Subject: Oil cooler air flow
From: Rotary Engine
Date: 6/29/2007, 7:32 PM
To: AARotary Engine


     Thanks..

     I'll bet the exit is in a region of reasonably thick boundary layer.
If
     so, the flow on the surface may be very slow or even reversed in places
     - leading to relatively high exit pressures and low flow through the
     cooler.  I think making the exit lip taller should help things, though
     not pretty.  Like a Gurney Flap, or the cowl exhaust on an older
     Cessna..
     Matt-
     --------------

     Matt;

     My little boundary layer program says the BL at the exit position (not
     considering the effects of the fairing) would be about 0.8"

     One of the photos that Paul posted (first photo) is my original
     configuration - just a curved ramping up to the exit.  This never flew
     because an "expert" told me that it would create a lot of turbulence
     and
     drag - that a better approach would be an airfoil shape that would
     be the
     flow attached.  This  supposedly would provide lower pressure at the
     exit
     and would also provide more spacing internally from the exit face of
     the
     core.  Convinced; I re-did the fairing as shown in the second photo.

     Based on the theory of low pressure above at the exit, and high
pressure
     below at the entrance, not much of a scoop was put at the entrance;
     just an
     airfoil shaped lip that extended below the surface about 1/2" (photo
3).
     The oil cooler, and second coolant rad are in the wing root behind the
     strakes (photo 4)

     Early flights showed the setup was not effective.  (BTW; custom oil
     cooler
     core dimensions of 5.5" x 21.5" x 3.3" thick.)  It seemed the most
     expedient
     solution was to make the inlet into more of a ram scoop, which I did
     initially with a temporary metal extension, the photo that Paul
     attached
     earlier.  This improved things enough to allow flight testing to
     proceed.  I
     then replaced it with a nicely glassed scoop of similar
     configuration (1.25"
     x 23" wide, sorry, no photo) and as we went into cooler winter
     weather; all
     was fine.

     This scoop does ingest the boundary layer, but if it's overall
     effectiveness
     is better than 65%, it should provide the necessary air flow.  I
     have place
     some VGs forward of the entrance to reduce BL effects.

     I have managed to get tubes in place to measure pressure before and
     after
     the core to measure pressures there, and will then move them to get
     pressures at the entrance and exit.  It'll be a few days before I
     have a
     chance to get data.

     I thought just cutting the exit fairing shorter would help, so I cut
the
     center section of the fairing back about 3" to see if it would help.
No
     detectable difference in the temperatures.  I have also tried VGs on
     the
     upper surface of the exit fairing to no avail.  Perhaps VGs further
     forward
     on the strake?

     The other wing has a second parallel coolant radiator still with the
     original inlet (no scoop), and it does very little cooling.  But it is
     interesting that the in-cowl rad with sort of p-51 style armpit scoop,
     handles the cooling just fine - total core volume of about 400 cu. in.

     Sorry to gone on here, but I appreciate any help I can get.  A solution
     (with minimal structure changes, of course) is necessary.

     Al


Maybe a boundary layer rake would be useful to figure out what's going on..
-------------
    Yeah; that would be great; just not something I have lying around
the shop:-)
--------------
I've been told that using an effective inlet scoop without an equally
effective exhaust scoop is like "pushing a rope"..  Both (air and rope)
tend to pile up and tangle.  Getting adequately low pressure on the
outlet side is at least as important as having high enough pressure on
the inlet side.

Since the shapes of the scoops appear to be roughly correct, that says
to me that the air pressures around the scoops are not favorable.  My
guess was that the exit was more susceptible to problems than the inlet
- higher vs. lower pressure, boundary layer thickness leading to slow
local flow, leading to poor suction/vacuum effect.

To improve things, either the exit scoop needs to stick up high enough
to get above the boundary layer, or something needs to be done to
thin/energize the boundary layer - more/bigger VG's than you tried
before?  Trimming away the exit scoop would seem to me to have a minimal
effect if the local flow is slow.
--------------------
    I agree with all that.  I tried small VGs on the exit fairing; maybe
larger VGs 2-3 feet ahead of the exit fairing would be more effective, ya
think?
I could try a Gurney flap on the rad side when I make a flight to measure
pressure on the oil side.  I can tell if it helps by a larger delta T on the
coolant.  I'm thinking 1/2 to 3/4" right at the trailing edge of the fairing
- sound right?
-------------------
Here's another idea that might work (later)..  I can't tell from the
pics which way the air flows while it's in the wing root - is it
basically vertical?  Or is one scoop ahead of the other?  I'm guessing
the inlet scoop is ahead of the exhaust?  It might be advantageous to
have the exhaust as far forward (up against the back of the wing spar)
as possible in order to put it in an area of thinner boundary layer.
----------------
    The inlet is right behind the spar, the outlet further back.  The
attached photo shows the ducting for the rad - oil cooler is similar.
Moving the exit 10" forward may not make that much difference; and is a
major undertaking.
---------------
Here's another thought..  Is there any chance that there's high pressure
air passing through the joint between the wing and strake (wing root)?
If so, that might radically interfere with the airflow over the exit
scoop..
---------------
    Interesting.  I thought about this the other day, and had considered
taping the gap to see if it helped. There is a gap there of maybe .040" that
resulted from making a wing incidence angle adjustment.  The space between
the spars connects to the engine compartment; which is likely higher
pressure.  One thought though - on the other side I adjusted the incidence
the other way and there is only a few thousandths gap, and that side doesn't
work either.  But it's worth a go..

Thanks for your thoughts, Matt.  I'll keep you posted.
The idea of the wing root coolers started with Alan Shaw (WingCo) who
installed a standard aircraft oil cooler in similar fashion on his Velocity,
and said it worked wonderfully. Why did his work, and mine not - that is the
question.

Best,

Al

P.S.  Paul; are we stuck with 'plain text' on this list.  Ability to use
'rich text' or 'HTML' would be a big help.

---------------------------------------------------------

Well you can send it anyway you like if it is important. I'll convert
it to plain text and jpg.

BTW boundary layer rakes are not hard to make. Here is one I made for Tracy.
The last two are from a ram jet in a museum.

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.