Subject: Glass Goose Sea Hawker cooling systems.
From: paul lamar
Date: 8/14/2016, 8:20 PM
To: A10-Me-Earthlink


 The RX8 core is a good core to use as are Chrysler Doge pick up
 cores. Large frontal area with thin cores. The Dodge core is 20" by
 30" and is very very effective in a cross counter flow configuration
 on the TTC engine. Two cores back to back.

 Let's take the Dodge in a non cross flow config.

 20 by 30 is 900 square inches.

 We have learned more about this subject since Neil Ungar started
 flying and on the time to climb engine. Neil complained about over
 cooling. It appears we can reduce the intake area of the wedge
 diffuser to 15%, of the frontal area of the core, from the 20%
 recommended previously.

 In other words 15% of 600 square inches is  90 square inches. The
 core is 20 inches wide so the height of the 4.5 inches high and 20
 inches wide.

 Both Neil and the TTC  rads are directly in the prop blast so if that
 is not the case then you may note higher water temp when taxiing.

 Here is the dodge core on top with the diffuser underneath. Yet
 another 3D :-) It is far enough forward to give room for the intake
 manifold behind it. This leaves room underneath for and oil cooler.

 The exhaust out the back of the cabin may help to attach more air to
 the aft sides of the cabin. This is a problem that was addressed by
 vortex generators previously (or in addition).

 Paul Lamar


 Excellent recommendations, thank you!   I was thinking of having the
 radiators on the top/back (low pressure) of the upper cowling with a
 long K&L duct.  Your version may be easier.   I will play with some
 radiator models on the actual engine.  The smaller the rad size the
 easier it will be to fit.  20x30 will be a challenge.

 In this model,  it seems that two 20x30x1 counterflow radiators would
 give a total volume of 1200ci?   Is it correct that we need about
 600ci for water cooling and 200ci for oil?   If so, I assume that
 using cross counterflow rads would each need to be about 1/2 the
 Dodge size as long as I have 600ci?

 My RX8 has twin oil coolers about 8x10x2.  The single RX7 oil cooler
 is about the same volume.  The latter should fit up front if the
 water cooler intake is a little higher.

 Has anyone used 3D printing to create their engine cowling?

 Scott Gettings

 Those volume numbers depend on the thickness of the core. Thinner
 core less cubic inches are required as thin cores are far more
 efficient for a given volume. Dodge were leading the way with other
 manufactures catching on.

 This is a cross counter flow rad for a Tango

 14" by 24" They are Dodge Neon cores.

 Some what more compact at higher cost to make. You also need a
 skilled tig welder.

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




Scott,

 We should change the standard to square inches of frontal area
 because all the recent cores are around 5/8 to 3/4 inch thick
 including the RX8. Rad core frontal area is more correlated to
 cooling than volume.

 600 square inches frontal area for 250 HP single core. 2 x 300
 square inches  for cross counter flow. Two Neon rads are about that.
 Maybe a bit more so they should handle a mild turbo charge.

 BTW have you considered a 8" prop extension to help the air stick to
 the cabin sides? You can lesson the angel at which they go back. If
 you go back at 7 degrees and then cut it off flat that will
 implement the Kamm effect. Same as the recent Preuses.


 Paul Lamar



 Hi Paul, I noticed during my re-search on the Glass Goose that the
 Icon A5 also has issues back in the cabin sides. They have also used
 vortex generators in that area, on the test planes. The GG has the
 lower wing and the landing gear that also adds to the problems.

 Steve Steve Carlisle

 Yes. It is a result of the side by side seating. That Icon is almost
 a scam. Be wary?

 With the Glass Goose it would not be that hard to change it to fore
 and aft seating and narrow the cabin. That lower wing is necessary to
 keep the prop out of the water spray which will destroy it.

 Here are a couple of great books on seaplanes. Extensive discussion
 of the unique design issues.

 Paul Lamar

 Correction fore and aft seating.

 The Kamm 7 degrees does not work out well both side by side and fore
 and aft.

 The first 3D is side by side the rest are fore and aft.

 Of course to build it fore and aft  seating one would have to start
 over. Not practical for those already built.

 Paul Lamar

 IMHO if I were designing a sea plane I would scale something that
 worked well like some of the Dornier's like the DO 26. Google it.

 Actually the Schneider cup racers found out a long time ago floats
 were lower drag than sea planes.

 I also think a coot is a better than a glass goose.


 Paul Lamar

 Hi Paul, I stumbled across this amphibian concept, made in
 PlaneMaker, for Xplane.  It likely has no turbulence issues, a lot
 like the SeaRay amphibian. Applying this parasol wing and aft cabin
 mods together with interplane struts would also lift the prop out of
 the water, no need to put a prop dam on the empenage.

 Inline image 1

 Steve Carlisle

 Looks like it violates the 7 degree rule.

 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 1998-2016 All world wide rights reserved.



On Aug 13, 2016, at 12:36 PM, Scott Gettings <sgettings@cfl.rr.com <mailto:sgettings@cfl.rr.com wrote:


 *Subject:* *Glass Goose Sea Hawker cooling systems.*



 The RX8 core is a good core to use as are Chrysler Doge pick up
 cores. Large frontal area with thin cores. The Dodge core is 20"
 by 30" and is very very effective in a cross counter flow
 configuration on the TTC engine. Two cores back to back.

 Let's take the Dodge in a non cross flow config.

 20 by 30 is 900 square inches.

 We have learned more about this subject since Neil Ungar started
 flying and on the time to climb engine. Neil complained about over
 cooling. It appears we can reduce the intake area of the wedge
 diffuser to 15%, of the frontal area of the core, from the 20%
 recommended previously.

 In other words 15% of 600 square inches is  90 square inches. The
 core is 20 inches wide so the height of the 4.5 inches high and 20
 inches wide.

 Both Neil and the TTC  rads are directly in the prop blast so if
 that is not the case then you may note higher water temp when
 taxiing.

 Here is the dodge core on top with the diffuser underneath. Yet
 another 3D :-) It is far enough forward to give room for the
 intake manifold behind it. This leaves room underneath for and oil
 cooler.

 The exhaust out the back of the cabin may help to attach more air
 to the aft sides of the cabin. This is a problem that was addressed
 by vortex generators previously (or in addition).

 Paul Lamar


 Excellent recommendations, thank you!   I was thinking of having
 the radiators on the top/back (low pressure) of the upper cowling
 with a long K&L duct.  Your version may be easier.   I will play
 with some radiator models on the actual engine.  The smaller the
 rad size the easier it will be to fit.  20x30 will be a challenge.

 In this model,  it seems that two 20x30x1 counterflow radiators
 would give a total volume of 1200ci?   Is it correct that we need
 about 600ci for water cooling and 200ci for oil?   If so, I assume
 that using cross counterflow rads would each need to be about 1/2
 the Dodge size as long as I have 600ci?

 My RX8 has twin oil coolers about 8x10x2.  The single RX7 oil
 cooler is about the same volume.  The latter should fit up front if
 the water cooler intake is a little higher.

 Has anyone used 3D printing to create their engine cowling?

 Scott Gettings

 Those volume numbers depend on the thickness of the core. Thinner
 core less cubic inches are required as thin cores are far more
 efficient for a given volume. Dodge were leading the way with
 other manufactures catching on.

 This is a cross counter flow rad for a Tango

 14" by 24" They are Dodge Neon cores.

 Some what more compact at higher cost to make. You also need a
 skilled tig welder.

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




Scott,

 We should change the standard to square inches of frontal area
 because all the recent cores are around 5/8 to 3/4 inch thick
 including the RX8. Rad core frontal area is more correlated to
 cooling than volume.

 600 square inches frontal area for 250 HP single core. 2 x 300
 square inches  for cross counter flow. Two Neon rads are about
 that. Maybe a bit more so they should handle a mild turbo charge.

 BTW have you considered a 8" prop extension to help the air stick
 to the cabin sides? You can lesson the angel at which they go back.
 If you go back at 7 degrees and then cut it off flat that will
 implement the Kamm effect. Same as the recent Preuses.


 Paul Lamar



 Hi Paul, I noticed during my re-search on the Glass Goose that the
 Icon A5 also has issues back in the cabin sides. They have also
 used vortex generators in that area, on the test planes. The GG has
 the lower wing and the landing gear that also adds to the
 problems.

 Steve Steve Carlisle

 Yes. It is a result of the side by side seating. That Icon is
 almost a scam. Be wary?

 With the Glass Goose it would not be that hard to change it to fore
 and aft seating and narrow the cabin. That lower wing is necessary
 to keep the prop out of the water spray which will destroy it.

 Here are a couple of great books on seaplanes. Extensive
 discussion of the unique design issues.

 Paul Lamar

 Correction fore and aft seating.

 The Kamm 7 degrees does not work out well both side by side and
 fore and aft.

 The first 3D is side by side the rest are fore and aft.

 Of course to build it fore and aft  seating one would have to start
 over. Not practical for those already built.

 Paul Lamar

 IMHO if I were designing a sea plane I would scale something that
 worked well like some of the Dornier's like the DO 26. Google it.

 Actually the Schneider cup racers found out a long time ago floats
 were lower drag than sea planes.

 I also think a coot is a better than a glass goose.


 Paul Lamar

 Hi Paul, I stumbled across this amphibian concept, made in
 PlaneMaker, for Xplane.  It likely has no turbulence issues, a lot
 like the SeaRay amphibian. Applying this parasol wing and aft cabin
 mods together with interplane struts would also lift the prop out
 of the water, no need to put a prop dam on the empenage.

 Inline image 1

 Steve Carlisle

 Looks like it violates the 7 degree rule.

 Paul Lamar



I agree, and the 7 -degree rule will be a problem with any center-pod pusher aircraft.  This aircraft is very similar to the GG, except that the inter-wing distance is much greater than the GG (which is a known problem partially compensated by increasing the upper wing AOI by a few degrees.)

I made some more detailed measurements on my GG.  The widest point of the pylon is just ahead of the upper wing leading edge and is 43.5" wide.   At the leading edge it is slightly narrower at 42.5". The walls regress at about  7-10 degrees for about 10", then rapidly angle in at about 27 degrees to the tip.  The total length of the pylon from the leading edge (LE) to the aft tip is 46".

The current prop position using only Tracy's redrive is 48" from the LE, or 2" past the tip of the pylon.  Most builders add a 6" extension, which would make the prop hub 8" behind the pylon. Looking at the pylon structure, it would seem safe to shorten it by about 10" without affecting the structure.  Limiting the wall regression to 7 degrees would give us about 4" taper on each side, which altogether would give a flat Kamm area about 37" wide.   This could put the prop from 12" to 18" behind a flat Kamm rear end. The engine cowling would also have areas that exceed 7 degree taper and may present a challenge.

I wonder if the lowered overall drag of a Kamm rear end also means cleaner air for a prop at some certain distance behind the flat wall?  The prop is only 62" so may be easily blanked out.   Some builders have used simple, vertical "turning vanes" just outside the separated boundary layer, usually mounted in between the upper and lower wings.  Some have used these in addition to the stock VGs. Is there a practical way to model the airflow at the prop with these various options (I"m not sure that pressure measurements or tufting would give the answer for the prop airflow itself).

Regarding the radiators:  the distance from the LE to the front engine pulley is 14".  The top of the engine is also 14".  A Neon radiator is actually 16.5x26",  so a rad of this size would need to angle up over the engine just a little to fit, which could be done by raising the cowling a little.  The 26"  width would not be a problem.

Thanks,

Scott Gettings

I adjusted the dimensions here and there and had another idea. I came up with a new diffuser
that would handle three heat exchangers in box formation.

The rads are 20 by 12. You might find them in a small Honda.
The top rad can be the oil cooler. The water rads are merely in parallel.
Parallel is better in theory but you might have to adjust the flow in each rad
to equalize the inlet temperature..

They are recessed into the cabin top with water pipes coming out under the spar.

I have another idea to get the air to stick to the pylon sides. A page out
of the SR71 design book :-) More about that later.

Paul Lamar