82 replies to this topic

[MRA]

10 to 15%? Sorry, can't say I've ever seen that much of a gain.

 

That sounds like an awful lot of power to be grinding in to the thrust washers, do you think they'd stand up to that?

 

These days with modern cars, they are trying to drag every ounce of gain and minimise every gram of losses, I've of no doubt that if the losses from Helical gears were as high as 5% they'd go to Herringbone gears to reduce the loses while keeping their drive trains quiet.

 

Of course you are only basing your theory on all the load going through the thrusts.  The thrusts work differently than bronze bearings, the main reason for bronze bearing failure is incorrect clearances causing judder, this doesn't occur on thrusts as they are under load or not depending on whether or not you are accelerating or decelerating.

 

What you are seriously forgetting is the wiping effect that helical gears have as they mesh and un-mesh, this is where a lot of your losses occur.

 

Modern production car gears are cut, heat treated then ground, this gives them a far better surface quality than any Mini gears thus a reduced friction due to better oil retention on the mating surfaces, leading to less metal to metal contact.

 

The reason you won't see herringbone gears in a production car is their cost is far too high, to machine a herringbone gear of the sizes we would need would require a larger case to fit them into as the gears would have to be wider.

 

Gears could be rolled, that would maintain the current width, but then how would you grind them round* ?  Then there is the distortion due to the heat treatment process.

 

*The forming process would distort the gears as they are being rolled and grinding is the current method to make them round again.

With herringbone gears there is currently a requirement for a radial groove (tool clearance groove) along the centreline, eg. as the left cutter cuts towards the center it needs a clearance groove to "fall in to" otherwise a wall of swarf would build up and very soon, bang goes you £2000 cutter, the same goes for the right hand cutter.

 

No, herringbone gears won't be used in production cars for some time yet, even with 3D printed gears the advantages may not be worth i

Edited by MRA, 29 June 2016 - 11:39 AM.

[Spider]

In regards to 'the wiping' effect, can you expand on what is actually occurring there, just so we can be on the same page. I think I see what you're saying, but just want to be sure we are both looking at the same acorn on the tree

 

I don't know what they did in later years (at Rover / Powertrain) for gear production, but certainly in the BMC / Leyland days, the gear teeth were flame hardened and then ground. The Crown Wheels were done differently, but still ground finished.

 

I'm of little doubt that with modern manufacturing techniques and volumes, Herringbones (or more likely a derivative thereof) could be manufactured at a cost that would fit in with car production, perhaps not at the bottom end, but certainly in the upper middle end and beyond. 

[tiger99]

As for power loss, remember that ALL of it, less maybe 1 watt of noise, ends up as heat. 1 hp = 746 watts, so if you lose 15% of the output from an 80 hp engine, that is 8952 watts with nowhere to go, so the gearbox is going to get very hot! That amount of heat is beyond the capabilities of the standard casing to dissipate.

If you thrash a tuned Mini flat out for 100 miles or so, does the gearbox casing melt? I don't think so, so the losses are likely somewhat less than 15%.

But the way to get them down is to use herringbone gears, and a chain drive to replace the drop gears. I know it is difficult and expensive, but surely someone is going to do it sooner or later? If sufficient sets of parts were made the cost might be bearable.

[Dusky]

I've started a herringbone thread before, I had a few people intrested. If I could Get more people intrested I could Get a good price for them. I got a good friend who has all the tooling to make them. But would quickly cost 500£ for a one off set of gears. More people would make it much lower quickly.

[MRA]

In regards to 'the wiping' effect, can you expand on what is actually occurring there, just so we can be on the same page. I think I see what you're saying, but just want to be sure we are both looking at the same acorn on the tree

 

I don't know what they did in later years (at Rover / Powertrain) for gear production, but certainly in the BMC / Leyland days, the gear teeth were flame hardened and then ground. The Crown Wheels were done differently, but still ground finished.

 

I'm of little doubt that with modern manufacturing techniques and volumes, Herringbones (or more likely a derivative thereof) could be manufactured at a cost that would fit in with car production, perhaps not at the bottom end, but certainly in the upper middle end and beyond. 

 

It is also known as tooth flank frictional load.

 

That's the point, it isn't possible to get the desired finish on herringbone gears currently at any price, anyone can cut herringbone gears, but not many can make them to the accuracy and surface finish requirements to give the benefits that a true herringbone gear would give, hence why they are still only used on large high powered gearboxes.

[MRA]

As for power loss, remember that ALL of it, less maybe 1 watt of noise, ends up as heat. 1 hp = 746 watts, so if you lose 15% of the output from an 80 hp engine, that is 8952 watts with nowhere to go, so the gearbox is going to get very hot! That amount of heat is beyond the capabilities of the standard casing to dissipate.

If you thrash a tuned Mini flat out for 100 miles or so, does the gearbox casing melt? I don't think so, so the losses are likely somewhat less than 15%.

But the way to get them down is to use herringbone gears, and a chain drive to replace the drop gears. I know it is difficult and expensive, but surely someone is going to do it sooner or later? If sufficient sets of parts were made the cost might be bearable.

 

Ask anyone with a SCCR kit installed and they will tell you the oil has a higher temperature, not sure how you derived your figures for melting a transmission case, but 8kw is nothing, your engine is about 25% efficient which means your cooling system has to cope with 179KW of heat !

 

Also if you are thrashing about for 100 miles then you will also have 100 miles of air removing the heat from the same transmission case.

[MRA]

I've started a herringbone thread before, I had a few people intrested. If I could Get more people intrested I could Get a good price for them. I got a good friend who has all the tooling to make them. But would quickly cost 500£ for a one off set of gears. More people would make it much lower quickly.

 

I would be interested in seeing the results.

[MRA]

I've started a herringbone thread before, I had a few people intrested. If I could Get more people intrested I could Get a good price for them. I got a good friend who has all the tooling to make them. But would quickly cost 500£ for a one off set of gears. More people would make it much lower quickly.

 

PM sent

[tiger99]

The point about the cooling is that the engine heat partly goes out of the exhaust and partly into the radiator. But the path from gearbox bearings to the outside world is rather less effective. The gearbox fins will not get rid of 8kw at speed, without a very large temperature rise. The best thing is of course a very good radiator and an oil-air heat exchanger so the gearbox oil can be cooled. That is a lot more efficient than relying on airflow over the casing. Not everyone has a system like that. Perhaps they should.

[MRA]

The point about the cooling is that the engine heat partly goes out of the exhaust and partly into the radiator. But the path from gearbox bearings to the outside world is rather less effective. The gearbox fins will not get rid of 8kw at speed, without a very large temperature rise. The best thing is of course a very good radiator and an oil-air heat exchanger so the gearbox oil can be cooled. That is a lot more efficient than relying on airflow over the casing. Not everyone has a system like that. Perhaps they should.

 

Apart from the fact that the oil removes heat from the engine, pistons etc...  really how can you be so sure an aluminium case cannot remove 8KW extra ?  where is the Engineering calculations to prove this ?

 

Quite a few people use oil coolers on Mini's and as the oil is shared it does help to reduce the oil temperature, the main reason for an oil cooler is due to the use of straight cut gears, of course more power would need more cooling as well.

[carbon]

The point about the cooling is that the engine heat partly goes out of the exhaust and partly into the radiator. But the path from gearbox bearings to the outside world is rather less effective. The gearbox fins will not get rid of 8kw at speed, without a very large temperature rise. The best thing is of course a very good radiator and an oil-air heat exchanger so the gearbox oil can be cooled. That is a lot more efficient than relying on airflow over the casing. Not everyone has a system like that. Perhaps they should.

Just a guess, but I reckon the gearbox casing will be able to dissapate more than 8kW once the oil is up to working temperature.

 

As well as 8kW from frictional losses in the gearbox, at full load the engine oil is probably picking up 20 kW or more from the crank, block, head and pistons. This compares about 80kW being lost as exhaust heat, and about 50kW into the coolant to radiator.

 

If you cover up a Mini gearbox casing with a sumpguard the oil temperatures can go very high under continuous load when no oil cooler is fitted. There's probably as much heat being lost from the gearbox casing as from a 13 row oil cooler.

[MRA]

Anyone who wants to work through the formula for losses in helical gear trains please feel free to follow the attached link....

 

http://www.geartechn...lical_Gear_Mesh

[tiger99]

Well I design electronic equipment for a living and if I could get rid of 8kw of heat with the surface area and fins of a Mini gearbox I would be delighted. But it depends on the acceptable temperature rise. On the other hand, I have given a Mini a good thrashing many times and not noticed the transmission case being hotter than the engine block.

I said a couple of years back that I was interested in instrument in a transmission with thermocouples to see where the hot spits were. Maybe I will be able to do that eventually. I would like to know gear temperatures too, but that is impracticable.

What I am sure of is that 8kw is unacceptable and is one area where there may still be scope for substantial improvement with an A series power plant. Less loss equals more performance. An electric drive, with alternator driven directly by the engine, synchronous rectification, and permanent magnet brushless dc motors driving the wheels would lose a lot less than 8kw. Possibly one reason why the rail industry uses diesel electric drive in most locomotives. There are other practical reasons too.

It would be interesting to know how the power loss in a modern car compares. It should be possible to bring the Mini up to the same level of efficiency. A chain drive instead of drop gears gains about 2% for a start, but I can't find data about running chains at 6000 rpm. Handling the torque is not a problem.

We are probably very close to the limit of reliable, efficient power output from the A series which only leaves the reduction of other losses as the means of improving performance and economy,which is why it matters. Other things to think about are rolling friction (tyres), hub bearings front and rear, and residual brake drag. All that may add up to a few kw. Aerodynamics would be next, but with too much butchery it would not really be a Mini so we can't do too much in that area except buy a Mini Marcos, GTM etc.

Coming back to the transmission losses, I wonder if conformal gears might be better than the normal involute type. They are certainly stronger, which means potentially smaller and lighter.

[nicklouse]

For chain drive from crank to gearbox look at the old SAAB 2.3 L engines. Triplex drive chains.

[MRA]

Well I design electronic equipment for a living and if I could get rid of 8kw of heat with the surface area and fins of a Mini gearbox I would be delighted. But it depends on the acceptable temperature rise. On the other hand, I have given a Mini a good thrashing many times and not noticed the transmission case being hotter than the engine block.

I said a couple of years back that I was interested in instrument in a transmission with thermocouples to see where the hot spits were. Maybe I will be able to do that eventually. I would like to know gear temperatures too, but that is impracticable.

What I am sure of is that 8kw is unacceptable and is one area where there may still be scope for substantial improvement with an A series power plant. Less loss equals more performance. An electric drive, with alternator driven directly by the engine, synchronous rectification, and permanent magnet brushless dc motors driving the wheels would lose a lot less than 8kw. Possibly one reason why the rail industry uses diesel electric drive in most locomotives. There are other practical reasons too.

It would be interesting to know how the power loss in a modern car compares. It should be possible to bring the Mini up to the same level of efficiency. A chain drive instead of drop gears gains about 2% for a start, but I can't find data about running chains at 6000 rpm. Handling the torque is not a problem.

We are probably very close to the limit of reliable, efficient power output from the A series which only leaves the reduction of other losses as the means of improving performance and economy,which is why it matters. Other things to think about are rolling friction (tyres), hub bearings front and rear, and residual brake drag. All that may add up to a few kw. Aerodynamics would be next, but with too much butchery it would not really be a Mini so we can't do too much in that area except buy a Mini Marcos, GTM etc.

Coming back to the transmission losses, I wonder if conformal gears might be better than the normal involute type. They are certainly stronger, which means potentially smaller and lighter.

 

Certainly the more heat you keep above the piston the better.

 

I think the main issue with chain drive is speed over packaging, ie size of the sprockets..  not saying it can't be done, but would require some sideways thinking.

 

I think the overhead catenary might be challenging ;)

 

Reliable efficient power ?  I have heard this for 30+ years, there is much to come, just add technology and stir !

 

Conformal gears are expensive unless high volumes are being made, they are currently being used in certain automotive applications, but that's about it, tooling is twice as much as for involute gears, it could be a way forward, even twice the tooling isn't so bad as it means it should last twice as long, initial tooling will be expensive but long term it will even out.