Subject: Oil to water heat exchanger data
From: Rotary Engine
Date: 2/26/2008, 8:33 PM
To: AARotary Engine




    >                     Mark,
    >
    >                     So you're not interested in the cooling
   redundancy that
    >         Paul has
    >                     suggested as a good idea - using oil-air and
   glycol-air
    >                 exchangers at
    >                     the same time?  I still think that would be the
    >         direction I'd
    >                 go.  In
    >                     fact, I'd make that a requirement, as I think
   it's
    >     something
    >                 that could
    >                     make for a more favorable outcome should the
    >     airplane suffer
    >             from a
    >                     single component failure..
    >
    >                     Maybe it's just anecdotal, but the vaunted
   P-51 Mustang
    >         has a
    >                 separate
    >                     oil-air cooler, as I believe do the other
   Merlin powered
    >             aircraft
    >                     (Spitfires, Hurricanes, etc).  I think the
   P-40 (Allison
    >                 powered) was
    >                     the same way.
    >
    >                     The Mustang had an induction air to glycol
   supercharger
    >                 intercooler, but
    >                     this ran with a separate cooling loop from
   the engine -
    >         and had
    >                 its own
    >                     coolant pump and air-glycol exchanger.  This
   allowed
    >         putting the
    >                     intercooler where the
    >     directness/compactness/packaging is
    >                 important.  It
    >                     wouldn't be practical to have a duct feeding
   an air-air
    >             intercooler
    >                     between the stages of the supercharger (on a
   Merlin).
    >
    >                     Please keep in mind, I don't have a rotary
   powered
    >     airplane,
    >             nor a
    >                     liquid cooled airplane, nor even an airplane
   I built
    >         myself, so
    >                 take my
    >                     ideas with a grain of salt..  :)
    >
    >                     I also commend you for putting together an
   airplane
    >     that's
    >             running
    >                     well.  I admire the accomplishment.
    >
    >
    >                     Regards,
    >
    >                     Matt-
    >
    >                 Matt,
    >                 I do have a glycol-air and an oil-air exchanger,
   but also an
    >             oil-glycol
    >                 exchanger.  I don't see where removing the oil-glycol
    >     exchanger
    >             is going
    >                 to improve the time I have to get on the ground
   should the
    >             glycol system
    >                 spring a leak.  I may explore the possibility of
   building a
    >             dedicated
    >                 exit duct for the oil-air exchanger though... in my
    >     spare time.
    >              If that
    >                 works out, then I'll consider removing the oil-glycol
    >     exchanger.
    >
    >                 Mark S.
    >
    >                 I am really interested in why it is not working
   better. If
    >         you don't
    >                 mind,
    >                 next time you are out there, please record the
   number of
    >         fins per
    >                 inch and the
    >                 tube spacing. Also if you have a water manometer
   measure the
    >             pressure
    >                 on the face of the oil cooler at speed. Use the
   porous
    >     foam hot
    >             glued to
    >                 the end of the sense tube trick. You have my
   cooling book.
    >
    >                 Here is a chart of dynamic pressure verses speed. The
    >     pressure
    >             should be
    >                 .3 to .4 of this number in the center of the
   core. If it is
    >         higher
    >                 or lower
    >                 there could be something wrong with the duct or
   the porosity
    >         of the
    >                 oil cooler.
    >
    >                 In other words at 120 MPH dynamic pressure is 6.9
   inches of
    >         water.
    >                 The oil cooler face should be in the ball park of
   2 to 2.8
    >         inches of
    >                 water.
    >                 The higher the number the less the porosity of
   the oil
    >     cooler.
    >             If it is
    >                 blocking the air flow through the core
   excessively it should
    >         be much
    >                 higher
    >                 than 2.8 inches.
    >
    >                 If it is lower than 2 inches you may have a leak
   in the duct
    >         or the
    >                 duct may
    >                 be improperly shaped.  You can do the same thing
   to the
    >         glycol rad
    >                 and compare
    >                 the two numbers. It would also be helpful to
   measure the
    >         pressure
    >                 behind the core in the cowl as well.
    >
    >                 You would not happen to have any data like this
   on the oil
    >         cooler
    >                 would you?
    >
    >                 My brain is kind of fried. Pardon the pun. Perhaps
    >     perhaps Bill
    >                 Schertz can tell
    >                 us where we want to be on the 3 rotor oil BTU
   rejection.
    >
    >                 Paul Lamar
    >
    >
    >             Paul,
    >
    >             Won't the delta-P of the oil cooler be affected by the
    >         properties of the
    >             glycol exchanger and its duct?  If the glycol portion has
    >         especially low
    >             drag, I think the pressure inside the cowl will be
   relatively
    >         high (low
    >             delta-P across glycol exchanger), making it look like
   there's
    >         something
    >             wrong with the oil exchanger/duct.
    >
    >             I guess if you check the pressure on the front side
   of the
    >     glycol
    >             exchanger, and in-cowl pressure, and compare all to
   ambient, you
    >         should
    >             have a pretty good picture of where the air wants to go.
    >
    >             The inlet duct on the oil-air exchanger looks pretty good
    >     (geometry,
    >             shape).  If it leaked air, that would lower the
   glycol-air
    >     exchanger
    >             efficiency (that doesn't appear to be the issue) by
   raising the
    >         in-cowl
    >             pressure.  So, it seems like the oil-air exchanger
   might not
    >     be well
    >             matched to the glycol-air exchanger - when feeding a
   common
    >     outlet
    >             chamber.  If this is the case, the only fix I can
   think of is to
    >         arrange
    >             for an isolated outlet from the oil-air exchanger.
    You could
    >         also block
    >             a portion of the outlet of the glycol-air exchanger,
   but that
    >         seems so
    >             wrong..  :)
    >
    >             Mark, is your cowl in final paint?  I think it would
   be pretty
    >         easy to
    >             cut louvers aft of the oil-air exchanger, build a
   small outlet
    >         plenum
    >             and connect it to the outlet of the oil-air exchanger
   with large
    >             diameter scat tube.  Maybe it would look something
   like this:
    >
    >             http://www.cs-ent.ca/6578o.jpg
    >
    >             If it doesn't buy you anything, you wouldn't have
   spent much on
    >         it, and
    >             they'd be easy to cover over later (assuming you
   aren't in final
    >         paint).
    >
    >
    >             Regards,
    >
    >             Matt-
    >
    >             Best to put a small flap on there to keep the oil temp up
    >             at high altitude cruise. Reduces the GPH.
    >
    >             --
    >             Paul Lamar ...No rotor no motor.
    >
    >
    >
    >         Sure..  I was just trying to come up with a cheap proof of
    >     concept.  A
    >         cowl flap would be ideal.  Could even be controlled by a
   picaxe
    >     or maybe
    >         simple a sealed bellows..
    >
    >         Matt-
    >
    >         VW thermostat. BTW it sucks air through the oil cooler
   when wide
    >     open.
    >
    >         Paul Lamar
    >
    >     Yes, a/c is still in primer.  I'll see what I can do on this.
    I'm
    >     pretty sure Lancair sells fiberglass gills that I could bond onto
    >     the lower cowl.  That's just what I wanted... more fiberglass
   work.
    >
    >     Mark S.
    >
    >     Make some measurements first and see if it is not something else
    >     such as an air leak in the input duct. The less porous the
   core the more
    >     important the input duct air leaks.
    >
    >     Use thin wood ply door skins from the Home Depot to make the
   duct.
    >     Then cover
    >     them with glass. Quick and easy.
    >
    >     As far as I know three people have been affected by coolant
   leaks. Two
    >     tried to turn back to the runway and crashed and one landed
   on the
    >     freeway.
    >     Non fatal crashes.
    >
    >     When water turns to steam it expands something like 6000
   times as I
    >     recall.
    >     What happens next is the pressure cap over pressure valve
   pops off
    >     and the steam pumps the remaining coolant overboard in seconds.
    >     Not something you want to fool around with.
    >
    >     Use a cowl flap Mark.
    >
    >     It will enhance the flow through the oil core during take off and
    >     climb out.
    >
    >     Aero drag matters not on climb out. Much of the HP just goes
   to lift
    >     the weight of the airplane. 2500 pounds at 1000 feet per
   minute is
    >     is 2,500,000 ft pounds per minute. Dividing by 33,000 foot
   pounds per
    >     minute which is one HP equals 75 HP just to lift the weight
   of the
    >     aircraft
    >     alone not counting aero drag. If it is 1500 FPM it becomes
   112.5 HP.
    >     The induced
    >     aero drag also increases due to the higher wing lift coefficient
    >     required.
    >
    >     I know you know most of this as you have my cooling book so
   this is
    >     primarily
    >     for the new guys.
    >
    >     Paul Lamar ...No rotor no motor.

    > Paul,
    >
    > I was very careful when fabricating the inlet duct to ensure a tight
    > fit, which it does.  But there is a small hole or two at the corners
    > that could use a dab of silicone.  I know it doesn't take much to
   cause
    > a big leak, but the holes I'm referring to are very small.  I'll plug
    > them and see if it makes any difference.
    >
    > Good idea of using door skin material for a form for the duct.  Of
    > course, it needs to be covered with packing tape or some form of
   release
    > agent or it will be forever married to the fiberglass.
    >
    > Mark S.
    >

   Put some duct tape over the piano hinge as well.

   You leave the door skin in there. It is very light weight sandwich
   and it is stiff.
   Almost like balsa wood core. I learned this trick from Perry Mick.
   You only need
   one layer of glass cloth. Actually 4 oz glass cloth would be OK if
   you can find some.
   --
   Paul Lamar ...No rotor no motor.



Yes, I see how he did that.  Good idea.  I'm guessing that my exit duct
will have to be curved to go around the exhaust.  First, heat may be too
great for the glass, but I have some fiberfrax material that I can use
to protect it.   Also, I fabricated a complete exhaust shield since this
picture was taken.  So, the radiant heat is much less than before.

I have lots of 4" open cell foam that is easy to cut and shape.  That's
what I used to make the form for the inlet duct.   I figured I  would
use it for this, if curves are required.  Either way...

Mark S.

Well I think we still need a scale 3D so we have a uniform reduction
in cross sectional area.

Paul Lamar

The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
Copyright 1998-2007 All world wide rights reserved.