Water pumps

I think it is time to take a look at the automotive water pump. There
are 2 major types of pumps, centrifugal and positive displacement, we
are interested in the centrifugal group.

Within that group there are 2 types of designs; open impellor and
closed impellor. Open impellor are used in almost all automotive
water pumps.

The open impellor type are used for their low cost of manufacture and
their relative ease of cleaning, should they become blocked.

The closed style is however far more efficient but is substantially
far higher in cost to produce, and far harder to clean if they become
blocked. Blockage in our situation is not a factor. Open style
impellor pumps are never used on flammable or explosive materials as
the clearance between the impellor and the volute can be reduced to
zero by thermal expansion. Automotive pumps run with very high
clearances to avoid the same problem, this makes them low
efficiency.

We must now consider the application they are being used on. In
Industry the pump is designed around a set of parameters that suit
the customers requirements, and in general used at a constant speed.
In the case of the automobile and other internal combustion engines
this speed can vary considerably.

The automotive pump must be able to deliver sufficient flow at low
rpm and or low speed and high engine load to cool adequately. If you
double the speed of this pump you require the cube in energy, i.e.
2x2x2=8.

Consider an engine running at 2000 rpm at idle and then at 6000rpm or
higher as we do at cruise, the auto environment never for domestic
use run at so high rpm for any length of time. You have increased the
speed by a factor of 3, so now you require the cube of 3, 3x3x3=27.

I have no argument with Pauls formulae for Volts x Amps = watts, and
the conversion between HP to Watts, what is in dispute is that you
have increased the rpm by a factor of 3 but have not increased the
flow by anywhere near that factor, but are consuming 27 times the
power. You could only possibly hope to increase the flow rate by a
factor of 3 by using variable geometry impellors and volute. This
aint going to happen.

In any case to increase the flow by this amount and be able to
exchange 3 times the heat is also not going to happen with a heat
exchanger designed to suit the output of the pump So where does this
leave us. It leaves us with a pump that is highly inefficient and a
vast waste of energy, but is acceptable to the auto manufactures, and
where do you think all that energy goes? Some is lost in slippage
within the pump due to high clearances, the higher pressure is able
to pass easily to the low pressure side thus reducing pumping
efficiency, but the rest goes into heating the cooling water!!!

Some European car manufactures have woken up to this and are using
electric pumps, also many of the rotary racing boys here and
probably elsewhere are using electric pumps. They consume a fraction
of the power of a mechanical pump, but are highly efficient. I will
be running an electric pump made by Davis Craig of Australia, check
them out. I also Know of a tri rotor here in Tauranga that is a race
car that has run an electric pump for some years.

Andrew. Campbell

What kind of race car? What brand of electric pump? What size is the
fuse?

Paul Lamar

Hi!: thanks for the good info.

I'd always consider having hot water going out of block from the
upper part, then entering, as usual, radiator in its upper part, and
the return piping from lower radiator part, entering the low part of
block or housing, in order that, provided the case in pumping
failures, no wall in pipes exist, some Thermal-siphon effect always
exists, making liquid circulate.

is this a reasonable concern, or too childish, already implemented
from very early times?

Regards, Salut †

Jose Gros-Aymerich

Madrid, Spain

The liquid cooled aircraft engines of WW II have pump impellers that
look just like the pump impellers available today. Therefor the same
pump HP needed per prop HP out is the same. The Mazda impellers have
proven to be effective in aircraft use. Use a smaller impeller or
less pump HP at your own peril.

If you want to try an electric pump use a small starter motor. They
will develop 3 HP at 200 amps out of the battery. You will need two
alternators :-) I know of no car alternators that will put out 200
amps. There may be one out there but check the weight :-)

Paul Lamar

You May well be using at least 3 Hp to drive the pump and possibly
much more. But that energy is not being used to cool the engine,

The piston engine aircraft runs at a considerably lower RPM than a
rotary, so the centrifugal pump is not being overdriven by the same
extent. If the automotive pump could be run on a variable speed drive
your power consumption would reduce dramatically, but not the cooling
ability. At high RPM these pumps are acting as a water break similar
but way cruder than the water breaks we used to use for engine
testing. Andrew Campbell.

So far I have not seen a picture of Davis Craig impeller. The Mazda
2nd gen RX7 impeller is about as good as they get.

That preclude a better impeller. The Mazda engineers are no dummy's
when it comes to designing anything. The last time I looked at a
electric water pump the impeller was nowhere near as good as the 2nd
gen

The RPM of the pump has nothing to do with it, Less RPM more
torque.

HP = RPM * Torque / 52 52 That is the bottom line.

The HP of the pump must be there.

1 HP = 746 watts 3 HP 2238 watts.

Watts = Voltage * Current

If it is coming from the battery the current must be there. Since the
battery is only 12 volts the current must be 200 AMPS!!!

Please tell me what fuse he is using :-)

It takes a lot of HP to pump water through the engine at high
velocity sufficient to cool it under worse case conditions. Climbing
the Grape Vine in SoCal in a car when the temperature is 100 degrees
F at high speed is a worst case condition.

You see at least one car pulled over to the side with cooling
problems. Taking off on a hot day and then running the aircraft
engine 70 to 80 % continuous is another problem. Usually the system
would fail during taxing.

We have been over and over about all this years and years ago. By
writing the book I fixed the problem forever :-)

. Pumping water from one container to another is super easy.

Paul Lamar

Here is a good paper on modifying a rotary water pump of better performance.

From the summary ...

Re-Design of Impeller of Centrifugal Pump of a Wankel Engine and Its Performance Analysis Using CFD

"Considering the above design modifications and their corresponding performances, for higher flow rate to be achieved modifications to done on exit angle, impeller inlet diameter and no. of blades. Following modifications are made to the existing impeller.

1.    Exit angle( β2 ) is increased by 100

2.    Inlet impeller diameter is decreased by 3.5mm.

3.    No. of blades are increased to 9 and thickness is reduced such that area occupied by impeller is same. Blade thickness is 2.4mm.

Volume flow rate and head developed by pump with this impeller is 76.8 LPM and 14m. Flow rate and head is increased by 22% and 6% respectively.
This improved design is fabricated and tested in test rig. About 12% improvement is seen in the volume flow rate, i.e. flow rate increased from 58 LPM to 65 LPM "

I can send the full paper in another resonance if needed.
Barry Bordes

Barry, try to send plain text only from your Ipad.
Apple inserts a bunch of hidden special characters in their messages.

They are probably talking about the RX8 pump or the 3rd gen Rx7.

They were selling it for awhile. Too much money :-)
They probably make them one at a time on an NC mill rather than casting them.

The stock RX8 was not as good as the stock 2nd gen RX7.

Please send the paper as a pdf if you have it.

Paul Lamar

The Rotary Engine free News Letter. Powered by Linux. If you want off
the list PLEASE let me know and I will remove you. ACRE NL web
site. http://www.rotaryeng.net You Tube http://tinyurl.com/beqqxas
Copyright 1998-2017 All world wide rights reserved.