Best “v” thread form

[SillyOldDufferModerator @sillyoldduffer]

If I understand it right, two factors decide thread angle.

One is strength: threads designed to transmit power such as ACME and Buttress etc, have low thread-angles. At the other extreme, high thread-angles ( 80° ) are used on thin pipe where a coarse thread is wanted for speedy assembly and strength isn't essential: electrical conduits and so forth.

The other factor is grip, which is vital in fasteners. Fasteners must be easy to tighten and remove deliberately, but unlikely to come loose by accident. Whitworth, Unified, Metric, BA, and most other threads discussed on the forum are fasteners for which middle of the range thread angles 45 to 60° are suitable. There are many other specialised threads, usually to meet a particular requirement, but also to ensure spares can only be bought from the original maker. Different logic may apply to their design!

Whitworth's thread was optimised for the heavy engineering of his time in Wood, Cast Iron, and Wrought Iron, but technology soon moved on. A few decades later, BA was optimised for light engineering, especially instruments, working in Brass, Steel and  alloys.

Whitworth's 55° came from averaging the thread angles of the multitude of home-grown fasteners made by British Industry at the time. Good choice because practical experience had proved 55° was effective. Later threads are a little more scientific. For example, BA thread angles are determined by the pitch which relates to the dimensions of the peaks and valleys applied to smooth out the crude V cut. BA rounds peaks and valleys to a specific formula, and this decides the thread-angle, hence apparent weirdness of 47.5°. The committee didn't start by specifying 47.5°, it's the result of their sums.

Whitworth's thread was a huge success, but not because it's the best of all possible threads. The real advantage was Whitworth nuts and bolts were interchangeable. As already noted Whitworth is unsuitable for light work, but manufacturing cost was also a major issue. Later engineers, notably Seller's in USA, but also the metric crew in Europe, simplified Whitworth by dropping complicated thread rounding and moving to 60°. Seller's and Metric threads are equivalently strong, but easier and cheaper to make. Not so obvious today because threads are rarely lathe cut, but simplified production was a big advantage when they were.

Dave

Edit: I hate smileys!!!

 

Edited By SillyOldDuffer on 16/11/2021 11:11:38

[File HandleParticipant @filehandle]

Thanks Dave, that does all make sense and answers my,possibly, simple question.

[EmgeeParticipant @emgee]

Dave

Your quote :
At the other extreme, high thread-angles ( 80° ) are used on thin pipe where a coarse thread is wanted for speedy assembly and strength isn't essential: electrical conduits and so forth.

Not quite correct, electrical conduits were 3/4"x16 TPI 55 deg currently 20×1.5mm 60 deg, if I remember correctly the pitch is the same across other sizes.

I wouldn't consider electrical conduit is thin walled either, older installations supported some heavy old light fittings but modern day units are built much lighter.

Emgee

[JasonBModerator @jasonb]

PG thread as used for conduit is/was 80deg

 

Edited By JasonB on 16/11/2021 19:01:57

[EmgeeParticipant @emgee]

Posted by JasonB on 16/11/2021 19:00:54:

PG thread as used for conduit is/was 80deg

Edited By JasonB on 16/11/2021 19:01:57

PG threads appeared in low numbers at the end of my working life but were restricted to tapped holes in equipment produced in Europe and required adapters to connect on metric conduit sizes, AFAIK metric conduits are still M20 and M25 in the UK.

Emgee

[Oily RagParticipant @oilyrag]

There was a case where a number of years ago I was involved with a 3 litre V10 F1 engine made in the UK by a Japanese manufacturer. The inter cylinder main bearing webs were breaking out of the main cylinder block casting after just 400km running distance in the car. The cracks were forming at the thread roots of the M11 main bearing / sump retaining bolts. The redesign to install M12 would add 7mm to the overall length of the engine and an increase in weight calculated as 2.8kg (the engine was one of the lightest in F1 at around 93kg for a fully dressed unit ). The suggestion was therefore made that we should use MJ11 form threads (MJ being the aerospace form with radiused root form similar to Whitworth ). The cost of MJ11 'specials' and delivery for taps was £400+ and 9 months.

A simple but effective compromise was reached by using 7/16 BSW which cured the crack formation problem. After this the Japanese head of design declared that "Mr Whitworth very clever man!" The taps were bought next day for less than £20. A subsequent re-design saw the adoption of BSW in most of the stressed areas with reductions in bolt sizes and overall engine weight. Sometimes it is difficult to improve on a good design – other times it is sometimes necessary to take a backwards view to see a way forwards

Martin

[clogsParticipant @clogs]

Oily Rag

thanks for that info/thread on the F1 engine….most illuminating…..

[File HandleParticipant @filehandle]

Martin, I think that you have posted this on another thread. it was partly the memory of it that made me think about the initial question in the OP.

[Oily RagParticipant @oilyrag]

Keith,

Apologies for repeating myself, it's one of the myriad problems of getting old!

However, in my defence, 'Clogs' has come to my rescue by seeing this (apparently) for the first time. So my repeating it was validated in his instance. It was an insight to the Japanese that Whitworth had calculated the best form of thread overall, when considering the whole application spectrum of threads.

As the Japanese Chief Designer, Mr Kanada, looked at the Whitworth form he declared that it was superior in all its calculations to any other existing thread. I think he was most surprised by his investigations and doubly so that it was created more than 150 years ago and even more so that it was now considered obsolete. Politics have ruled over Engineering in the case of threads and their forms.

The 'cure' of the 7/16 BSW thread was then extended to the use of both 3/16 and 1/4 BSW threads elsewhere on the engine.

Martin

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Oily Rag on 18/11/2021 16:16:37:

Keith,

Apologies for repeating myself, it's one of the myriad problems of getting old!

[…]

.

The story merits repeating … as does Whitworth’s thread-form !!

MichaelG.

[JAParticipant @ja]

There was a similar posting to this about two years ago.

The greater the thread angle the more the thread will "lock". That, together with the fine pitch of 26 t.p.i., is why the British Cycle thread uses 60 degrees. Conversely it is why ACME form uses 29 degrees.

During the previous posting someone, David I think, challenged me about this and I offered to show the mathematics behind the friction from pitch and thread angle. Other things took priority and the posting ran its course so I left it. If anyone is really interested I will provide it (in my own time).

JA

[File HandleParticipant @filehandle]

JA is it as simple as a greater angle means greater thread surface area, therefore greater static friction and greater lock?

[Oily RagParticipant @oilyrag]

….As my post above, friction is important but it is but one element of a threads function. To re-emphasise my earlier post….

…..It was an insight to the Japanese that Whitworth had calculated the best form of thread overall, when considering the whole application spectrum of threads.

Martin

[JAParticipant @ja]

Keith

In mechanics surface area has nothing to do with friction. The friction between two surfaces is solely dependent on the force vertical to the surfaces and the coefficient of friction. Mechanics gives a very good model of what is actually happening

Martin

Yes, agreed. I think the Whitworth form, including BSF & ME, is the optimum. Perhaps because it is product of one person's work and not a committee?

I will probably be driven out of the workshop next week by the cold. Therefore I will some of my time to write down the friction part including compound angles. It will be too late for this posting but may be useful.

JA

[File HandleParticipant @filehandle]

" Yes, agreed. I think the Whitworth form, including BSF & ME, is the optimum. Perhaps because it is product of one person's work and not a committee? "
Thanks – that sort of answers my original question. I had a feeling that this was true, but felt that standardisation at 60deg might mean this was better, I also understand that the rounded profile of the thread reduces the chance of failure due to stress, although there is a version of the Unified thread that copies it?

[SillyOldDufferModerator @sillyoldduffer]

Posted by JA on 19/11/2021 11:34:56:

…

…I think the Whitworth form, including BSF & ME, is the optimum. Perhaps because it is product of one person's work and not a committee?

I will probably be driven out of the workshop next week by the cold. Therefore I will some of my time to write down the friction part including compound angles. It will be too late for this posting but may be useful.

JA

Pedant alert: the Whitworth thread wasn't designed by an individual, it was derived by averaging British industry best practice at the time. Similar threads produced by different firms wouldn't fit together, but most of the daft ideas had been eliminated by trial and error. Whitworth standardised on the centre of the spread. His approach exploited statistics based on experiment rather than individual genius. Thury took much the same approach to Swiss instrument threads.

Whitworth's genius was understanding why a standard was important and then getting industry to accept it. His ideas took many years to catch on because all the old boys resisted. The belief that all change is bad remains rock-solid to this day!

alert over!

Whether Whitworth's form is optimum seems to be an open question. I've never seen a technical comparison of the two. Be really good to see some figures if JA can manage them.

Looking at the property list suggests the metal is the main driver, not the form of thread. They are Tensile Strength, Lower Proof Stress, Yield Strength, Proof Stress, Elongation after fracture, Contraction after fracture, Ductility, Hardness, Fracture Torque, Impact Strength and Surface Defects. Deciding which is best would seem to involve making a lot of nuts of bolts out of the same metal and then breaking them on measuring machines.

In theory, I doubt flattening the top of metric threads rather than rounding them makes any difference. Rounding the valleys as per Whitworth must reduce stress raising but flattening does a not quite so good job too. I suspect the difference is small. If stress raising in a bolt is important, my gut feeling is the job is exotic and probably shouldn't be using an ordinary fastener. Don't take my gut seriously. In my experience gut feelings are always inferior to measuring and/or doing the sums!

Dave

 

Edited By SillyOldDuffer on 19/11/2021 14:44:49

[Michael GilliganParticipant @michaelgilligan61133]

Posted by SillyOldDuffer on 19/11/2021 14:43:31:

[…]

Pedant alert: the Whitworth thread wasn't designed by an individual, it was derived by averaging British industry best practice at the time. […]

True for the 55° angle … but I’m not so sure whether it’s also true of the form

MichaelG.

[SillyOldDufferModerator @sillyoldduffer]

Posted by Michael Gilligan on 19/11/2021 17:33:07:

Posted by SillyOldDuffer on 19/11/2021 14:43:31:

[…]

Pedant alert: the Whitworth thread wasn't designed by an individual, it was derived by averaging British industry best practice at the time. […]

True for the 55° angle … but I’m not so sure whether it’s also true of the form

MichaelG.

You might be right though I'm sure I read it in a book. Memory is untrustworthy though. I reckon I heard on the radio recently that 17½ Cats provide enough heat to keep a fully insulated house warm. Impressive argument in favour of insulation given the average cat outputs about 10W.

Or did I dream it? I'm woken by a clock radio and often conflate stuff due to being half-asleep! Whatever the truth it's time to throw another kitten on the fire…

Dave

Dave

[Michael GilliganParticipant @michaelgilligan61133]

Sorry, I’m not authorised to share my copy, Dave … but you may be able to find this online:

On an uniform system of screw threads : communicated to the Institution of Civil Engineers, A.D. 1841

MichaelG.

[John ReeseParticipant @johnreese12848]

This explanation is going to be a bit simplistic. For any given pitch the thread thickness at the pitch line is the same regardless of the flank angle. That thickness represents the area in shear that resists stripping of the thread. Actually the total shear are is the tooth pitch line thickness multiplied by the length of the pitch line helix for the length of thread engagement. That length will be identical for a given diameter and pitch. The flank angle of the thread will make little, if any, difference in stripping force for the same length of engagement.

[Oily RagParticipant @oilyrag]

Just to explain further in the example of the F1 engine given earlier, the failure was not the fastener but rather the failure of the parent metal the fastener entered. The failure was the parent metal cracking out from the thread root, the thread itself was not stripping. Therefore the root radius, or lack of it, of the female thread in the parent metal was the weakness in the metric thread form. In recognition of this stress raising, the MJ series thread addresses this as does the UN(C)J form.

Martin

[John ReeseParticipant @johnreese12848]

A wider root like 60* will give a bit more strength at the root as would a radius. Unfortunately many taps are ground to a sharp point which exacerbates the problem of breakout. A UNJ tap or the metric equivalent will give the greatest root strength. Check the length of screw engagement to see if that is a contributing factor, ie, insufficient engagement.

[Author]

Posts