Overdriven transmission

[autodoctor911]

I wonder has anybody successfully downsized the gearbox in a race car by overdriving the input from the engine, effectively reducing the torque requirement for the standard gearset. It seems like the gear set would be able to handle less torque at higher rpms with reduced mass. Of course the inertia may remain the same, but overall mass of the car may be reduced.

I was thinking of using either a spur gear or planetary right on the engine in place of the flywheel to increase rpm, with a downsized clutch and transmission, and of course an equal reduction built into the pinion and crownwheel to turn the axles at the normal speed relative to engine rpm.

Alternatively, a reduced mass standard crown wheel and pinion could be used and a further reduction introduced at the upright with another set of spur gears, allowing the axle mass to be reduced as well, at the expense of extra mass in the upright, but also allowing more ground clearance for aerodynamic gains with a larger tunnel/diffuser or a lower Cg if inverted allowing a low mounted gearbox.

[J.A.W.]

Yeah, its an ages old approach applied to motorcycles..
But as for F1, check the 'Inline 6...' thread topic a bit down the page - here in the E,T & C section..

[Tommy Cookers]

the OP seems to suggest driving the gearbox' (gearset) at greater then engine speed, this has not been done in motorcycles
(and the Heidegger engine appears to be driving it at engine speed ie 1:1 crank takeoff, presumably to use existing gearbox parts)

F1 gearboxes have anyway in recent years been running at very high engine speeds of c. 20000 rpm
and have moved the clutch to the input shaft (traditionally something to be avoided eg the H-16 BRM)

a gearbox designed to run at less than engine speed has less inertia (referred to the engine)
and the lower rpm also helps with dog engagement, this is what motorcycling knows well
(even shaft-drive motorcycles now have gearboxes driven at less than engine speed, if only to allow one CWP with various engines)

modern car gearboxes are all-indirect, this also could somewhat help the designer (in running at other than engine speed)

in principle you could run a 2013 F1 gearset on a 2014 car by running it at greater than engine speed
but they chose not to go that way

[-Felix-]

I always thought the Porsche 912 engine had a different number of teeth on the spur gears on the crankshaft, had a number of 30/31 on my mind, thus lowering the engine revs (although just by 30/31..). But choosing such a "small" different number of teeths is more often used to change the teeths that are driven. Because if you have the same number, the 2 teeth on each wheel interacting are always the same, which can result in damage and or problems with the gear surface because of uneven machining (you can only grind to a certain degree of tolerance of course). If you have a choice you better try to avoid that so each teeth interacts with another one each revolution, resulting in a more even wear. (Hard to explain in englisch, I hope you get what I tried to say..)

So opting for this kind of transmission is more a thing of reliability and wear because 30/31 isn't that big at all. Although I found a picture of the spur gears of the porsche engine where it looks like both are 30. Have to do some further research when I'm at home.

a gearbox designed to run at less than engine speed has less inertia (referred to the engine)
and the lower rpm also helps with dog engagement, this is what motorcycling knows well
(even shaft-drive motorcycles now have gearboxes driven at less than engine speed, if only to allow one CWP with various engines)

But if you design a transmission to run at less than engine speed you have it to take greater torque, which itself needs use of bigger shafts and gears, resulting in a higher inertia and overall weight again...

modern car gearboxes are all-indirect, this also could somewhat help the designer (in running at other than engine speed)

What do you mean by that? All modern car gearboxes are indirect?

[Tommy Cookers]

.....
So opting for this kind of transmission is more a thing of reliability and wear because 30/31 isn't that big at all. Although I found a picture of the spur gears of the porsche engine where it looks like both are 30. Have to do some further research when I'm at home.

a gearbox designed to run at less than engine speed has less inertia (referred to the engine)
and the lower rpm also helps with dog engagement, this is what motorcycling knows well
(even shaft-drive motorcycles now have gearboxes driven at less than engine speed, if only to allow one CWP with various engines)

But if you design a transmission to run at less than engine speed you have it to take greater torque, which itself needs use of bigger shafts and gears, resulting in a higher inertia and overall weight again...
TC RESPONSE - maybe

modern car gearboxes are all-indirect, this also could somewhat help the designer (in running at other than engine speed)

What do you mean by that? All modern car gearboxes are indirect?

generally modern car gearboxes don't have coaxial input and output shafts so there is no need for a 1:1 ratio (all ratios are indirect)
this all-indirect layout would/does allow some reduction or multiplication effect to be designed in without additional parts

btw - your 30/31 example is/was known in the UK as a 'hunting tooth'
to prevent some load peak in the cycle arriving on the same gear tooth every time
expected eg some places in the camshaft drive of a dohc single-cylinder engine (but little reason to have it in a gearbox drive ?)

PS - sorry this post is messy
PPS some of the various transverse GP and F1 gearboxes (from 1937) might have run at other than engine speed ?

[J.A.W.]

Was the O.P. really asking - & if the intention was to reduce the torque requirements of the transmission
gears ( & necessary mass - with issues in train) by speeding them up, via a primary reduction gear ratio?

This is certainly a practice used in motorcycles, indeed for example - the famous Kawasaki 2T 500cc triple
shared basic clutch/gearbox component dimensions with its later 750cc sibling, but the primary ratio was
reduced numerically ( sped up) from 2.41:1 in the 500 to 1.88:1 in the 750 to cope with the increase in torque..

[riff_raff]

While running gears and rolling element bearings at higher pitch line velocities will reduce the torque related forces acting on them allowing them to be made smaller and lighter, it is not good for efficiency. The viscous losses in rolling element bearings, and contact sliding and windage losses in the gears, become exponentially worse as speed/PLV increases. You should also consider that the extra gear stage required at the gearbox input will add weight and parts.

Plus, 100% of the engine output power must pass thru this gear stage. The mechanical loss from the gears and bearings in this stage are probably around 1% of the transmitted power. So if the output from the engine is 650hp, there will be a 6.5hp loss from this extra gear stage alone. Then there is also the need for a higher gear ratio at the final drive, which will increase the diameter of the single largest gear in the transmission.

[J.A.W.]

Actually r-r, transmission losses from crank shaft to driven tyre are closer to ~10%, AFAIR,
& that is with a high efficiency (~98%) chain final dive.. (& where 'hunting tooth' is a standard fit),
- for shaft final-drives, esp' those with extra 90`transmission gearing direction shifts - its even more..

However the 'purity' of trans' efficiency is often traded off..
..for a performance advantage in quick/accurate shifting - with preference for minimal losses in those areas..

[gruntguru]

2% loss per mesh is average. I would bet that F1 transmissions are a lot less than that.

[Tommy Cookers]

the (nominally 25 MW) planetary gearbox of the P & W geared turbofan is over 99% efficient
hundreds of orders have been placed for this engine product eg for the A320 Neo
RR have claimed 99% or better at a nominal 50 MW

the secret seems to be tailoring the gear design and manufacture around a rather constant load, presumably F1 etc also does this

[-Felix-]

generally modern car gearboxes don't have coaxial input and output shafts so there is no need for a 1:1 ratio (all ratios are indirect)
this all-indirect layout would/does allow some reduction or multiplication effect to be designed in without additional parts

I'm still not sure what you're getting at, I mean, there are several transmissions in modern cars with coaxial input and output shaft? Or is it that you mean, when you have for example a simple manual Inline 6 gearbox, that you have 3 shafts and when you have a direct gear for example 5th, that you have a overall gear ration of 1:1, but because you have a ratio like 12/28 from input to intermediate shaft and 28/12 from intermediate to output shaft, so you have in fact none single stage 1:1 gear ratio? Is that what you meant that you never in modern car gearboxes have a gear pair with same number of teeth?

btw - your 30/31 example is/was known in the UK as a 'hunting tooth'
to prevent some load peak in the cycle arriving on the same gear tooth every time
expected eg some places in the camshaft drive of a dohc single-cylinder engine (but little reason to have it in a gearbox drive ?)

Yes, that's what I tried to explain Of course there is no reason in gearbox, but as I said I thought it was in the central spur gear of the 912 engine or also the Heidegger inline 6.

@r-r when you have the central takeoff anyway, it doesn't add any mass if you just go for a ratio other than 1:1..

[Tommy Cookers]

now I'm not sure what I was getting at either

I just posted about the GP and F1 transverse and other remote gearboxes eg W125, W196, 50s Italian F1, 75 Ferrari F1
whether they were all at engine speed or lower (or higher) etc ?

the P&W geared turbofan at 3:1 reduction has less M of I referred to/'seen' by the rest of the engine compared to a direct drive fan
due to the reduction ratio, and despite the geared fan being much bigger than the direct equivalent
important for engine response etc

EDIT re the following post by Felix
our F1 engine has to accelerate its own mass moment of inertia (crank, rods, pistons etc) - this is a fixed amount
and a MMoI equivalent to adding the MMoI effect of the gearset on the engine - this is a variable amount
(it varies with the actual gearset MMoI and the square of any reduction or multiplication ratio applying between engine and gearset)
and a MMoI equivalent to adding the effect of the CWP, wheels, car mass etc on the engine - this is another fixed amount

[-Felix-]

but gear ratio does not have influence on inertia? Inertia is related to the body and mass distribution itself, if you drive a wheel through 10:1 or 1:1 gear ratio doesn't affect it's inertia and hence the power to accelerate it at a given rate. The torque yes, but to accelerate a rotating body at a given rate, the power stays the same always..

[gruntguru]

Yes, but the ratio between engine and wheels remains the same regardless of whether you speed up the GB input shaft or not. To put it another way, accelerating from one road speed to another in the best available gear will always result in the same change in kinetic energy for the engine moving parts and for the vehicle as a whole.

[-Felix-]

EDIT re the following post by Felix
our F1 engine has to accelerate its own mass moment of inertia (crank, rods, pistons etc) - this is a fixed amount
and a MMoI equivalent to adding the MMoI effect of the gearset on the engine - this is a variable amount
(it varies with the actual gearset MMoI and the square of any reduction or multiplication ratio applying between engine and gearset)
and a MMoI equivalent to adding the effect of the CWP, wheels, car mass etc on the engine - this is another fixed amount

That's right, I see your point. Wether it adds to a better engine response i kind of doubt, since the inertia of car mass and wheels etc. is a great times bigger than the one of the engine and gearbox itself?