Intake vacuum energy recovery

[JRMichler]

Those of us who understand ecomodding eliminate that source of loss by:

1) Shifting to a higher gear to reduce manifold vacuum.
and
2) Shutting the engine off and coasting.

That gives us the gas mileage benefit of a hybrid without the cost of a hybrid or the complexity of an intake manifold vacuum energy recovery system.

[teoman]

Originally Posted by Frank Lee

the concept is valid.

That is how Lolo Ferrari died. Her enhanced breasts put a burden on her lungs and she died in her sleep.





Same may apply to a car

[serialk11r]

Roots blower won't do it, the efficiency is too poor. I had this idea before (two Roots blowers, bypass one of them and shut off the air supply to turn the second one into a turbine, open the bypass to provide supercharging).

Given that this happens at very low load, you need a really tiny turbine to provide sufficient restriction.

It's far easier to just avoid the power loss by increasing EGR.

[Ecky]

Aside from reducing RPM and increasing engine load (easily the best way to improve this), I would think another way would be to just reduce the volume on the other side of the throttle plate(s). One could:

1) Reduce plenum volume
2) Reduce runner length/volume
3) Advance the intake cam considerably, so the valves close early, meaning more of the vacuum production would be inside the cylinder and less in the intake manifold (smaller volume)
4) Individual throttle bodies (ITBs)

[freebeard]

https://en.wikipedia.org/wiki/Scuderi_engine

The Scuderi engine, formally called the Scuderi Split Cycle Engine, is a split cycle, internal combustion engine invented by Carmelo J. Scuderi (April 13, 1925 – October 16, 2002).[1] Scuderi Group, an engineering and licensing company based in West Springfield, Massachusetts and founded by Carmelo Scuderi's children, is testing a working prototype of the engine that was officially unveiled to the public on April 20, 2009.[2][3][4]

The inventors heirs rode this one into the ground. I think it could have worked as a box four. They sunk all their investment capital into a proprietary inline four block, that wasn't modifiable for test purposes.

[cRiPpLe_rOoStEr]

Originally Posted by Ecky

Individual throttle bodies (ITBs)

It seems to be a good approach, but in this bean-counting era I wouldn't hold my breath to see it becoming available in any newer production car with a naturally-aspirated engine. Assuming it's unlikely to work in a turbocharged engine.

[serialk11r]

Originally Posted by Ecky

Aside from reducing RPM and increasing engine load (easily the best way to improve this), I would think another way would be to just reduce the volume on the other side of the throttle plate(s). One could:

1) Reduce plenum volume
2) Reduce runner length/volume
3) Advance the intake cam considerably, so the valves close early, meaning more of the vacuum production would be inside the cylinder and less in the intake manifold (smaller volume)
4) Individual throttle bodies (ITBs)

1,2,4 are all basically the same thing. Only 4 would have a noticeable effect, and it would still be small. Once you have a plenum, there's a big enough path to other cylinders that the pressure becomes fairly stable. ITBs only trim a small % of the pumping losses (e.g. at runner volume = 1/2 cylinder displacement I'm estimating around 7% pumping loss reduction at typical idle, but that number gets better if you can drop the runner volume more). They also don't reduce pumping loss as much if you have LIVC.

3 has the obvious problem of residual exhaust. Every previous gen DI engine used internal EGR and were limited by the ~20% charge dilution limit. If you don't care about high end power, you can regrind the cam for very little duration to get EIVC, but it's better to go the opposite way and use LIVC like every OEM does, since that preserves top end power, and reduces pumping loss.

[Ecky]

I'm fairly certain BMW uses EIVC, but I suppose there also have infinitely variable lift.

It doesn't seem unreasonable to be to have two cam profiles, perhaps one very low RPM profile with a switchover as low as 2500RPM, which is intended for EIVC.

[serialk11r]

Originally Posted by Ecky

I'm fairly certain BMW uses EIVC, but I suppose there also have infinitely variable lift.

It doesn't seem unreasonable to be to have two cam profiles, perhaps one very low RPM profile with a switchover as low as 2500RPM, which is intended for EIVC.

That's right, Multiair, Valvematic, Valvetronic, and VVEL are EIVC, since they reduce lift and duration together.

What you see with the Honda R18 and K20C2 is 1 LIVC and low lift cam and 1 normal duration cam. The R18 would use the normal cam all the way to the rev limit but the K20C2 uses the big cam at >6100rpm for powah.

LIVC without excessively high lift is better for pumping losses than EIVC since you draw air through a bigger opening, and there's also a little less negative work from compressing air heated by the cylinder walls. Ideally, you'd be able to close the intake valve at 300 degrees ATDC while idling.

[Bicycle Bob]

If you have a bad muffler, you can hear wasted pressure bursting past the exhaust valves whenever you reduce manifold vacuum substantially. You have to start the compression stroke well below atmospheric pressure to wind up with a heated charge not much above it. Any audible bang is an inefficiency alarm, however glorious they may sound in rapidly rising succession.

I think that the easiest way to test this is with an Eaton electric roots supercharger. It has about the right pressure differential range, but might want some high-vac oil. It could be rigged to go from idle vacuum right into boost if you wanted to, either producing or drawing power. However, doing that would make a variable compression ratio even more desirable. At high manifold vacuum, we should have a very high compression ratio to take full advantage of the charge not detonating. It might be best to never run even wide-open, and build the engine lighter. Such an engine would have a power density more suitable for air cooling, and I hear that 3-D printers make much better fins than casting.