EVERYTHING is compliant. It is just a matter of to what degree. Your engine's crankshaft is a bunch of torsion bars. The shafts that drove Carrera camshafts are torsion bars- they even have flywheels to help stabilize the ignition and valve timing. Your car's chassis is a spring.

I understand that. It's simple physics to know that everything has compliance. I'm just saying that the pulleys/belts were probably designed in for added compliance.

I'm sure an engine crankshaft was not designed to offer a certain amount of compliance. I would want it as rigid as can be.

I think the decision was more cost and convenience of manufacture based. You can always have a stiff shaft and a guibo like on 935s or 917s.

Certainly you want a stiff crankshaft, but the 70.4mm strokers aren't quite stiff enough for 8,000 rpm sustained.

I would think that the maintainance of a critical gearbox and fatigue lifing of a shaft would be more expensive than replacing belts more frequently. Can a non-A&P change the belts? (Though the shaft would be more elegant than several stages of pulleys)

A guibo coupling will not provide slip like a pulley/belt system. I'm sure the decision was not based on cost, maybe convenience. If you've ever been involved with anything to do with aerospace, you would know that cost-over-function is an absolute no-no.

Well they both clearly work, so why would cost not then come into play? I can't think of a reason why slip would be necessary rather than just torsional compliance/damping.

I have no idea. I'm not a mechanical engineer. I'm just a mere electrical engineer. Most of the mechanical engineers I have worked with started out trying to be electrical engineers, but couldn't cut it.

You would truly have to know the mechanics of the device to determine why it was design the way it was. Many things aren't observed until it has gone through extensive testing.

Redundancy? Four belts so one can fail and no-one dies?

http://forums.pelicanparts.com/uploa...1389193312.jpg