To Pin or Not To Pin

[Martin KyteParticipant @martinkyte99762]

Reading some comments on pinning crankshafts on another thread and getting ready to complete a loco crankshaft raised the question do I or don't I pin the webs. Could someone with a brain more used to doing stress calcs than me solve this question once and for all and maybe find the fault in my logic too.

Scribblings gave me a value around 6630lb's force yeald strength for a 3/4 inch dia 3/4 inch crankpin in a 3/4 inch web. Assuming a 1/4 inch pin is used diametrically (rather a large size) I came up with 2000 lb shear strength or 4000lb for the double shear which is less than the failure point of the loctite joint. So if the loctite fails so will the pin.

 

If the loctite fails at a higher torque than the pin the pin must then fail too making the pin redundant. If the arrangement is that the pin fails at a higher torque then the loctite is redundant. And before you say that the pin and the loctite share the load that can only happen if there is movement which there cannot be. and assuming we haven't come up with some cunning plan to pre-stress the pin before hand.

The only conclusion I can come to is pin it or loctite it but not both.

regards Martin

PS loctite do encourage the use of chemical retainers with interference fits and I can see the point of that because it's clear that the bond and the shrink fit will both add to the strength of the joint.

 

 

Edited By Martin Kyte on 10/09/2019 14:12:47

[Martin KyteParticipant @martinkyte99762]

[Neil WyattModerator @neilwyatt]

The function of the pin is to stop nagging doubts in a corner of your mind

Neil

[Jeff DaymanParticipant @jeffdayman43397]

Not debating strengths or stress, but just an observation – the loctite statements assume that there was enough loctite, it was chemically OK, and the cure went perfectly. These things usually do occur just as it says on the bottle. But if it doesn't and something goes wrong with the cure, and you don't also have a pin in place, bad things are going to happen.

I like to hedge my bets on cranks and links by using loctite and also using spring pins, the type at link below. Just my opinion, and food for thought.

**LINK**

[michael howarth 1Participant @michaelhowarth1]

Apocryphal only, but I have heard from a loco man that caution should be exercised in using WD40 type lubricants as they can adversely affect Loctite joints. FWIW.

Mick

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Neil Wyatt on 10/09/2019 14:49:35:

The function of the pin is to stop nagging doubts in a corner of your mind

Neil

.

I thought Religion was an unacceptable subject for discussion on this forum.

MichaelG.

Edited By Michael Gilligan on 10/09/2019 15:40:25

[BazParticipant @baz89810]

Where has religion been mentioned?

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Baz on 10/09/2019 15:49:16:

Where has religion been mentioned?

.

Neil wrote:

"The function of the pin is to stop nagging doubts in a corner of your mind"

… Which reads like Religious dogma to me.

MichaelG.

Edited By Michael Gilligan on 10/09/2019 15:52:41

[Andrew TinsleyParticipant @andrewtinsley63637]

I wonder where the sense of humour went? Nice one Michael.

Andrew.

[Nigel Graham 2Participant @nigelgraham2]

On those original calculations, which appear to be simply the static strengths of the crankshaft's components, the dimensions suggest to me a 7-1/4"g loco, so hefty piston areas and fairly long strokes hence crank radii.

However, what is the maximum shear load on the crank-pin likely to be, and perhaps more importantly, that on the web-to-shaft joint? Would the crank-pin shear load really approach a ton, even as a resultant of two cylinders at maximum cut-off and steam-chest pressure?

You don't give the details but a quick calculation gives a piston load of about 700lbs with 100psi in the cylinder, on a 3" bore; and that will apply until cut-off, although the torque developed on the joint between crank-webs and driving-axle (not crank-pin, as the big-end journal is sliding round it) will obviously go from 0 at dead-centre to maximum ft-lbs at slightly before mid-stroke. It's that the web-shaft joint, subject to a strong torque rather than simpler lateral (though rotating) load in the crank-pins and their joints, I'd regard as the potential weak-point.

The greatest shearing stress on the crank-pin, but with minimum axle torque, for each cylinder is at dead-centre or very slightly after, ignoring lead and admission pressure-rise effects. Short of taking indicator-diagrams, the actual pressure will be unknown but somewhat less than what the boiler gauge indicates. The crank load is partly relieved by connecting-rod angularity transferring some of it to the cross-head guides as the stroke proceeds, but the load obviously orbits the pin without trying to turn it, so alternating the stress directions.

Whilst appreciating it's necessary to ensure everything being strong enough, I think there will be a reasonable Factor of Safety in that proposed crank-pin assembly, but I'd worry more about that web-to-shaft joint. Still worth using both glue and pin though – and making everything as near to optimum as possible.

'

On using 'Loctite' and 'WD-40'…

I'd be happy to err on the side of caution and keep WD-40 away from 'Loctite' or indeed any adhesives apart from fully-cured epoxies, but despite what it says on the tin, it is not really a lubricant!

Nor really a penetrating fluid for anything but very light seizing, for which the good old 'Plus-Gas' is the purpose-made and better.

"WD" was chosen by the manufacturers as it stands for "Water Dispersant", its primary function, and it's good at that. It seems to be a paraffin compound, so is also good for washing proper lubricants out of bearings.

It is also good for cleaning a locomotive or other steam miniature after running, used sparingly with a clean cloth, but the water-repellent film it leaves is thin and not very tenacious, so for storage longer than a week or so I'd still wipe lubricating-oil over the cleaned bright parts.

It's quite handy as a cutting fluid for aluminium though!

[Martin KyteParticipant @martinkyte99762]

You are absolutely correct and of course I should have said web to shaft.

The point is why it's worth using glue and pin if indeed that is the case.

The question is about absolute failure modes really rather than what a loco would realistically produce. I'm really trying to acertain if there is any point at all in pinning as well as loctiting. The calcs (of which I am not best placed to do) should show if the pin is stronger or less strong than the loctite.

The crank pin dims were pulled out of the air as an example and as I said a 1/4 inch pin through the web is perhaps rather larger than would be typical.

The other side of the coin is the logical argument as expressed in my post.

At the moment I feel that there is no good argument for pinning if using loctite on the basis if the one fails then theother is going to as well.

The peace of mind argument is the same as Fred Dibnah banging in the ladder irons harder the higher he got. There was no point but he couldn't stop himself from doing it.

Hopefully someone will take the bait and do a proper analysis.

regards Martin

 

 

Edited By Martin Kyte on 10/09/2019 16:43:51

[JasonBModerator @jasonb]

Taking your figures, you would have a joint good for upto 6000 if just using loctite and a perfect joint. Or a joint good for 4000 if just using pins but to get them both to fail you would need 10,000 so the combined use is stronger than one or the other.

That gives me peace of mind as per my reply in the other thread even if over any load that is likely to be applied..

[not done it yetParticipant @notdoneityet]

Posted by Jeff Dayman on 10/09/2019 14:55:33:

….

I like to hedge my bets on cranks and links by using loctite and also using spring pins, the type at link below. Just my opinion, and food for thought.

Belt and braces are rarely both required but not finishing with your trousers round your ankles, when either of the two fails, certainly gives one peace of mind.

[duncan webster 1Participant @duncanwebster1]

Posted by JasonB on 10/09/2019 17:11:17:

Taking your figures, you would have a joint good for upto 6000 if just using loctite and a perfect joint. Or a joint good for 4000 if just using pins but to get them both to fail you would need 10,000 so the combined use is stronger than one or the other.

That gives me peace of mind as per my reply in the other thread even if over any load that is likely to be applied..

It's all about stiffness. As the OP says, if the loctite hasn't failed, the pin isn't loaded, or at least not very much, and if the loctite fails then the pin follows in quick time. They are not really sharing the load. Drilling a diagonal hole for a pin (if that is what is meant) reduces the strength of the shaft considerably. I once literally blew up a 1te jack trying to get some loctited wheels off, 17mm dia * 20 long seat. Scaffold pole on the jack handle and the cylinder burst. Get it nice and clean, use new loctite (it has a shelf life), and observe the manufacturer's recommendation for fits

Edited By duncan webster on 10/09/2019 17:37:19

[nigel jones 5Participant @nigeljones5]

Also bear in mind that whilst loctite may state that you have 10 minutes playing time before it cures to such a degree as to make manipulation by hand impossible I have found almost without exception that the true time is just a few seconds (I favout 272 for general fixings as it doesnt run). I would pin the crank.

[Ron LadenParticipant @ronladen17547]

I agree with Fizzy I was recently fitting the wheels to the axles on the class 22 using Loctite 638 and I didnt seat one correctly, when I realised which was no more than 30 seconds later I couldnt budge it, it was solid.

So despite what Loctite quote re working time in practice it is somewhat different.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by fizzy on 10/09/2019 18:02:13:

Also bear in mind that whilst loctite may state that you have 10 minutes playing time before it cures to such a degree as to make manipulation by hand impossible I have found almost without exception that the true time is just a few seconds (I favout 272 for general fixings as it doesnt run). I would pin the crank.

.

Not disputing your findings, but just a couple of thoughts:

1. Copper is a powerful catalyst for the methacrylate curing process
2. 272 is high-strength for a threadlocker … but may not be optimum as a 'retainer'

MichaelG.

.

https://www.henkel-adhesives.com/us/en/product/threadlockers/loctite_272.html

• Breakaway torque on M10 bolts of 23 Nm

Edited By Michael Gilligan on 10/09/2019 18:29:10

Edited By Michael Gilligan on 10/09/2019 18:31:15

[duncan webster 1Participant @duncanwebster1]

Old loctite goes off more quickly, if you adjust it after it has started going off it weakens the bond. Don't ask how I know. If you do have to get it apart, warm up the joint, then clean it all up with paint stripper to get the residue off before you remake it.

[Nigel McBurney 1Participant @nigelmcburney1]

I think in a large loco I would use square keys with keyways in shaft and crank ,made tight and driven in,plus loctite. An lternative I have used on small o/crank i/c engine with one inch shafts and crankpin was shrink fit plus 1/4 inch taper pins driven in hard and not meant to come out, no complaints from customer.

[John OlsenParticipant @johnolsen79199]

I've just received the latest "Funnel" magazine from the Steam boat association and there is some discussion there about pinned and loctited crankshafts for the Leak compound engine. Apparently the published build instructions from Camden press suggest doing it that way, and apparently quite a lot of people have had trouble with cranks made that way. I didn't have the Camden book when I built mine, just the original Model Engineer articles, so I went for a crank locally cast in SG iron, which is also what the Funnel now suggests as a solution. It is interesting that the pinned and loctited ones should give trouble, since in a steam launch there should not generally be a lot of shock loading, unless you are in the habit of running hard aground. I would have thought that a locomotive would be more demanding.

Press fits with no keys work fine on many motorcycle crankshafts, but getting the right degree of interference could be a bit of a challenge in the home workshop

John

[Robert Atkinson 2Participant @robertatkinson2]

I would dry assemble. drill for a small (1/8" / 3mm) pin and then clean and assemble with loctite retainer (correct fits etc). The pin is only an assembly aid to make sure everything is aligned properly. If you do use a large pin and loctite there will be some load sharing between the two as nothing is completely rigid.

Robert G8RPI.

[Clive FosterParticipant @clivefoster55965]

As the load flow is circumferential round the outside of the shaft and inside of the bore any drilling or pinning will cause stress concentration where the load lines scrunch up to go round the pin and its hole. Which can make afair difference to maximum load capacity.

As ever the more you look at it the more complicated it gets if accurate answeres are needed. Probably best approach is to follow standard, known to work, practice and run the numbers to ensure you have a decent factor of ignorance.

If you fancy a taper press fit, motorcycle stye the Velocette self gauging system is relatively easy. Basically the two tapers slide together half way before pressing on. Still got to do the maths as to load carrying ability, roughly proportional to the engged area of the taper, and hoop stresses around the female taper to be sure that it won't stretch. Riveted and pressed BSA system is a good example of the pros not getting it right. The DB34 Gold Star is known for wallowing its drive side shaft loose. Especially when used on the road witha doube R box. The basic parallel press fit of the shaft in the flywheel is actually slightly heavier than the Velo taper system giving a theoretically stronger joint. But Velo drive side shafts stay put unless seriosuly abused but Goldie ones will loosen. For all practical purposes of power performance and general engine loads the Thruxton is the "same" as a DB34.

Clive

Edited By Clive Foster on 11/09/2019 08:05:02

[Martin KyteParticipant @martinkyte99762]

So far I think the case for the use of pins in a loctite joint is not proven , at least to me. Couple of comments.

Yes keyways would be a lot stronger but we are talking about diametrical pins here.

Drilling for a pin reduces the area of adheasive contact and so reduces the strength of the bonded join in my example by about 5.5%.

You cannot possibly have equal load sharing unless either pre-stressed or the bonded joint can move, so were are still in the realm of the bond failing followed by the pin.

Personally I feel that the thinking may be somewhat historical. If in the past you were going to pin a crank and rely just on the pin 'gluing it all together with loctite holds the thing together whilst the pinning operation takes place. In this scenario we would be relying on the strength of the pin and any chemical bond is just a bonus. Early chemical bonds were really not as strong as current offerings and it takes time to build confidence in new techniques.( especially for us). Perhaps a s time went by the bonded joints became stronger with better the introduction of better products. and the pinning was done because that was what was always done. I don't know, but I'm interested in comments.

I do think the peace of mind statements are valid, just but for psychological reather than engineering reasons.

Finally how catastrophic is a joint failure on a miniture loco?

regards Martin

[SillyOldDufferModerator @sillyoldduffer]

My take on pins and retainers like Loctite is that neither are ideal for this job – a pin takes all the stress on a relatively small bit of metal and is likely to sheer, while the retainer is a chemical bond liable to be weakened by dirt or if moved while setting. Adhesives are also effected by sunshine, excessive heat, and solvents, including oil and water. That said, both could well be 'good-enough' to outlive the builder!

I thought adhesives made weak joints until I followed the instructions! Extreme cleanliness – a quick wipe with a cloth isn't enough, especially if any oil is present. Fussing the joint into position while the adhesive is setting is bad news, as is stressing the joint before the glue has fully set. Leave it alone!

The advantage of pinning is it's quick, easy, locates the two parts positionally and can be dismantled. But pins aren't strong. Done properly adhesives are a good way of making a strong joint with minimum fuss, but the parts aren't positively located during assembly and dismantling is difficult.

When fixing a crank to a shaft, I'd see the pin as being used to fix the angle but have strength provided by the adhesive. If the pin is strong enough on it's own, no point in using glue.

If the joint is to be dismantled, and a pin isn't strong enough, look to keyways or splines. Much stronger but far more trouble to make.

Thinking about the failure modes of a combined pin and retainer joint has left me deeply confused. I can't even decide if they share the load like strands in a rope, or if one does all the work such that its failure then stresses the other. It might depend on which strains first under load. The glue and the pin will both move when force is applied and probably at different rates : if the glue gives more than the pin, then the pin will share the load, otherwise the glue will protect the pin. Perhaps!

But back to failure-modes. In practice, I think sub-optimal glue is more likely to fail due to indifferent preparation and its environment than the pin. If a glued joint is slightly dirty, takes too long to position during assembly, and is then overheated, the joint is at risk.

Cranks in reciprocating engines are subjected to dynamic shock loads. The maths is beyond me, but I've read any calculation of safe values based on static loadings should be halved at least.

Dave

[Nigel BennettParticipant @nigelbennett69913]

I had the wheels come off (well not absolutely off, you understand) my 5"G Edward Thomas a few years ago. The Loctite gave way as a result of some over-large stresses caused (I think) by a hydraulic lock and as a result, the motion ended up in an interesting arrangement of angles and unsurprisingly it Wouldn't Go. I removed the coupling rods and ran it as a 2-2-2 for a few laps, rather than waste the nice bright fire, but it sounded a bit off-beat.

Later investigation showed the quartering on the driving wheels was about 120 deg not 90. I fitted some axial pins once I'd got it all back together.

The load on the Loctite had obviously gone far beyond expectations – but I've never relied solely on Loctite after that for sticking wheels on to axles. Crankshafts would, I imagine, be prone to similar unexpected stresses when in use.

I suppose it's down to where you want the weak point to be – did the Loctite failure on my ET wheels save me knocking a cylinder cover off or bending a rod or two?

[Author]

Posts