Barry Gardner wrote:
Paul,
Let me offer my own spreadsheet for turbo selection for consideration and
critique by the assembled wisdom on this list. It is based on reading both
Hugh MacInnes and Corky Bell and summarizes what I was able to figure out.
It is different from Jeff Spitzer's spreadsheet because it answers the
questions in a different way. I took a turbonormalizing approach but assumed
that pilots may be interested in a little turbo boost on takeoff. Therefore,
the question to ask was, how thin will the air be at various altitudes and
how much boost will be required to maintain the power that prevailed on the
runway?
If you deconstruct the calculations in the cells, Jeff Spitzer's horsepower
calculation is dependent upon an estimated BSFC and airflow ratio. While I
believe his numbers are in the ballpark, the HP calculation is very
sensitive to these estimates.
I used Corky Bell's approach of applying the boost % increase to a normally
aspirated engine. I used 146 HP for a naturally aspirated engine, which may
be too conservative given the ability of this group to port the engine,
remove the emissions controls, tune for best power, etc. A better number may
be 160 HP if all that stuff is done.
I was also interested in converting these estimates to lbs of air/min. Corky
Bell uses cfm, which he acknowledges to be the "semi-incorrect" term (p27).
The problem is that the Turbonetics catalog uses lbs of air/min in its
compressor maps. This spreadsheet gets you from one to the other.
By allowing users to plug in compressor efficiency, it computes the pressure
ratio, the other variable necessary to pick from the Turbonetics catalog.
User variables are highlighted in green at the top of the sheet (though none
of the cells are protected and formulas can be intentiionally or
unintentionally modified).
I have a couple questions I wasn't able to find answers for. If you or the
rest of the crew have comments or suggestions, I'd be pleased to listen.
1) In converting air density in slugs per cubic foot to lbs per cubic foot,
I found a conversion constant of 32.174 which was said to be roughly
accurate. However, I don't know if that's accurate at altitude though I
didn't know why it wouldn't be. The density figure changes as a function of
altitude so it seems to me that this cures the conversion problem. But
"slugs" are not units I deal with daily.
2) The tables for figuring the standard lapse rates all use 59F or 15C as a
starting point. Higher temps affect pilots of course but we term our
adjustments in terms of equivalent higher altitude ("density altitude''). To
tweak the accuracy of this model, I'd like to know the standard lapse rates
at sea level for temps above the standard 59F--but wasn't able to find any.
Users can crank in how many pounds of boost they are willing to put through
the engine and figure out what altitude they wish to attain that at. Then
they can figure out the pressure ratio and lbs of air/minute and go window
shopping in the Turbonetics compressor/trim size maps.
I didn't mess with A/R ratios. A/R's below 1.0 are used to get better
low-speed response and bigger A/R's are used to get more power. Bell writes
(p33) that getting the right A/R ratio is hard, and "...trial and error is
usually necessary." Given this group's emphasis on power and given that low
speed throttle response isn't much valued by pilots, I suspect our bias is
towards higher A/R ratios.
Greg Richter has real-world experience and Jeff Spitzer is one heckuva
engineer, so they have different and valuable perspectives to add to this.
But this spreadsheet is my attempt to synthesize the theory specifically for
turbo selection. Please critique it and let me know where errors lie.
Barry Gardner
Wheaton, IL
The AirCraft Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. Copyright 1998-2002 All world wide rights reserved.
http://home.earthlink.net/~rotaryeng/ http://www.linux.org