Audi CVT efficiency

[wangp1283]

One of the main loss of efficiency for the variable diameter pulley (such as that used by Audi) is due to the high clamping force needed. The clamping force around 10 times the transmitting force. Coefficient of friction of the fluid is about 0.1.

According to Luk's site:

http://www.luk.de/english/Bibliothek/Download/k7/luk_k7_chap06.pdf

This loss can be up to 15%. 6% is lost in the hydraulics and 9% is lost in the variator.

I have a couple of questions.

1) One obvious way to decrease the clamping force needed would be to use a force multipler such as a gear train. Why not use it? I talked to an engineer and he said the response will decrease. So is the response of the system dependent on its displacement?

2)I recently made a design that will require only 10% of the present clamping force, everything else being equal. Can anyone give me an idea of how much the efficiency can improve? 10% of 15% is 1.5%. So instead of lossing 15% to the clamping unit, if only 10% of the clamping force is needed, then the loss can be reduced to 1.5%. However, I have a feeling the relationship is not linear.

3. Why not use an electrical mechanical actuator instead? That'll
help with the pump loss. The only reason I can think of is probably
weight and cost. How much would such a system weigh? In Audi's case the actuator must exert up to 15000 lb of force.

Thanks. I need the answer so I can finish my design.

 

[crysta1c1ear]

1) As described in the text, inadequate clamping results in belt slip which can rapidly destroy the variator. (See figure 7 PDF page 11, over-clamping, optimum range, and destructive slip.)

2) As you said yourself, the losses are 6% plus 9%, the clamping forces being the hydraulic 6%. So if you could make that 90% better and have 0.6% losses, your total would be 0.6% + 9% ie 9.6% losses. So IF you could do it, I think the losses would drop to 9.6%.

But I don't think you can do it. When you mention gearing, that's effectively leverage isn't it? You get a smaller force but have to move the lever a larger distance. Work, equals Force times Distance, stays the same.

There must be other factors contributing, since the text describes the single piston arrangement as requiring less force (oil pressure) but larger distance (flow rate) and shows the LuK double piston arrangement as being an optimisation.

 

[DesignOutsourcing]

The link above didn't work for me, try this one and look up gearbox systems in Technical Presentations:

http://www.luk.de/content/en/products/library/library.jsp

So 6% of the losses are hydraulic and 9% mechanical. As I understand it, the belt is clamped between the pulleys and it takes some force to pull the belt out of the pulleys and then pull the belt into a clamped position again on the opposite pulley. I think that this is where the mechanical losses are (not in applying the clamping forces to the pulleys.)

The system works because of friction between the belts and pulleys. Any eforts to reduce friction will also reduce the torque you can transmit through the system. They don't want to reduce the clamping force per se but better control it so that it is low when torque is low and higher when torque is higher. Hope this helps, cheers, derek


Join us at

http://groups.yahoo.com/group/opensourcecars