First, couple of things – Tracy is absolutely correct;
back pressure on the pump reduces pump power. Why? Because it reduces the flow
rate, and power is a function of the mass being moved around the loop.
Lynn is correct that the largest pressure drops in the cooling loop; in
a car, certainly; is the radiator. May not always be true in the airplanes
since some folks, like Tracy, are using small diameter lines to reduce weight. Lots of
entry and exit losses on the radiator tubes; however, the total tube x-section
on a single pass car rad is generally much larger than the x-section of the
hose.
Now this issue of pump cavitation may be
worth a bit more exploring. The cavitation occurs on the back side of the
impellor blade where the pressure is low – more representative of the
pump inlet pressure; not the outlet. Now let’s say I’m
running a 20 psi cap; and that cap is seeing the pressure on the outlet side of
the pump. And let’s say we’re running at high power and high
rpm where we would be most concerned about cavitation, and the engine is hot
and the system pressure is at cap pressure.
Now let’s say the pressure drop
around the loop, pump inlet to pump outlet is, whatever; 8 psi. The pump
is now seeing 20 psi (above ambient) at the outlet, and 12 psi at the
inlet. Now let’s add a restrictor at the pump outlet which reduces the
flow, so maybe the drop in the rest of the loop is now only 5 psi. Pump inlet
is now higher; 15 psi. That’s good, reduces the likely hood of cavitation
a little bit.
But now let’s move the connection of
the filler – pressure cap – to the inlet side of the pump, as I
suggested in the earlier post. Now the inlet side of the pump is seeing
cap pressure of 20 psi. And the flow has not been reduced. Seems to
me like a better mousetrap.
And keep in mind that for high altitude
flight, the system absolute pressure is going down in direct relation to the
ambient pressure, so that 20# cap that gives an absolute pressure in the system
of 34# at sea level is giving about 10# less at 14,000 ft.
Al G
-----Original Message-----
From: Rotary motors in aircraft
[mailto:flyrotary@lancaironline.net] On
Behalf Of Lynn Hanover
Sent: Tuesday, August
18, 2009 7:24 AM
To: Rotary motors in aircraft
Subject: [FlyRotary] Swirl pots
Ed has all of the smarts on this. So pay attention
when he writes anything.
And now the Larch.....................
I hate to see a rotary engine go down the drain
because the engine overheated on the first start. And a lot of them do die that
way. The water pump is installed at the top of the engine, and will use any
excuse to cavitate and stop pumping coolant. It has no head on it. Pump
talk for water pressure caused by the weight of water above the pump. So a very
small amount of air behind the pump will stop it.
I have a Shrader valve on my make up tank, and charge
the system to relief pressure before starting the engine.
In the olden days, that center iron had a flat spot
with some kind of fat sensor stuck in it with two nice threaded holes along
side. That hole got a flat plate with a Shrader valve sans core installed. A
nice metal cap on the valve body kept the coolant inside. On a new fill up, you
take that cap off and pour in coolant until it came out of the Shrader valve
with no bubbles. Then put the cap back on. Fast, simple,
I should have removed the restrictor from the drawing
before posting it. It begs for comment. It is probably of no value in an
aircraft application.
Here is my thinking on that piece of the system,
backed up by observation only and no technical understanding or evidence.
On my flow bench I see a vena contracta around all
exposed tube ends, (like the raw end of the shop vac hose) that reduces the
effective diameter of the hose. The air making a 180 degree turn around the end
of the tube impinges on air flowing into the tube at any angle less than 180
into the tube. So the CFM you would calculate for the tube diameter and
pressure will be about half of the calculated flow when tested.
The smaller the tube diameter the more profound the
effect. Now look into the tank on a radiator and notice the flattened tubes
sticking up from the flat tank floor. Any vena contracta going on in there?
And it is in a liquid, way worse than flowing air.
While raising the pressure around the end of the
tube, flow does not improve as an exact function of pressure. Or, doubling
the pressure does not double the flow. And now more bad news. The
same construction at the outlet end of the tubes creates another
interesting condition, where high velocity flow from the raw tube end, creates
a high pressure differential between the flow and coolant at rest in the tank,
that caused the low speed coolant to constrict the flow exiting the tube. You
can see this on a humid day when a war bird revs up for take off. The vapor
rings from the prop tips are compressed inward and stay closely attached to the
fuselage. Or, a vena contracta observable in broad daylight.
I deduce from this that there is a considerable
restriction involved in forcing coolant through a radiator.
I am not alone in this thinking. Notice that the lower
radiator hoses on cars have a spiral of wire installed to prevent the collapse
of the hose at high engine speed. Even with the use of modern higher pressure
caps now common. Notice also that the lower hoses are larger in diameter
than the upper hoses.
So, the radiator, like the power valve or restrictor
in a Freon system, where a restriction of some sort is required to generate a
pressure differential. Unlike a Freon system that pressure differential is of
no value in our cooling systems.
So, the radiator(s) are the biggest restrictor in the
loop, and the lowest system pressure is likely to be found between the
radiator(s) and the pump inlet. When that pressure drops below local air
pressure
the lower hose would collapse if it were not for
the wire inside, your 22 pound pressure cap be damned.
So, the restrictor, limits the pressure differential
across the radiator, and helps prevent pump cavitation that shortens the vanes
on the pump and renders it ineffective. When you are shifting at 9,600 RPM and
miss the shift, (the rev chips are 9,600s) or the driver selects 1st instead of
3rd successfully (easy with dog rings) and the revs go to the moon, then the
restrictor is a help.
For airplanes, not so much.................Sorry for
any confusion.