Unusual ways to change gear, cont'd

"Cé hé sin" (michael-m-mouse)
08/28/2018 at 18:25 • Filed to: Beier variable drive

Latest in my series of unusual and often forgotten (and often deservedly so) means of varying ratios is the Beier variable ratio gear, or CVT as we’d call it today. It was designed by a Herr Beier, an Austrian gent whose name would therefore rhyme with Mayor, more or less.

Here’s a diagram, which unfortunately tells us absolutely nothing without a bit of explanation.

We’ll start by looking to the right. The sketch shows two shafts, each of which is equipped with several discs. The lower shaft provides the output and has discs which are able to move axially and are pressed together by the spring visible to the right. Each disc is machined so that the rim is its thickest part. In section then the rims look somewhat like jaws. The upper, input shaft has discs which interleave with the output ones and which are slightly conical in section with surfaces which are about 3 deg to the vertical. To proceed the input shaft is brought closer to the output shaft so that its discs force themselves between the output discs and come into contact with their raised rims. The output discs, we remember, are able to move along their shaft against spring resistance. The input discs are now able to turn the output discs, initially at approximately a 1:1 ratio. Now we can move the input shaft further towards the output, as shown on the left hand side of the shaft. The input discs still maintain contact with the raised rims of the output discs, but the contact point has now moved closer to the input shaft and the effective diameter of each input disc has now decreased. In simple terms a small wheel is now turning a bigger wheel and so reduction gearing has been set up. Continue to move the input shaft closer and the gearing reduces still further.

As each disc can transmit only very limited torque there are many discs and (not shown in the diagram) typically three input shafts clustered around the output shaft.

What’s not obvious is that the discs were not in mechanical contact with each other as oil jets endured that there was a thin film of oil between them which transmitted the torque.

Beier drives are used with electric motors, and more exotically they featured on the Napier Nomad two stroke opposed piston turbocompound diesel (with afterburning) aircraft engine from the 1950s.

The Beier drive was used on buses early in the 20th century but so far as I know nobody has tried it for automotive use since except Citroen who designed their own take on it in the early seventies with the aim of using it on the GS. Nothing came of it.

DISCUSSION (10)

RacinBob > Cé hé sin
08/28/2018 at 18:38

My guess is the friction and wear were problems too difficult to overcome.....

Cé hé sin > RacinBob
08/28/2018 at 18:45

Actually, no (and I forgot to put that in!)

The discs didn’t have any mechanical contact. Torque was transmitted by a film of oil, just as in a modern (steel) belt and pulley CVT. Slip page was about 1 or 2 % and efficiency about 90%.

RacinBob > Cé hé sin
08/28/2018 at 19:16

Well isn’t that interesting. I wonder what it’s limitation was, though 10% losses may mean a lot of heat so maybe that was it. Also 10% is a ton to give away in the efficiency world.....

50ford500 > Cé hé sin
08/28/2018 at 19:31

I wonder what the power limitations are for it though? It's not like early 20th century vehicles were known for extreeeeme power output

bhtooefr > Cé hé sin
08/28/2018 at 20:12

For some reason I’m reminded of toroidal CVTs - I feel like a lot of the same principles are at play, it’s just that toroidal CVTs didn’t need clamping force. (And, having to deal with clamping force affects shift behavior, so is that the problem, maybe?)

Cé hé sin > 50ford500
08/29/2018 at 04:51

From what I can find Sumitomo offer electric motors with Beier drives up to 200 hp.

Cé hé sin > RacinBob
08/29/2018 at 05:04

How efficient are conventional belt and pulley CVTs though? As I understand it they’re worse than torque converter automatics, let alone manuals.

Cé hé sin > bhtooefr
08/29/2018 at 05:07

M aybe so. They’re used up to about 200 hp with electric motors, but I haven’t been able to find why they’re not used in the autom otive business. Efficiency of 90% may no t sound great but I doubt if it’s worse than a belt and pulley CVT.

RacinBob > Cé hé sin
08/29/2018 at 07:59

We ought to find that out. I just think about the 10% loss as all going to heat. at 750 watts per HP. lets say the box is transmitting 30 HP, that is 2.2kw heat which is not insignificant.

RacinBob > Cé hé sin
08/29/2018 at 08:22

I assume they are pretty high considering AT’s have lock up torque converters and dual clutch are lockup so they should be in the high 90's. I can’t imagine any designer spotting themselves 10% with a transmission if efficiency is the goal.

I do note that it appears that CVT’s do have external coolers. So I could be wrong. I don’t know if I believe it but s ee below.

https://www.physicsforums.com/threads/cvt-efficiency.304935/

The following web site has a good discussion of the various types of continuusly variable transmissions (CVTs) http://en.wikipedia.org/wiki/Continuously_variable_transmission None of the CVTs have efficiencies as high as manual transmissions: about 96% in all but the 1:1 gear which is about 98%. Manufactirers of CVTs will not disclose the efficiencies of their designs; it is proprietary. Note that manual transmissions in standard automobiles do not require water cooling. If a particular CVT requires water cooling, that is proof that it is less efficient than a manual transmission.

Reference https://www.physicsforums.com/threads/cvt-efficiency.304935/