Re: electrical supercharging

[Otter <kdsebe0@xxxxxxxxxxx>]

>Problem 1.  It takes about 25 hp to drive the compressor for a 3 liter
>engine 
>at 4000 RPM at 1.8 ATA manifold pressure.  This means for your system
>we'd need to replace a shaft and bearing assembly with a 25 hp motor,
>control electronics, energy storage system, and a 30 hp generator (to
>account
>for electrical and mechanical losses in the conversion process).
>(numbers for a MB 617.952 engine, 123 chassis 300D at 80+ MPH)

How'd you get these numbers?  I've heard everything from 5hp to 30hp now.
There are so many variables, especially the viscous losses through the
various pipings.  Unfortunately, I don't know how to calculate these
losses, but I think it will _greatly_ reduce the amount of power input
required to sell a given output pressure increase at the intake valve.

At proportianately higher speeds, I'd think that the turbo would *not*
require a proportianately higher energy input.  Ram air induction has a
small(<10% from what I understand) increase in pressure after 40mph.

And anyways, all I want is a low boost, so the motor doesn't have to be
_quite_ so big.

>Problem 2.  The turbine is located as close as possible to the exhaust ports
>to minimize heat losses from the exhaust gas stream.  This is done to
>maximize
>the efficiency of the turbine and allow it to extract the needed energy to
>drive
>the turbine from the exhaust while minimizing back pressure.

True, true.  But I don't want to move the turbine, simply the compressor.

> going to an electrical system would have dubious
>benefits
>and would entail a serious amount of extra weight.

How much is "serious"?  I'd think 50lbs on a 3500lb MB is negligible.

And if viscous losses are very high, which I think they are, then the
benefits aren't quite so "dubious".  

>If a centrifugal flow
>compressor
>is used, it would still have a lag time on spool up;

But the lag might be almost unnoticeable.

>if a positive
>displacement
>unit is used the lower efficiencies would mean a higher engine inlet
>temperature
>unless inter/after cooling was added.

Well, still, an intercooler can only cool down so much, so I guess I'd have
to avoid positive displacment.

>For a gasoline engine, boost is seldom used (depending on the driver, of
>course,
>but normal cruise mode does NOT require boost.  My engine (117.963) needs
>less than 4 psia
>in the inlet manifold to maintain 80+ MPH). 

So you convert that exhaust gas into something useful, say electricity to
power your car, since you don't need the boost.  And you manage the
compressor so that it provides maximum efficiency.

>Sizing the turbine presents
>problems,
>as the exhaust gas volume available varies greatly and while there are
>various 
>approaches to dealing with this (vanes, wastegates, multiple turbines) they
>all
>add to the complexity and cost of the system.

I'm confused.  Why would this present any more problems than already exist
with a mechanical setup?

>I'm not aware of too many
>off-shelf
>generators that run at 80,000-120,000 RPM, either.

Is that what it would have to run to???  WOW!  

How about a gear differential?

>With the light duty cycle, it would seem to be more efficient to simply tap
>crankshaft
>horsepower as required to run the blower, using an electrically activated
>clutch and
>a belt assembly.  True, you'd steal the power from the crankshaft but you'd
>use
>almost as much through back pressure on the turbine and there would be
>considerable
>savings in weight.

So you lose engine efficiency by taping into the exaust?  Or do you just
lose output power?

Just an idea I want to fully understand.  I can't do it at the moment, but
maybe sometime in the future...

But there must be _some_ way of getting better efficiency than what we've
got now.  Maybe high pressure sonic waves tuned to engine rpm and intake
valve timing! :)

Kenn
http://www.euromajic.com

------------------------------