practice is optimal at all times. I started the thread in the hope that it would get people to think.

Thread-locking compounds are probably the answer… ; )

Edited By Kiwi Bloke 1 on 26/10/2018 08:56:49

[Ian S CParticipant @iansc]

With the thin nut on the bottom, and the standard nut tightened down on top, the bottom nut becomes a washer once the load is taken by the top nut. The item in ME had the stress diagrams etc, and the modern science, rather than the 19th century science repeated in 20th century text books.

One reason that our metal work teacher gave for putting the thick nut on the bottom was that you could never find a thin spanner to tighten and hold the thin bottom nut, he trained as an engineer in the UK before WW2. AN military specifications say that the AN-315(thick nut)shall be placed on first, the AN-316 check nut is then placed on top, the AN-316 L and R nuts are also used for locking turnbuckles.

Ian S C

Edited By Ian S C on 26/10/2018 11:15:56

[Author]

Posts