Subject: New pump test at 20 psi pre pressure
From: ACRE
Date: 7/21/2003, 3:32 PM


Bill,

Here is the data you requested.
The data is at 4300 RPM for the new water pump. 20 psi  pre pressure.
Pressures in psi. Pump outlet, flow meter, inlet respectively.
Flow in gallons per minute.

flow      p1   p2     p3
closed    18   17     5
15        18   18     6
20        18   17.5   6
25        18   17.5   6
30        18   17     6
35        18   17     6
40        18   17     7
45        18.5 17     9
50        19   17.5   11
55        19.5 17.5   12
75        20   17.5   14

Stopped   18.5  18    19

Pressures are a bit higher than the stock pump but in the same ball park.
It does not have the 1/8 to 3/16th bleed hole like the stock
pump. Could be added by a small external line.

Same characteristics of a closed loop car cooling system.
Pressure has little to do with flow over a wide range.
That is why the bleed hole does not seem to hurt the flow all that
much in the stock pump.

It might be there to prevent cavitation at high speed if the lower
rad hose collapses. I suspect this might be what happened to Perry's pump.
The bleed hole got blocked with trash. Or it could be there to circulate
a bit of water in the block when the rad is plugged up.

On second thought looking at these pictures again  the system could
have simply been run low on coolant for awhile at high RPM.

Paul Lamar

Tracy Crook wrote:

Paul, I'm not prepared to offer any real world test data but I may have an
explaination for why your data looks so different from what we might expect.
Your test setup is very close to a true closed (rigid) system where there
are no significant pressure gradients.   A typical cooling system is not a
true closed system because of the elasticity of hoses, large variation in
path cross section, etc.  It would be interesting to see the effect of a big
chunk of rubber hose on the inlet side of the system.

Just a thought.  Back to programming hell for me.

Tracy

I do have some hose in there and yes it does expand and get hard.
A lot of liquid cooled airplanes use aluminum tube plumbing anyway and
yes I know nothing is infinitely ridged.

BTW There is one mode of operation I have not tried yet.
When the expansion tank and pressure cap is located between the rad and the inlet
to the pump, and the pressure on the inlet side of the pump is
below atmospheric pressure, fluid will be drawn in from the overflow
tank through the pressure cap one way valve and the pressure on the outlet 
side of the pump will increase until the negative pressure is 
normalized to atmospheric pressure.

The pressure differential across the pump will remain the same however.
That can be simulated by putting about 15 psi pre pressure in the
test cooling system.

I want to point out that no amount of increased fluid volume on the inlet
side of the pump is going to make a hill of beans difference
in the closed loop system situation. 

You must also realize that any open loop system
sucking from an open container at roughly the same inlet pump level will have 
atmospheric pressure on the inlet side of the pump. The same thing
cannot be said about a closed loop sealed cooling system such as we
are dealing with here.

BTW A block of water one foot cube exerts a force on the table it is
sitting on of 65 pounds. That is 144 square inches so that is 65
pounds divided by 144 or .45 psi. That means a 12" column of water one inch
square  weighs .45 psi. A stand pipe on the inlet side IF OPEN AT THE TOP
will exert .45 psi for every foot in length. It matter not how large in diameter
the stand pipe is. It could be the size of the lake behind Grand Coulee
damn and it would still be .45 psi per foot. 

This is also the principle of the water manometer which we should all
be very familiar with by now.
 
Unfortunately it is not
open on the top so the air pressure above the column tends to hold
the water from running out. Just like a pipette or soda straw works when
you dip it into a glass of water sealing the end with your finger.

This is a whole order of magnitude less pressure than we are dealing
with in these pump system so a stand pipe is a waste of time.

Paul Lamar
 
The AirCraft Rotary Engine NewsLetter.  Powered by Linux.
ACRE NL web site. http://home.earthlink.net/~rotaryeng/            
Copyright 1998-2003 All world wide rights reserved.