Subject: Estimating HP of a gas turbine
From: Rotary Engine
Date: 4/13/2006, 6:30 PM
To: AAme


I had an idea for estimating the HP of a gas turbine.

Turbines generate max torque when
the blades are stationary and then there is a straight
line drop off as RPM increases and blade velocity approaches
exhaust gas velocity. See the chart. This chart was a wild
guess on my part so it could be far from the truth.
The following estimate would indicate it is too optimistic.

The blade area of this stock Mazda turbo charger turbine is
about .45 square inches not counting the flap. There are
eleven blades for a total area of about 5 square inches. At
600 MPH exhaust gas velocity the dynamic pressure at standard
SL density is 936 pounds per square foot or 6.5 psi above
atmospheric.

6.5 psi times 5 square inches is a force of 32.5 pounds
assuming the blade lift coefficient is one.

The radius of the blades is about 1.25 inches so
the static torque on the turbine is 41 inch pounds
or 3.4 foot pounds.

Mark a point on this chart on the left vertical axis at 3.4
foot pounds. We are almost done. Now we need to figure the RPM
when th blade velocity is the same as 600 MPH. The mean radius is
1.25 inches so the circumference is 7.85 inches or .65 feet.
600 mph is 880 feet per second. That is 52,800 feet per minute
so the RPM is 52,800 divided by .65 feet or 81,230 RPM.
So mark a point on the x axis at 81,230 RPM and draw
a straight line up to the max torque point of 3.4 foot pounds.
Now you must calculate the HP and plot it as a function
of RPM. You are now done.

These numbers are amazingly close to what you see in a real turbo
charger. Actually they are kind of conservative as turbo
chargers consistently rotate faster than 81,000 RPM as most
of the compressor charts indicate. That means either my
gas squirt velocity assumption is too low or my exhaust gas density
assumption is too low. Perhaps the gas squirt velocity is supersonic
so I don't know how to deal with that. I have no idea what
happens in a nozzle when the gas velocity approaches supersonic.

Come to think about it the pressure in the exhaust manifold
is probably the same as the dynamic pressure on the blade
as it is a closed container more or less at zero turbo charger
RPM. Exhaust manifold pressures have been
measured at way more than 6.5 psi. First column in third
attachment Table III. 30 inches of Hg is 15 psi atmospheric
pressure and 60 inches of Hg is 15 psi above atmospheric.

I was assuming less than half that at 6.5 psi. Maybe my
chart is not optimistic after all. The heat balance of the
rotary would indicate it is close to correct.

BTW This is also why the HP from the turbine increases with
altitude as the back pressure on the turbine is reduced.

Other possible errors:
My lift coefficient assumption of one on the blade could
also be too low. Any self respecting wing will will have a
lift coefficient of 1.2 or above.

Another possibility is the blade area chosen is too low
and I need to include the flap area.

Paul Lamar ...No rotor no motor.

Paul,
Another way to get a handle on the exhaust gas energy is to think of the
engine as an air pump and heater, then use the equation for power:
power = (pressure)  x  (volume flow rate).

At 6500 rpm the engine will take in about 293cu ft/min of air.  If the
temperature rise is from 72F  to 1300F (532R to 1760R) then the exhaust
flow will be (293 cu ft/min)(1760/532) = 970 cu ft/min.

Using your pressure across the turbine of  6.5psi (936psf), the power
equation gives:
power = (936psf) x 970cu ft/min) = 908000 f-lb/min = 27hp.

Of course, working with gasses and a centrifugal turbine it will not  be
possible to get anything more than an order of magnitude estimate from
any of these calculations.  The concept of power balance is rarely
mentioned in turbo charging discussions.   A given compressor/turbine
combination will reach an equilibrium rpm when the power balances on the
two ends of the connecting shaft.  The beauty of  the turbo-compound
turbine is that  its rpm will be known once it receives enough exhaust
power to contribute to the engine output.

Aubrey

Sounds about right. Nice and simple formula. Thanks Aubry.

Double check. Coming at it from another direction.
Lets see 200 HP at a BSFC of .47 is 94 pounds of fuel
per hour. At stoichiometric 14.7:1  that is 1381 pounds of air
per hour or 23 pounds per minute with an air density
of .0767 pounds per cubic foot that is 300 cubic feet
per minute. Close.  Real close!!! :)

Here is a chart from Hugh MacInnes book "Turbochargers"
showing the air flow requirements of a 2.6 liter turbocharged
piston engine at 5 psi and 10 psi boost. The VE is 85% giving credence
to the fact the NA side intake port rotary VE is more like 100%.

I think a 120% VE P-port turbo compound would be the cat's meow
but I am not going to build one just yet.

Paul Lamar ...No rotor no motor.



Hi Paul;

Regarding the airfoil selection.
IIRC - Most of the laminar sections I have looked at had a maximum
lift/drag somewhere
between  0.9 and 1.0 for sections with slight camber (flaps).  I assume
this would be better
even though the exhaust flow is extremely turbulent.
Transonic, supersonic, Reynolds number and airfoil may alter this
considerably.

Another thing to think about, I'm guessing you chose a maximum blade
velocity of 600 mph to avoid transonic tip losses.
I would have to calculate it, but at the exhaust temp of 1500F and
pressure?, the speed of sound is probably
more than twice as fast.

Cheers
Cary

Good point.
Well that could partially explain the descripancy. We might
be dealing with 900 MPH air.

Lets see. One slot in the dual scroll turbine housing
is .33 wide and 2.8 inches in diameter or an area of 2.9
square inches. That comuncates with only one rotor.

The other slot has a width of .2 inches so the area is
1.76 square inches.

Lets see 970 cubic feet per minute half of that is 485 cubic
feet per minute or 838,080 cubic inches per minute going
through 2.9 square inches is a velocity of 289,000 inches
per minute or 24,000 feet per minute or 400 feet per
second or only 272 MPH. Hmmmmmm. Check my arithmetic.

838,080 cubic inches per minute going through 1.76 square
inches is 476,000 inches per minute or 40,000 feet per
minute or 661 feet per second or 450 MPH. Interesting.

Now I am really out of the ball park I think.

Perhaps I need to take the area (the A in A/R) entering
the scroll. Hard to say what that is without sawing
the turbine housing apart.

I liked Aubrey's method better :)

The airfoil is just a bent up piece of sheet metal
basically.

Paul Lamar ...No rotor no motor.

The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
Copyright 1998-2006 All world wide rights reserved.