Twin Centrifugals VS Single

[Walter Sherwin]

----- Original Message -----
From: BamaInstrument
To: diy_efi at diy-efi.org
Sent: Tuesday, April 17, 2001 2:37 PM
Subject: Re: Twin Centrifugals VS Single


I like the concept and have been looking at something similar.  It seems
that a large centrifugal supercharger would work okay, but I don't like the
pressure/flow (and thus) torque curve.  The results from a centrifugal unit
are something like the square of the tip speed.  If the diameter of the
charger is larger then the tip speed can be accomplished with a lower RPM.
I've been looking at gearing directly to the crank, as opposed to belt
drive.  The cam is a VERY high and variable load to the crank and seems to
be handled quite well

I like the concept of some type of variable gearing.  The early Paxtons (?)
did this by changing the driven pulley diameter.  Mainly I'd like to have a
much broader torque range than would be available with the current engine
centrifugal super chargers.

dh
---------------------


Yes, unfortunately the response of a belt driven centrifugal unit is
proportional to the square of the impeller tip speed, together with a few
other minor influences.  It would be ideal to either: 1) vary the belt drive
ratio with respect to the crank speed [ala turbo's], or  2) oversize the
unit, within the impeller speed confines, and then dump the unwanted
manifold boost/flow via an intelligent dump valve controller [assuming a
blow-thru fuel configuration].

#1 has been used in various reincarnations over the years, but,  is fraught
with two problems.  The first problem is what sort of a reliable variable
ratio device to stick into the drive arrangement.  Large belt driven
centrifugals, at high pressure ratios,  require significant shaft power.
The second problem is adherence to the ultimate impeller speed [you don't
want the impeller to try to escape its housing at an inopportune moment].

#2 has not been commonly attempted, but might work?  Would require some
further study and noodling.  Again, one would have to adhere to the ultimate
impellor speed issues.

The "twin" belt driven centrifugal approach mentioned earlier, is yet
another concept, and might allow a designer to sum the aggregate effects of
two smaller compressors, and ?possibly? derive a more suitable broad range
net air/boost/pressure curve with respect to crank speed [and perhaps with
higher adiabatic efficiency].  As Greg said in another related post, one
would have to sit down with the individual compressor maps in hand and
crunch the numbers to examine the net benefits [if any?].  Twins also open
the door for compounding, down the road, should your pressure desires become
wild.

Interesting topic.  I'm  just noodling thru it myself too, with an eye
toward a project at hand.

Oh ya, EFI content........all the above would work well with just about any
form of blow-thru-compatible EFI system that understands 2-3 bar boost, even
TBI :)


Walt.

----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org