IC Engine Bearings...

[Jon H]

This thread is to extend the discussion on crankshaft bearings in IC engines & is to discuss the specific issue of how they carry engine loads.

It came about as a discussion in the "Engine Failure" thread which began....

"In all of our engines its the front bearing that carries all of the load so it makes sense for it to wear out and not the rear. Normally something else finishes bearings off before they wear out but its not impossible. Contributory factors here would be things like prop strike or any other impact that would transmit load into the front bearing. An out of balance spinner or prop would also not help the poor thing out."

Jon - Laser Engines

And was followed up with...

"One thing I don't understand is why Jon says the front bearing takes all the load in lasers, I've changed bearings in my 70 and 80 at least twice and the 100 once (the 70 and 80 are from about 2003/4 and the 100 is much earlier) and in those instances the rear bearing was showing signs of distress whilst the front one seemed to be as good as the day it was installed, I changed them anyway but it seems to me all of the firing loads and prop thrust are focused on the rear bearing and the only time any undue thrust is placed on the front one (unless it's a pusher) is when a starter is used or when you inadvertently stuff the plane in nose first at high speed.

On some of my OS types I've not been able to get the front bearing for one reason or another and the original has been recycled though I didn't remove the front bearing from the casting. One of those, a 52 surpass, has since been pulling a southerner major around since it replaced a 2 stroke in the 90s.

I always try to ensure that I run the engines at full throttle for a short while after flying and store them nose up so gravity drains any residue from the bearings onto the backplate. It seems to work."

Braddock, VC

Now...read on...

Steve Hargreaves - Moderator

 

Its not just Lasers. In all ball raced model engines from Laser, Saito, OS, Enya, ASP, Irvine, Super tigre, ys (and all the others) the front bearing does all the work when it comes to thrust loads generated by the prop.

I just grabbed a crank and bearing and put a picture at the bottom. You will see that tightening the prop nut pinches the front bearing between the rear crank journal and the prop driver collet/thrust washer etc. This means the crank cannot float back and forward as it is held by this front bearing and normally there is a clearance between the face of the rear bearing and the face of the crank web to allow for expansion of the crankcase. If this clearance was not there the expanding crankcase would pull load onto the bearings as the steel crankshaft will not expand as much as the ali case.

Also, if the engine was not designed to put all the load on the front bearing tightening up the prop would also load up the bearings and you are likely to lock them solid.

Rear bearings take the loads of the crank and piston flying about but do not take any of the thrust loads of the engine.

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:43:04

Edited By Steve Hargreaves - Moderator on 20/03/2020 20:00:02

[Frank Skilbeck]

Plus don't forget the gyroscopic precession loads these will apply mainly to the front bearing.

I recently replaced some really old bearings on a 75 because they had got noisy, the old bearings were slightly notchy, engine had been run on castor 15+ years ago but more recently on synthetic on opening there was still castor residue gum inside. Got a 62 which had been laid up for quite a while, it didn't turn smoothly so i changed the bearings out, they were a bit gummed up and sticking as Jon notes above, new bearings and it runs a real treat.

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:44:26

[Shaunie]

I’m sorry, but how does the front bearing handle thrust loads?

surely it only handles out of balance forces generated by the prop and gyroscopic precession forces.
all thrust loads must be handled by the rear bearing not the front one.

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:53:15

[Martin Harris - Moderator]

I would have thought that would be the case. The front bearing of almost all engines is only retained by friction and many come out of the casing very easily with only gentle heating. The rear bearing is retained by a large lip and the crankshaft can't move forwards due to the crank web. The thrust loads are only a few pounds though and I've always assumed that the larger size of the rear bearing was to accommodate the hammering/rocking loads applied via the con rod.

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:53:54

[Jon H]

To be honest it did surprise me when i first learnt about it goodness knows how many years ago but it is the front bearing that carries the thrust of the engine. when you consider the mechanics of the thing it has to be the front bearing for the reasons i explained before.

When we make crankshafts one of the most critical dimensions is the distance between the rear face of the front bearing and the front face of the crank web. The same is true of the bearing housing, the spacing has a +0 tolerance on the dimension.

 

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:54:32

[Martin Harris - Moderator]

Sorry Jon but I really can't see it despite your explanation. What retains the front bearing? It can only be the crank web via the rear bearing pushing against the crankcase which stops the crankshaft moving forwards and the critical dimension simply prevents the driver locking up against the bearing housing...

Edited By Martin Harris on 20/03/2020 00:00:18

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:55:02

[paul d]

I tend to agree Martin, when a engine is running the prop is pulling forward taking the thrust loads off the front bearing (it's pulling) your right in thinking it's the rear bearing (providing the crank web is behind it) that takes the thrust.

I believe front bearing failures are caused by gyroscopic forces, out of balance props , prop strikes and of course the over enthusiastic use of electric starters.

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:55:35

[Jon H]

This confuses me a little. As a manufacturer of glow engines i am presenting a fact about how bearing loads are carried and yet no-one seems to believe me. Why? I appreciate that it is not what is expected and might go against popular wisdom, but it remains true none the less.

The front bearing is only held into the crankcase with a press fit, but that fit is tight enough to mean the bearing will not fall out. That is, apart from on my old and worn saito 45. Over time the front bearing creeps forward and i have to give the crank a well measured clout to put it back in the right place. Some engines, like the enya VT240, have a circlip ahead of the front bearing to aid in retention but its still the front bearing that takes the load. If i could measure the force my press puts on the bearings when i insert them i would. I would suspect its many hundreds of kg of force. More than the thrust of an engine that is for sure.

You cant use the rear bearing for thrust as doing so would mean all of our ham fisted propeller tightening loads just went into the rear bearing as we tighten the nut. It would be the same as putting a bolt through the two bearings and just tightening it up as there is nothing to bite down on to. To allow us to tighten the prop we need to pinch the front bearing so that the loads are essentially trying to crush the steel and not pull the bearings together. We then hold this front bearing in the front housing or crankcase with a sufficiently tight press fit that it wont fall out. In theory the engine would run like this but its best to support the other end of the crank too so we slap a bearing there as well.

This is a photo of a laser 70 crank installed in its bearings. As you can see, the crank web is nowhere near the rear bearing face so it cannot be carrying any thrust load. I chose the 70 as this clearance is larger than our other engines for various reasons so is easier to take a picture of but the principal is universal. There will be a small clearance between the front face of the crank and the rear face of the rear bearing. It will be really small but it will be there.

 

 

If my explanation has been poor and you guys still cant see it then we can start another thread and i will try again. I am more than willing to try and explain it as long as you guys are aware that what i have said is an engineering fact and will not be changing on account of your belief's.

 

EDIT

I just did a very crude test but i estimate i use up to 320kg of force to insert the bearings into our front housing as my press has a force multiplication of about 16 and i apply up to around 20kg of force to the handle. 

Edited By Jon - Laser Engines on 20/03/2020 09:35:18

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:45:27

[Andy Stephenson]

One effect on the entire engine/airframe arrangement that seems not to be understood is the effect of gyroscopic precession when using a 2 bladed prop. These precession forces act when the airframe is rotated in pitch or in yaw but this force is pulsating as the prop rotates. This is because the prop disc is not continuous. For example during a pitch up manoeuvre when the prop is horizontal the airframe won't experience much precessional force, however, when the prop is vertical the force is at it's maximum. You can see from this that there is a severe amount of fatigue cycling going on during aerobatic manoeuvres and this takes place at twice the engine rpm. Using a 3 or more bladed prop will even out this effect this is why full-size aerobats use 3 bladed props because you can imagine what effect this resultant vibration would have on the pilot otherwise.

A.

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:45:49

[paul d]

Jon, any chance of picture of that 70 crank minus it's bearings?

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:46:09

[Nigel R]

crank (nearly) floats in the rear bearing gents, the crank is not pressing against the bearing's inner or outer races... remember how easy it is to get the crank free with a light tap when disassembling (assuming you don't gum your motor with castor, but that's a different argument)

front bearing is held tightly as jon says by prop driver against step in crankshaft

rear bearing takes all the pulsed rotary loads from conrod, I guess it needs the extra size due to magnitude of torque peaks

Edited By Nigel R on 20/03/2020 11:19:34

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:46:29

[Nigel R]

Posted by Jon - Laser Engines on 20/03/2020 09:23:53:

I just did a very crude test but i estimate i use up to 320kg of force to insert the bearings into our front housing as my press has a force multiplication of about 16 and i apply up to around 20kg of force to the handle.

 

Does that hold true when the operating temperature climbs?

I'd expect a slackening of the fit and reduced holding force.

When we swap bearings as amateurs, the first tool in the box is heating the whole motor, and often the bearing just drops out

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:46:51

[Steve Hargreaves - Moderator]

So effectively the crank is held tightly to the front race by virtue of the inner race of the bearing being clamped between the step on the crankshaft & the collet holding the prop driver?

Think I'll stick to Electrical Engineering...it seems easier than Mechanical Engineering....

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:47:18

[J D 8]

If you want to feel how powerful Gyroscopic precession is take an angle grinder and hold it firmly in front of you and start it. Then turn your whole body quickly through 90 degrees. You will feel the grinder want to twist in your hand.

Do this with a 9 inch grinder and heavy disc be very careful as the effect is very powerful.

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:47:38

[Jon H]

Posted by paul d on 20/03/2020 11:09:32:

Jon, any chance of picture of that 70 crank minus it's bearings?

Its in an engine now. It looks the same as the one i posted before, just the journals are different diameters and the throw different. If its real important i can grab an unfinished one a shoot a photo?

Nigel and Steve, you got it. The bearing is held against the step on the crank, and then that is held into the engine. The rear race then 'floats' a bit.

Heat does make a huge difference to the holding force, but the heat needed for the bearing to drop out is higher than the operating temperature of the engine so its all good.

 

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:47:57

[Martin Harris - Moderator]

I really hate to argue with you on this Jon as I know and appreciate your expertise and I freely acknowledge that I'm very much an amateur engineer but unless we're talking at cross purposes, I can't accept that the front bearing takes any appreciable part in opposing the thrust of the propeller.

That front bearing is locked to the crankshaft when the propeller is tightened. Your picture shows a gap between the rear bearing and the crank web - as would be expected or the crank web would rub on the outer race of the bearing - or more correctly, have zero clearance. Looking at an old OS crank shows a small shoulder to provide that clearance and this video should audibly prove that the crankshaft hits the bearing:

While you may press your bearings in cold, I know from my own experience that front bearings often drop out of casings when the crankcase is heated although rear ones require a sharp tap on a block of wood and are sometimes a little reluctant to seat when replacing them - but they pull in by tightening the prop nut. While there is no argument from me that the front bearing takes axial loads from propeller forces I've never seen one fail - I normally change them as a precaution when working on an engine with a suspect or failed rear bearing.

Edited By Martin Harris on 20/03/2020 14:09:37

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:48:21

[Former Member]

[This posting has been removed]

[Dickw]

I must admit to being a bit confused by this discussion, but then I am an electrical engineer not a mechanical engineer. I would have thought that which bearing takes the thrust is a design decision not a universal principle.

I have attached a simple drawing and would be pleased if someone could explain why it might be wrong.

From memory of my pre-electric days I had a number of "single ball-race" engines with plain front bearings in which the rear bearing must have handled the thrust. That could be changed by adding a front "ball-race" but it surely doesn't have to does It?

Dick

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:49:27

[Martin Harris - Moderator]

A well made point, Dick - although you'd struggle to find a ball raced engine with a plain front one these days.

Perhaps the thrust load could be opposed by a front bearing retained by a circlip but the majority, if not all current engines simply retain the front bearing with an interference fit. I believe that the front shoulder/rear bearing locations are designed to allow enough end float in the crankshaft to avoid any preload on the bearings.

This is the precision tolerance referred to by Jon - to prove it, grab any cold ball raced engine and with a finger across the driver to crankcase joint, pull and push on the propeller - you will feel movement unless the crankshaft is rusted to the rear bearing!  If the front bearing was controlling the end float then it would have to be moving in its housing...I've just tried it on 3 engines and I'm not strong enough to exert anywhere remotely approaching the insertion pressures referred to by Jon!

Edited By Martin Harris on 20/03/2020 16:14:57

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:49:48

[Steve Hargreaves - Moderator]

Do we need to more carefully define "thrust" perhaps?

Our natural inclination is to think of the "thrust" produced by the prop but "thrust" in a bearing sense would be through the bearing from the inner to the outer race would it not? Effectively at 90 degrees to the central axis of the crankshaft.

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:50:13

[Jon H]

Ok...one last time before corona kills me (not the virus, its looking like im going to starve to death thanks to panic buyers)

The front bearing carries all propeller thrust loads in dual ball raced engines.

Dick has nailed it with his two drawings. Both are correct and only the lower one applies to model engines. The top one is the only way you could make a rear loaded setup without crushing the bearings as you tighten the prop. As you can see there is a clear step ahead of the front bearing. This means the prop nut forces tighten down on this step. The crank can then move fore/aft as required and the load is taken by the rear bearing. Model engine cranks are not this shape.

The lower drawing is essentially the shape of all model cranks. The step is behind the front bearing so the prop nut forces grab it. Once the front bearing is pressed into the crankcase it cant move, it is absolutely fixed solid. This means that you would need the distance between the two bearings, and the distance between the crank journals to be absolutely identical. If they were, the two bearings would share the load. However, aluminium expands more than steel so that would all go out the window when the engine was hot. This would load the bearings and destroy them. so, we deliberately make the crank too long and carry all the load on the front bearing.

To consider Steve's point about heating crankcases. Its valid, and completely true. Front bearings always come out easier than rear bearings. The reason is that in general rear bearings have a great deal more metal around them than front bearings and so its more difficult to heat them to the same temperature. Front bearings tend to be in nice round casting's that you can heat all over. Rear bearings are buried in a case with cooling fins one way, chunky mounting lugs another way, and thin material in another direction. This means the crankcase distorts when heated and so it can grab the bearing.

So, a shoulder on the crank web. Yep, we have them as well. As stated it prevents the crank hitting the outer race as that would be bad. But, that shoulder on the web will not touch the rear bearing, there will be a small clearance. And by small we are talking thousandths of an inch. In an ideal world expansion of the engine when hot would pull the rear bearing tight and it would share load with the front, but thats a really dodgy thing to aim for as getting it wrong would lead to mashed bearings in no time. In any case, that 70 crank has a mile of clearance between the shoulder and the bearing inner race. Other engines have less of a clearance, but its still there.

End float. This is caused by play in the front bearing. Lots of float is a sign the bearing has had it.

 

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:51:27

[Dickw]

Posted by Martin Harris on 20/03/2020 16:01:40:

A well made point, Dick - although you'd struggle to find a ball raced engine with a plain front one these days.

......................................................

Edited By Martin Harris on 20/03/2020 16:14:57

Memories, but still available

Dick

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:52:15

[Scott Edwards 2]

 

The more I read on this the more I realise how little I understand, and how grateful I am that other people do !

Jon - I'm only around the corner from you mate if you need milk for a cuppa 😁👍

Edited By Steve Hargreaves - Moderator on 20/03/2020 19:45:03

[Martin Harris - Moderator]

Posted by Jon - Laser Engines on 19/03/2020 16:57:52:

Also, if the engine was not designed to put all the load on the front bearing tightening up the prop would also load up the bearings and you are likely to lock them solid.

Rear bearings take the loads of the crank and piston flying about but do not take any of the thrust loads of the engine.

I'm genuinely interested in understanding this hypothesis so I do need to question your explanations.

I don't question that the propeller nut locks the driver to the crankshaft and the central race of the bearing is locked solidly to the flange at the front of the crankshaft. That bearing cannot move backwards as it is in a blind bore. The distance between the bottom of the front and rear bearing housing is fixed - let's say for the sake of argument it's 50 mm. There is a flange in front of the crank web to ensure clearance between the web and the bearing and provide an accurate seating for the rear bearing. If the distance between that flange and the front flange is 42.1 mm and the rear bearing is 8 mm thick then there will be a 0.1 mm clearance [I don't know what the real world figures would be] allowing the prop nut to be tightened as much as you feel fit.

We now have an assembled bottom end that turns freely. There is no positive retention method for the front bearing in the crankcase - but it can't move significantly as it is firmly locked to the crankshaft which cannot move forward because the rear bearing can't move forward. The thrust load wants to pull the crankshaft out but the above makes this physically impossible. The load path is through the crankshaft to the interface between it and the rear bearing which is transferred to the face of the rear bearing housing via the balls and the outer race of the bearing. I've yet to find a roller bearing inside an engine!

I can see no way that this load - other than a fairly insignificant frictional resistance between the outer edge of the bearing and its bore, especially at running temperature - can transfer to the crankcase via the front bearing, whose function is to maintain concentricity and absorb out of balance rotational and gyroscopic loads.

[Stephen Smith 14]

Rather you than me Jon trying to explain this, simple concept but it looks to me like your banging your head on a brick wall. What's so difficult to understand?