12

This great question discusses the disadvantages of a lighter flywheel:

  • More twitchy to drive
  • Shorter window of opportunity to complete gear changes
  • Possible stalling at low RPM

It seems to me that the disadvantages could all be resolved in modern engine control systems by modulating the fuel and FBW throttle positions. Why isn't this done? The lighter fly wheel would lead to increased engine performance and fuel economy.

  • 1
    So your idea would be to use software to make a lightweight flywheel feel like a normal one to the driver? Sounds like an interesting idea. Looking forward to people's thoughts on this. – JPhi1618 Apr 21 '16 at 14:00
  • 1
    My guess would be that it's not a trivial thing to do in software and is probably not worth it considering the main advantage for road cars would be insignificantly lower weight. – I have no idea what I'm doing Apr 21 '16 at 14:07
  • My car has a dual mass flywheel which combines the best of both worlds plus shields the transmission against shock damage. Fitting a fixed mass flywheel on a car like mine would rapidly destroy the gearbox. A lightened flywheel may work well on a race car that travels a few thousand miles between engine and gearbox rebuilds / changes but a road car has to run for many tens of thousands of miles before major overhaul work. – Steve Matthews Apr 21 '16 at 14:37
  • The disadvantages can be resolved by a good driver. – jedd.ahyoung Apr 21 '16 at 20:24
  • 1
    I'm confused by most of the commentary here. I have driven two different Datsun 280Zs with both stock and lightened flywheels (9lb vs 23 stock). I recall that the car idled high after installing the flywheel, which considering the lighter flywheel having less resistance to rotation (it spins up and down more quickly) made sense to me at the time. I never experienced any issues with "shorter shift windows". I did stall it a few times, but so what? The rest of the engines were basically stock except for blueprinted heads and headers. I drove on the street, track days and autocross. – Tim Nevins Mar 28 at 19:05
5

The hardest part of a lightweight flywheel is getting used to taking off. Its like learning to drive a stick again. You need to rev it a bit more, or you will stall. There may be some 'Anti stall technology' you can use, especially with drive by wire. The shorter shift window pairs nicely with a short throw shifter. Again, takes a bit of getting used to.

The ECU should be able to compensate some of the 'jerkiness.' When going on/off the gas, there is less momentum, so the RPM changes more rapidly and causes a jerk. The ECU could open the throttle a little when coming off the gas, or open it slower when getting on the gas.

  • But commanding the throttle to stay open longer when the driver is commanding less throttle is a safety hazard. If the driver is in an emergency stop and the ECU is commanding higher throttle just to keep the engine running at idle you have to take the car out of gear or you lengthen the stopping distance. And that's very bad for safety. – cdunn Apr 22 '16 at 11:32
  • I am not talking a lot of throttle, more like the amount of throttle a fast idle valve would be able to supply. I am pretty sure most cars with a manual transmission do this already to keep the RPM up for a second to smooth out shifting and to help people from burning out clutches when downshifting. I remember seeing that this was in the 80's dodge ECU. It could also check the brake pedal switch. – rpmerf Apr 22 '16 at 12:57
8

Well for one thing the mass of the flywheel is what it is because manufacturers have to strike a balance between performance and drive-ability and emissions figures. A lighter flywheel would require the car to idle at higher RPM's to keep from stalling because the inertia of the flywheel is what keeps the engine firing when you're not on the gas.

So yes, it can fairly easily be done, but the reason it's not is because most consumers will not be happy with a car that idles at 1000RPM and the relevant country's ministry of transport/EPA/whatever won't like all the polarbears you're killing.

As for dual-mass flywheels. They're great and all, but they're hellishly expensive to replace WHEN (not if) they go bad. E.g. my Subaru is pretty old school with its single mass flywheel, but it would cost a third to replace it compared to the dual-mass setup in the equivalent Mazda.

  • I'm not so sure that the engine would need to idle at a higher RPM. The computer could keep the idle lower, and when a stall-suspected speed is detected (say, below 600 RPM) then the computer could open the throttle slightly. That is the whole crux of this question. – dotancohen Apr 21 '16 at 15:10
  • 1
    The flywheel keeps the engine turning at a certain speed in between each cylinder firing. If the flywheel has too little mass, a low RPM would be jerky since you could feel the engine speed up on every cylinder firing (feels like the car is about to die, running rough). Worst case, the engine doesn't have enough momentum to even get to the next cylinder. The only way to make that go away would be to speed up the engine (or add mass). – JPhi1618 Apr 21 '16 at 15:27
  • 3
    @dotancohen the results of what you are describing, the computer opening the throttle, would be a higher idle speed. – Ukko Apr 21 '16 at 15:29
  • You could get away with that if the pistons and crank were made of lighter materials. But that would get expensive and hence manufacturers wouldn't do it except for very high-end models. Things are as they are for a reason. Usually economics. – Captain Kenpachi Apr 21 '16 at 15:34
1

I think the real approach to this is not adjusting the throttle, it's by changing the valve timing to synthetically lower the compression ratio. This will decrease the opposition to the crankshaft turning, conserving its momentum just like the flywheel does. Leaving the intake valve open during the first part of the compression stroke should do it.

  • Problem with having the intake open during part of the compression stroke it creates back pressure in the intake path. It would create oscillations that change frequency with RPM. I'd have to think about it more but it seems like this would decrease the amount of air making it into the cylinder. I think the MAP sensor would pick that up but it means fuel trims would have to oscillate to keep the AFR what it should be. It seems like this would cause more problems than it would solve with a light flywheel. – cdunn Apr 22 '16 at 4:34
  • In addition, you would not be able to change the valve timing on individual cylinders without a very complicated mechanism and very precise timing and sensing. If the timing is only changed for all the cylinders you wouldn't get the benefits of opening for just one. The system would be incredibly complex to design and difficult to keep running. All to allow a lighter flywheel. With the timing changed, wouldn't you loose all the benefits of having a lighter flywheel? – cdunn Apr 22 '16 at 4:54
  • You could have a separate valve for venting the "early compression" air (sort of like the old Mitsubishi jet valves, but in reverse). The fuel mapping would have to be revised to compensate for the fact that you're expelling already-metered air, but I don't think there's any way you're going to get out of tweaking the ECU no matter which approach you use. – TMN Apr 22 '16 at 10:47
  • Tweaking the ECU is a given, and not a problem as long as it's not asked to do more than its capable of. But so far, I don't see any hard limits that would be hard to overcome. The bigger picture thing we're missing here is if it was possible to use software to make a light flywheel drive like a heavy one, then why do it at all. If you make it drive like a normal flywheel, why not just have a normal flywheel and avoid all this complexity? What would be the gain of having a light flywheel that drove like a normal one? – cdunn Apr 22 '16 at 11:21
  • I thought we were just addressing the high idle speed issue. I think OP still wants the performance advantages of a low-mass flywheel, but doesn't want to pay the cost during low-RPM operation. – TMN Apr 22 '16 at 11:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.