Twin Centrifugals VS Single

[Greg Hermann]

At 10:25 PM 4/12/01, Walter Sherwin wrote:
>Forwarded from another list.....
>
>
>
>>Here's something I've been tossing around in my mind, for street toy use,
>>and I'd love to hear more from the group........
>
>>Let's say you have a large displacement stroker motor, and you desire to
>>artificially stuff it's hunger with centrifugal supercharging (non
>>manufacturer specific at this >point).  The typical approach today would
>>be to seek out a largish single centrifugal unit that would do the dirty
>>job, and pulley it to gain the max desired air flow/boost at >a specific
>>engine speed.  This would net you a large diameter centrifugal, operating
>>at a somewhat less than maximum impeller speed, hung from one side of
>>your >engine.  This would work, and you would prolly have a cog belt
>>drive arrangement as a result, if you wanted serious manifold pressure.
>>Of course this will generate a new >force/belt vector load at the nose of
>>the crank, that may or may not affect your front bearings.  One might
>>even direct the output of this large compressor to a single
>>>inter/aftercooler device for post cooling.
>
>>Let's consider a different approach......
>
>>What about mounting two smaller centrifugals, low on the motor, perhaps
>>directly opposed to the crank centerline so as to cancel the belt load
>>vectors.  Further imagine >the discharge(s) from these compressors being
>>directed thru two parallel inter/aftercooler exchangers (ala: Porsche
>>twin turbo setups).  The twin intercoolers might be a >bit easier to
>>position within a street vehicle.  At an esoteric design level, twins
>>could allow a person to fabricate an engine bay setup that has a
>>symmetric and artful >appearance.
>
>>Cost aside, does anyone see any further advantages/disadvantages to
>>running two slightly smaller compressors in place of one larger unit?
>>Obviously, the cost will be >higher with twins.  I'm more wondering about
>>airflow versus crank speed, throttle response, boost response profile,
>>parasitic drain at a given combined flow/pressure, >etc.
>
>>Thoughts would be appreciated;
>>Walt.

The problem with a crank driven centrifugal is that boost varies more or
less with the square of crank speed. Just the basic law of centrifugal
devices. One can play games with restricting the inlet to keep the curve
from peaking quite so dramatically, but----this leads to higher pressure
ratios and even more problems with intercooling in the higher rev ranges.

Had never though about the idea of using two of them, it almost seems like
a deal where one would have to look at individual compressor curves to make
any sense of what would happen in terms of air flow. You have got the
mechanical advantages of using two pretty well nailed.

It might be possible to flatten out the boost curve to some degree by
playing with using two different size units, or running tow units at
different speeds---

I just flat don't like crank driven centrifugals because of the fact that
you have to go to a bazillion engine rpm to get any real boost. Mebbe a
variable ratio belt drive??? (Like from a snowmobile??)

Greg


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