Srtaight Cut drops V Standard Helical ?
Turbo Mini Forums

Right, last thought before I go and do something useful....

Robert suggested calculating the safety margin for thrust load from oil film strength and current washer size.....

I've just thought, action and reaction being equal and opposite, the gearbox input bearing sees exactly the same side load.

It's an automotive special so not in my SKF book but physically is half way between a 6305 and a 6306.

Average value C(o) from the two is ~25,000 N

Apply SKF's rule and C(axial) is ~12,500 N

Makes some of the others worth a thought now.

EDIT - anyone had the input (1st motion shaft) bearing fail on a high torque motor ??? It's usually the third motion shaft one that goes I think, not the input one.

Edited by Rod S on 8th Oct, 2008.

Schrödinger's cat - so which one am I ???

You could just use a bloody good oil like castrol r for a decent film between the gear/washer. MoS2 is not as good as it could be here.

How's this for (probably a very crap idea) but one you could have a go at doing........

1. Machine half of the helical gear off.
2. Get the opposite made of the bit you have machined off.
3. Etch the 2 surfaces somehow
4. Grub screw them together with some seriously high quality grub screws.
5. Install your double helical drops.

Probably a crap idea!!!!!

I think double helical gears are still very difficult to machine by conventional methods, you end up running the cutting head into the other side....if you get me.

Since you hardly need the thrusts, how about slimming them right down and making the drops wider so they have more tooth area.

I can't say I've seen too much tooth wear on a SC set.

First gear on the laygear is a different matter though........

Bugger off, I'm getting there.

PMSL.

if you turn your fan round does it blow the other way?

Edited by Joe C on 8th Oct, 2008.

Stu,

What about the primary gear?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I might be getting mixed up but I think all this machining work would work out very expensive I know the last item I had ground with a special tool post grinder worked out hellish dear, why not go with a 7000 series bearing that fits the already ground idle gear shaft and maybe just a small amount of machining to the drop gear housing and gear box after all there alloy and will machine very easily??

I'm just thinking out loud..... !!!

Assuming we both mean the same thing by 7000 series (angular contact ball bearings) they are very wide and large in diameter at the benchmark size I quoted, so much so that I don't think there is enough alloy to machine. You would certainly have to remove the existing cylindrical needle roller bearing and then the 7000 would be carrying the radial loads too so it's axial capacity is significantly reduced.

Any machining of the alloy is going to weaken it and increase flexing, so my initial thoughts were "what could you possibly do to the gear and leave the alloy alone?"
The flat needle roller thrust bearings were the only thing I could think off that might meet that criteria.

But these are all just ideas being floated - whether any of them can be made to work remains to be seen.

If it was going to be easy, someone would have alreday done it by now !!!

Schrödinger's cat - so which one am I ???

Why not fit tapper rollers? seeing as we have to shim the idler gear any how you can use the same method to shim-up tapper rollers to apply the correct loading to the races when every thing is torqued down. so long as you use the same gaskets you would require only minimul adjustment if any at each rebuild.

Own the day

taper roller is a very interesting option and with very little machining they should fit,

Rod you spot on sorry if the last post seemed a bit dismissive, Its just that I have tried machining gears in the past with very little success lets keep on this I feel we are making good progress to a real solution here

thank you one and all for your input keep it up,

surly the existing solution is the timkin idler gear?

Any one looked at the RPM potential for these thrust roller bearings?

Own the day

Schrödinger's cat - so which one am I ???

interesting stuff guys!

didnt turbo harry do a flat roller bearing arangement on his?

also Rod I would assume that speed rating is in a sealed enviroment where there is a few cc of oil. with the constant wash of oil in the gearbox I would think they would stay plenty cool enough to up the speed significantly.

Why not fit a chain?

You guys are barking up the wrong tree on many occasions here, but I thought I'd see where this post went before jumping in.

If you have a set of three helical gears (a driver, idler and driven) with a 1:1 ratio, all positioned such that their axis are in the same plane, then the resultant thrust force at the idler is zero.
That's right - if you apply torque to the driver gear, and stall the driver gear, the idler will in theory sit there quite happily, and not move axially.

Reason is that the contact resultant force between the driven and idler pushes the idler in one direction, but the contact resultant force between the idler and driven pushes the idler in the opposite direction with the same force - hence cancelling it out.

Now, there is a large 'tipping force' at the gear, which increses with torque, ratio between the driver / driven gears and idler, and of course the helix angle.

This force in the situation described with the axis on the same plane translates to radial forces on the idler bearings - equal and opposite force on the two idler bearings....

Note though, that the axial force on the driver gear IS being applied - hence for helical gears, you are actually pushing the crank against the thrust washers in the block. That is a fact - and probably something you have never seen before - I've never, ever heard anyone else state this (although I'm sure there are a select few who have always known this - naturally)...

Same applies to the driven gear - the deep groove ball bearing on the trans input (AKA first motion) has a significant axial force as a result of the drop-gear helix angle...

BUT!!!! Note that on a 'factory' helical trans, the headset drive gear (the gear that drives the laygear) has a 'similar' opposed helix angle, and hence goes a long way to cancel out the thrust force from the (helical) idler gear.
Hence, as regards the single row-ball bearing at the first motion shaft - this has a theoretical lower loading when the drivetrain is all helical, or all SC - and a higher overall loading when SC and helical are mixed.


So, in an ideal world of minis, there shouldn't be a problem with the idler in helical form
However, there are other factors at work as regards the idler train in an A-series:

1) The idler / drop-gear train (driver, idler, driven) do not have their shaft axis all on the same plane - the idler is offset. What this does is create forces that do not 'cleanly' cancel each other out. Instead, it applies a reaction that does indeed thrust the gear (axially) in one direction or other - depending on the direction of the applied torque (drive or coast). Note though - this NET FORCE IS significantly less than the axial force on the driver or driven gears as it is generated from two 'nearly' equal and opposite forces, instead of one force.

2) The idler gear is larger than the driver or driven gears, hence the axial force at the 'top' and 'bottom' teeth on the idler apply a greater tipping force to the idler than if the gears were all the same size. More tipping = higher loads on the (conventional) idler bearings.

3) The aspect ratio of the idler gear relative to the bearing contact points on the idler shaft is pretty high, which basically means if the bearings were furhter apart, the loads on them would be lower (simple leverage).


So, what does this all mean? Well, looking at the idler alone DOES NOT answer all the points that need addressing, The whole picture needs looking at. suffice to say though, that the axial resultant forces are less than might be expected - the radial combined forces are also higher than you might first imagine. As regards needle/roller thrust bearings being used instead of thrust bearings - these themselves are not wholly efficient as they have a significant sliding of the rollers (rollers want to roll in a straight line, not round corners). Better? Potentially - but they need to be designed and specc'd correctly. Too little load on a thrust bearing will result in damage of rollers sliding ather than rolling.


Three final thoughts:


As regards (helical) idlers falling apart under high load - Here is a thought. How good would SC gears look after 'OE' levels of mileage / duty cycle. In almost all cases, i'm sure they'll be worse. Do you REALLY know the history of your parts?

For the record - I don't believe that 'helical gears' consume more power than straight-cut; especially when you look at the course, crude gear forms that are used on drop gears.

I feel this is very much an un-proven statement from folks who don't fully understand. In theory, when assembled into a driveline, a complete set of SC drops, main gears, and FD
'could' consume less energy as a full helical as they are transmitting only radial rotation (torque), wheras helical will always have a residual thrust that needs to be accounted for - somewhere...

Final thought: The worse thing you can do to a bearing short of giving it no oil, is giving it dirty or contaminated oil. In our industry we see oil straight out of a drum (even quality synthetic oil) that is of a worse ISO cleanliness than specified. Your engine filter has an uphill battle from the outset.


Incidentally - I'm welcome to comments / discussions on the above. I'm far from an expert in transmission design - but it is what I've fortunately in a position to do at work, including competitive analysis of many other automotive transmissions, so see everything in perhaps a slightly larger picture. I don't know it all though - so digest what I've put above, think about it a little more, then carry on this great topic. :)

Edited by TurboDave16V on 8th Oct, 2008.

You guys are barking up the wrong tree on many occasions here, but I thought I'd see where this post went before jumping in.

If you have a set of three helical gears (a driver, idler and driven) with a 1:1 ratio, all positioned such that their axis are in the same plane, then the resultant thrust force at the idler is zero.
That's right - if you apply torque to the driver gear, and stall the driver gear, the idler will in theory sit there quite happily, and not move axially.

Reason is that the contact resultant force between the driven and idler pushes the idler in one direction, but the contact resultant force between the idler and driven pushes the idler in the opposite direction with the same force - hence cancelling it out.

Now, there is a large 'tipping force' at the gear, which increses with torque, ratio between the driver / driven gears and idler, and of course the helix angle.

This force in the situation described with the axis on the same plane translates to radial forces on the idler bearings - equal and opposite force on the two idler bearings....

Note though, that the axial force on the driver gear IS being applied - hence for helical gears, you are actually pushing the crank against the thrust washers in the block. That is a fact - and probably something you have never seen before - I've never, ever heard anyone else state this (although I'm sure there are a select few who have always known this - naturally)...

Same applies to the driven gear - the deep groove ball bearing on the trans input (AKA first motion) has a significant axial force as a result of the drop-gear helix angle...

BUT!!!! Note that on a 'factory' helical trans, the headset drive gear (the gear that drives the laygear) has a 'similar' opposed helix angle, and hence goes a long way to cancel out the thrust force from the (helical) idler gear.
Hence, as regards the single row-ball bearing at the first motion shaft - this has a theoretical lower loading when the drivetrain is all helical, or all SC - and a higher overall loading when SC and helical are mixed.


So, in an ideal world of minis, there shouldn't be a problem with the idler in helical form
However, there are other factors at work as regards the idler train in an A-series:

1) The idler / drop-gear train (driver, idler, driven) do not have their shaft axis all on the same plane - the idler is offset. What this does is create forces that do not 'cleanly' cancel each other out. Instead, it applies a reaction that does indeed thrust the gear (axially) in one direction or other - depending on the direction of the applied torque (drive or coast). Note though - this NET FORCE IS significantly less than the axial force on the driver or driven gears as it is generated from two 'nearly' equal and opposite forces, instead of one force.

2) The idler gear is larger than the driver or driven gears, hence the axial force at the 'top' and 'bottom' teeth on the idler apply a greater tipping force to the idler than if the gears were all the same size. More tipping = higher loads on the (conventional) idler bearings.

3) The aspect ratio of the idler gear relative to the bearing contact points on the idler shaft is pretty high, which basically means if the bearings were furhter apart, the loads on them would be lower (simple leverage).


So, what does this all mean? Well, looking at the idler alone DOES NOT answer all the points that need addressing, The whole picture needs looking at. suffice to say though, that the axial resultant forces are less than might be expected - the radial combined forces are also higher than you might first imagine. As regards needle/roller thrust bearings being used instead of thrust bearings - these themselves are not wholly efficient as they have a significant sliding of the rollers (rollers want to roll in a straight line, not round corners). Better? Potentially - but they need to be designed and specc'd correctly. Too little load on a thrust bearing will result in damage of rollers sliding ather than rolling.


Three final thoughts:


As regards (helical) idlers falling apart under high load - Here is a thought. How good would SC gears look after 'OE' levels of mileage / duty cycle. In almost all cases, i'm sure they'll be worse. Do you REALLY know the history of your parts?

For the record - I don't believe that 'helical gears' consume more power than straight-cut; especially when you look at the course, crude gear forms that are used on drop gears.

I feel this is very much an un-proven statement from folks who don't fully understand. In theory, when assembled into a driveline, a complete set of SC drops, main gears, and FD
'could' consume less energy as a full helical as they are transmitting only radial rotation (torque), wheras helical will always have a residual thrust that needs to be accounted for - somewhere...

Final thought: The worse thing you can do to a bearing short of giving it no oil, is giving it dirty or contaminated oil. In our industry we see oil straight out of a drum (even quality synthetic oil) that is of a worse ISO cleanliness than specified. Your engine filter has an uphill battle from the outset.


Incidentally - I'm welcome to comments / discussions on the above. I'm far from an expert in transmission design - I'm generally just adopting a sheep scaring gaylord a position at work, including competitive analysis of many other automotive transmissions, so see everything in perhaps a slightly larger picture. I don't know it all though - so digest what I've put above, think about it a little more, then carry on this great topic.

Edited by TurboDave16V on 8th Oct, 2008.

Team www.sheepspeed.com Racing

Bollox, dave you beat me to it, That wasn't there when I posted. I was going to say exactly the same thing! Great minds think alike and all..

Team www.sheepspeed.com Racing

I agree Jim, it is very close what we both said. Strange that.

OK,

I'll plead guilty....

For the no axial load on the middle one I said it was hard to get my head around but superficially I thought the reverse rotation and opposite helix angle would create axial load, not balance it - drawing it out properly I see the middle one is neutral and only the bottom one has opposite axial load to the top.

And as for ignoring the first motion gear helix on the opposite side of the input bearing when thinking about it axial loading.... no excuse !!!

So two main comments/thoughts.

1 - Input shaft (first motion shaft) bearing - if by nature of the opposite helixes either side it is partially balanced axially (can't be exact balance because the number of teeth are different ???) surely swapping either the drops or the box to straight cuts (but not both) should create a significant axial load and lead to problems with that bearing.

2 - Idler gear - if the only overall thrust is generated by the three gears not being in a straight line (and as such is small) and the predominant loading is radial in the supporting cylindrical bearings from the "tipping" effect, why is it the thrust faces that suffer and not the needle rollers ??? Did Leyland increase the diameter of the shaft and needle rollers when they increased the helix angle from A to A+ because of increasing this "tipping" load and did they sacrifice a small bit of thrust area as a result ???

Other thoughts

I agree flat needle roller thrust bearings are far from ideal as the two ends of the needle are trying to be driven at different speeds (or they want to go in a straight line as you put it) - presumably this is the main reason they have a much lower speed rating than a similar sized taper roller - but are they worth any further consideration ??? Has anyone tried them ???

What do you think about pressure feeding the (plain) thrust washers ??? Would it compromise the cylindrical needle roller bearings with excess oil in that area ???

In short, are there any sensible improvements (tried and tested or even just theoretical) for someone who want to retain helicals ???

Schrödinger's cat - so which one am I ???

Edited by TurboDave16V on 8th Oct, 2008.

To expand on a way of preventing 'overfeeding' oil, the simplest way is to basically fill a chamber adjacent to the bearing, and then allow the oil to flow out of the top of the chamber. Result is zero pressure, and sufficient head to keep lubrication in there.

That is pretty simplistic, but a method found in many applications.

Thanks for all that Dave, it got rid of a few misconceptions I had and made me realise I don't alway think straight...

Schrödinger's cat - so which one am I ???

Well done guys. Excellent couple of posts. I'm added to!

Bugger off, I'm getting there.

Well, thinking about this last night - I realised I gave bum info about the 1st motion bearing... this must be rated quite happily for the axial thrust from the input (transfer) gear alone, as when driving in 4th, there is no torque being transmitted to the laygear - hence this bearing MUST be rated for 85%+ duty (or whatever duty cycle cars use) in 4th gear. Hence, running a SC box and helical drops would make very little difference to this bearing in reality.

Another thought - dangerous I know, but here goes....

Understanding what causes the thrust wear when using helical drops doesn't stop it happening - it's good to have had a few misconceptions put right but can the wear be stopped ???

Dave's already given his thoughts on forced lubrication (4/5 posts above) and I've given up on the idea of needle roller thrusts (speed rating and rollers shouldn't be forced to go round corners), others have suggested alternative bearings but all would need machining somewhere to fit, so what about the standard arrangement ???

It occurs to me that every time I take one apart, the thrust washer is allways "stuck" to the idler gear face, I've never seen one left on the transfer or gearbox case. In fact I usually have to prise them off (surface tension of the oil film?) to be able to measure them.

I realise you can't equate a static situation to the dynamic one, but this at least implies that most of the bearing rotation is between the washer and the alloy housings, which then wear.

So what if we pinned the thrust washer to the alloy and forced the movement to be between the washer and the gear ???

First thoughts are hard against hard (even with an oil film) is bad, but hard against soft is good, after all, a lot of overhead cams run direct in alloy cylinder heads and don't wear - but would the two hard surfaces survive ???

Or take it a stage further - how about replacing the current hard steel thrust washers with purpose made bronze ones and pinning them to the alloy housings ??? Would that not be a replica of the crank thrust washers ???

And finally, if the wear is because of "tipping" of the gear (albeit within the confines of the cylindrical needle rollers) and the slight misalignment of the flat washer face so induced accelerates the wear, what about a two part spherical washer to allow for random misalignment ???

Schrödinger's cat - so which one am I ???