[RV8RIVETER]

Allan,

I understand what you are saying. All factors influence cooling. I've got to assume Milt built his airplane according to spec, and it appears he has been fighting cooling problems for almost 300 hrs. Now what.....

I'd start another thead, but I don't know how. Somebody do it.....let's call it RV Cooling!

John

You asked for it!

Actually makes sense and could be considered an extension of Bob Axsoms speed mod idea post.

All of the poeple with 200hp cooling problems I have met or read about, used the stock set-up and Van's oil coolers. Van's cooling set-up is pretty much spam can stock and leaves alot of room for air to go all over the place.

It seems pretty logical that if you want to take air in, you want to take only the amount needed to do the job, hence you want to make sure all of the air is cooling something. You also want to smoothly flow and reaccellerate the "used" air back into the airstream. The NASA research seems to bear this out. Dave Anders 200hp RV-4 had (it may have changed) an inlet area of 34 sq inches and an outlet area of 24.7 sq in with exhaust augmentation, per CAFE. A Lancair pilot with has downsized his stock cooling 19.2 sq in of inlet area and his temps dropped 50deg. His website is very good reading and he took thorough plenum/cowl pressure measurements to boot! Here it is.

http://www.n91cz.com/cooling/webCowlrep.htm

Also very good reading is a varieze with inlet AND outlet plenum. he also uses exhaust ejctor pumping. Something I plan on doing.

http://www.ez.org/feature/F0502-1/F0502-1.htm

There is also a good NACA paper 1942 about exhaust ejector configurations for engine cooling, wich increased pressure drop 8 inches of water. Don't have the link, but I can email it to anybody who can't find it.

 

[arffguy]

Dave Anders lecture

Dave recently gave our group a lecture on cooling drag and other speed mods. Here is a link to the handout he gave us.

http://sacrvators.com/Aircraft Efficiency N230A.pdf

A real nice pair of folks, Dave and his wife. Thanks Dave.

 

[pierre smith]

Dave Anders

Hi Mikey,
Did you happen to see how Dave's augmenters were made? Is it an item we can buy or did he make them?
Regards,

 

[Yukon]

Wade,

Great post Wade. This guy is quite a scientist! Wonder what his day job is?

I think the plenum is the way to go. I'm going to use one on my engine. His cowling mods though sure seem like alot of effort for 5 kts.

John

 

[N941WR]

John,

Do you have any plans for your plenum? Like you, I have been thinking of doing one out of aluminum but am looking for some plans or good pictures.

 

[RV8RIVETER]

Depends

The Lancair guys day job is aero eng for BAE Systems. He was nice enough to answer my email questions. He spent the extra time building molds for fellow Lancair guys, which added to the complexity.

As for work, we all know these things are relative. I am on the very slow track, 2yr old twins, new house, budget, ect. So I have alot of time to think and theorize, as well as spend extra time in glass. I have read where the James cowl, while good quality, is alot of extra work. So, my thinking is if I am going to spend alot of extra work anyway, I might as well do the whole thing myself. That way I get to apply my brainstorming, hopefully learn something, and if I save some $$ that's a bonus. My current thinking is to make a carbon/kevlar plenum, and modify the std Van's cowl. Cutting off the nose inlet area and making my own inlets, and then making an ejector tunnel on the exit.

I have yet to decide on inlet geometry (current choice is oval inlets higher than stock to keep the air flow smooth) or if I am going to go to the trouble of some sort of steered exit plenum. Of course things may change, which is why I started this thread.

One other resource of interest is;
http://www.rotaryeng.net/how-to-cool12.html

While its main focus is rotary engine radiator cooling, it has very good info on inlet design and pressure recovery. It also led me to the book by Kuchemann & Weber , Aerodynamics of Propulsion.

 

[Yukon]

Your babies

Wade,

I hear you. Slow process when you are raising a family. Spend lots of time with your babies......they grow up so fast! My little girl starts college next year, and I don't know where the time has gone. Enjoy them while you can.


Have fun!

John

 

[Yukon]

Plenum

Bill,

No plans at this point, still looking into it.

John

 

[Walter Atkinson]

Plenums can be wildy variable in effectiveness. We've tested some really nice looking ones that were not very good and we've seen a few that do a nice job. The issue seems to be that the plenum needs to be as large as possible under the cowl to allow the air coming in to expand and slow down.

As has already been mentioned, the key is getting the air to go where it's needed and exit smoothly. The only way we've ever made much progress in altering baffle designs to to TEST each little change independently as it is made. It's slow, tedious research. We've made some really great, obvious improvements that on testing actually made things worse!

That's why there's a "back to the drawing board" phrase.. Arrrgh!

Walter Atkinson
Advanced Pilot Seminars

 

[RV_7A]

My new plenum

Heres what I've come up with and so far its turned out pretty well. No leaks here guaranteed. Initially it was formed using the upper cowling and then the inlet ramps added and then part molded using high temp epoxy.

-Jeff

 

[RV8RIVETER]

Plenum

John, yes you are right they grow fast. I have noticed already.
It even looks like I have a future building partner,
the boy can't go out into the garage without picking up a cleco!

From the lean of peak thread.

The first thing we learned is that what we learned in A&P school was just, flat, wrong.
The air does NOT come in, go to the back plate, pressurize the cowl and go
down through the fins. Nothing could be farther from the truth.

The BASIC airflow is that the air enters the cowl openning at the outboard
edges of the openning. It passes along the outboard edges of the cylinders
until it gets to the back plate whereupon it turns inboard and meets the air
that came in the other side right over the spine of the engine. This results
in the air moving forward along the spine of the engine case toward the
spinner where it exits the cowl and goes over the windscreen aft! That's
why you get oil spots on the windscreen when you have an oil leak!

Yep, about 70% of the air that comes in a cowl goes right back out the FRONT!
That's why smaller opennings, limiting the air IN and baffles blocking the front
are very helpful. If one tufts the cowl openning you will be surprised to see
about half of those tufts pointing straight forward TOWARD the prop when in flight.
That's the air going OUT forward.

The biggest advantage in cooling is created by slowing the air down on top
and creating a large deltaP with the lower deck. That allows more air to work
it's way through the fins and allow engine heat to move into the air. It is
essentially impossible to make intelligent changes in cowl and baffle design
unless you can WATCH the airflow with the changes. They can be very counter-intuitive.

We've fixed several problems that have existed for decades and we've run up
on some to which the solutions remain elusive. One in particular really has
me frustrated and downright chapped--dozens of brilliant ideas and sure-fire
solutions have failed over many months of efforts.

Walter Atkinson
Advanced Pilot Seminars

Walter I hope you don't mind that I copied this over.
I thought it applied well to this discussion.
That being the case, it would seem the best plenum design
is a two piece seperate plenum for each row of cylinders,
like the Varieze link I posted above.

Does anyone have an opinion on that?

 

[Walter Atkinson]

That's beautiful work.

How difficult is it to remove to pre-flight the engine compartment?

 

[RVbySDI]

Jeff,
Your baffling looks great. I keep getting drawn to the floor though when viewing your pictures. I hope it is just a matter of my not seeing other parts of your plane that are hiding the true stability of that plane but those landing gear legs precariously perched on a small block of 2 X 4 on your nice looking tile floor scares the beegeezus out of me. Please tell me you have that airplane secured so it will not fall off of those miniscule blocks and damage your beautiful work!

 

[Mike S]

Questions for Walt

R.E. your comments about in at the outboard edges of the inlets-----have you expiremented with vertical inlets,tall and narrow, mounted as far outboard as the cowling shape will tollerate??

Have you tested annular, or "Smile" inlets??

What about inlet guide vanes such as my Stinson has??

Thanks, Mike

 

[RV_7A]

Thanks Walter- Its just a matter of removing a hand full of screws with a power screw driver. Takes about 2 more minutes to access. This was designed to work with the stock baffling setup that Vans has you build. So it is almost a plug-n-play plenum.

Steve- They is no way it could fall of the blocks. The threads do a good job of forcing the blocks to slide on the porcelain tile floor. Careful now Steve... "that paranoia will destroy ya"

-Jeff

 

[Tom Gresham]

Jeff: If you guys keep this up, before we know it, Walter will be building one of the cute little planes.

I got him a ride in Newt Courtney's RV-8 (he lives in Lufkin, TX). I could see Walter smiling from 1,500 feet!

 

[gmcjetpilot]

Awesome info

Wade, Great post Wade. This guy is quite a scientist! Wonder what his day job is?

I think the plenum is the way to go. I'm going to use one on my engine. His cowling mods though sure seem like a lot of effort for 5 kts. John

I know 5 kts does not sound like much but its like finding 10 hp laying around. I do think a solid plenum is the way to go and you will pick up efficiency by simpling reducing leaks. (see next post)

Well I do know (ed) Dave Anders is a dentist and holds the cafe foundation dot org "Triaviathon" record in his RV-4. A distant second place was the previous 1st place record holder, John Harmon's Rocket II.

I talked to Dave Anders several times at airshows. He's a very nice Gent and willing to share. Here is a link to the Cafe report and Triaviathon flight test ("Tri" as in a time to climb, min and max speed test, with points awarded per a weighted formula):

http://cafefoundation.org/v1/aprs/RV-4.pdf

(Before the flaming starts from Rocket pilots, I will mention Dave is a maniac when it comes to performance. Pictures of Dave in the report show he is gaunt. Dave lost a lot of weight just for this flight test. Now that's commitment. Of course the IO360 angle valve in his RV-4 is also making who knows how much hp, but it's way more than 200.)

Between Dave and Tracy Saylor, they got me thinking about the topic of RV drag reduction years ago. The lecture notes do not get into the "secrets" or details of how to do it. Dave last I heard was part of Nu Venture (formerly, Questair Venture, you know the fast flying egg). Dave was at Reno a few years ago in his Venture. He had a landing accident. He was fine but the plane was trashed. I guess Alan Tolle is the NuVenture principal, here is the site: http://www.nuventureaircraft.com/

Going 250 mph in a RV-4 is a good trick with gear down and welded. In my opinion if you have dreams of going over 250 mph, than retract gear starts to come onto the radar as NEEDED.

John here are some pictures of plenums with the stock cowl. I stole these off builders sites, none are mine. I do think the "whole package", plenum and round inlets, aka sam james, is better than just the plenum, but the plenum is a big plus. It's also little less wear and tear on the cowl. The wight for the DOG HOUSE may be a little more and access to the top plugs a little restricted:

 

[gmcjetpilot]

One big dog house

it would seem the best plenum design
is a two piece separate plenum for each row of cylinders,
like the Varieze link I posted above. Does anyone have an opinion on that?

Wade you would think two separate left/right plenums would be better, but if
you read the NASA report that I mentioned you referred to (and linked on the
Lancair web site), you see one large plenum is best.

The flow into the right cowl inlet in general is much more than the left. This
is due to the prop wash (p-factor). This effect is made worse by a standard
cowl (like Van's) verses a "NASA" 0.30 or 0.60 cowl (round inlets). The
decimal numbers refer to ratio of inlet air velocity to the free air stream
velocity (Vi/Vo).

Also the plenum volumn acts as a reservoir, so one big plenum should be
better for volumn and balance. With that said I have seen well done left/right
split separated plenums work fine. Why. Well they are better than soft leaky
seals.

The NASA report made this one basic observation LEAKS are terriable for
efficency. It's air doing nothing and robbing the pressure differential between
the area above to below the cylinders. Air just leaking into the lower cowl is
a double whammy. More air inlet area is not better if its not going where it
should. Reducing leaks means you can reduce inlet area and reduce cooling
drag. The Delta Pressure across the cylinders is the mechanism that cools
the engine.

Soft seals will leak no matter how well they are done. A vary difficult area is
the front where the cowl inlets and engine front near the spinner seals. I
have seen excellent soft seal baffles that are works of art, and some not so
good. Bottom line a sealed solid hard cover DOG HOUSE is as good as it gets.
Also as Dave Anders said it's easier to seal the cowl inlet (round) to the
plenum. Here are some pictures I cut and pasted from the NASA report.


http://img133.imageshack.us/img133/9441/nasaexcerpt5bg.jpg

Notice the coefficient of pressure is higher on the right. Its not bad using
a so called NASA/S.J. Miley/Miss State/Texas A&M cowl, designated as 0.30
or 0.60 in the report. However with the standard "STD" stock Piper cowl (not
much unlike vans cowl) you see a big spread between the right to left inlets.

Here is why widely spaced inlets work better, in addition to the reasons that
Dave Anders gave in his lecture notes:

http://img133.imageshack.us/img133/2219/vanscowl3yy.jpg

The prop beats the air to death near the spinner. Prop blade shanks near the
hub are just blunt awaful things that are just a little better than a 2x4 board.
The pressure in the inlet is much higher on the outboard edge, so air just
goes in the outboard edge and out the inboard edge if the cowl inlet, back
into the breeze. This "SPILL" air re-merging with the free airstream is spill
drag and lost efficiency.

That is why the round inlets wide spaced away from the spinner is best. Look
at a P-51. The Mustang spinner is a huge airflow improvment. We can not
do this on our plane with the engine and cowl we have.

**NASA report on Horz engine cooling, 1981 is the tip of the iceberg. There
was more work done in the mid late 70's, published by Miley and Cross in SAE
and AIAA papers. I have seen adaptations or use of this info with varying
degrees of success. The good news is the basic Barnard Holy Cowl/Sam
James Cowl makes good use of the info, but not perfect. I am using a stock
cowl and stock baffle kit to start, but modifying them to be similar to a Sam
James cowl but with differnt ratios, areas and shapes.

Dave and Tracy both run custom cowls and plenums, not off the shelf or
after market. Nothing wrong w/ a Sam James cowl if you are not prepared to
size and shape your own cowl inlets, diffuser and plenum. This could be a fun
area to play around to get a few Ka-nots.

 

[penguin]

Cowl to Plenum transition

The question I have is what is the best way to transition from the cowl inlet into the plenum? I have an RV-6A, using a stock cowl, with the lips glassed into the upper cowl - how have builders transitioned to the plenum? I am suffering from high CHTs and am thinking of installing a plenum to lower the temps. All of the pictures I have seen show hard edged plenums with no baffle strip to mate to the cowl. I am thinking of using the lips, but might have to cut them off to get a good entry for the cooling air - does anyone have any advice or pictures?

Pete

 

[RV8RIVETER]

George

FYI:The guy Yukon was talking about was Christopher Zavatson, Lancair driver and BAE aero engineer.

I did not read in the report where they tried two seperate upper plenums. I will read it again. But in the meantime, from the graph of the .03F configuration, the two sides are almost identically symmetrical. I don't think there is enough difference there to be significant. But, on the other hand, a one piece is easier to make and maybe the addition of a flow divider or vane to break-up any circular flow patterns would be a good idea.

Of course it is a great thing that they mention that ground testing is a valid method for gathering flow data. I will definitely be purchasing a high flow blower to take preliminary data with different configurations, while building.

I just thought of something! What about you guys out there with one piece plenums, making a plexiglass plenum cover. Then tuft various areas of your engine compartment and hook up a blower. It may prove an interesting sight.

 

[RV8RIVETER]

The question I have is what is the best way to transition from the cowl inlet into the plenum? I have an RV-6A, using a stock cowl, with the lips glassed into the upper cowl - how have builders transitioned to the plenum? I am suffering from high CHTs and am thinking of installing a plenum to lower the temps. All of the pictures I have seen show hard edged plenums with no baffle strip to mate to the cowl. I am thinking of using the lips, but might have to cut them off to get a good entry for the cooling air - does anyone have any advice or pictures?

Pete

Most of those I have seen take the plenum ramp right into the cowl inlet. You could also make a fiberglass scoop that mates to the cowl inlets and the other mates to the plenum.

Though, I agree with George that to get really good effect from the plenum you need to abandon the stock inlets. Of course you could try, hey we are all learning here hopefully. But, you would definietely want to take steps to prevent any flow from one inlet just going out the other.

 

[gmcjetpilot]

Yep times 2

Pete and Wade:

Pete, yea the transition is a pain, and that is why I went with the round inlet. With the stock rectangle cowl inlet, bottom & sides are easy. The top I'm not so sure? How to bridge the gap between fixed cowl and shaking plenum with a smooth leak free seal is a challenge. When you have one lower seal and a separate upper seal you will have corners to deal with. The SJ cowl uses a separate ring, that's secured to the plenum via a flex duct. When you take the cowl off the rings are just hanging there. With the cowl on, the rings sit in a groove. http://www.romeolima.com/RV8/Cowl.htm
(rectangle is a little more difficult but not impossible. With round you can use hose clamps.)

Wade: Thanks I got confused with who Yukon was asking about, got it Lancair guy not Dave.

The left / right separate plenum vs one piece, yes, I agree in cruise the diff is small but not totally insignificant. In climb there is a bigger difference, of course from (p-factor). However the separate L/R plenum works ok, agree. I just don't think it's better. From a construction standpoint, one big box and cover is easier to build than two small ones; one single volumn plenum is my prefrence. As far as tuff idea, interesting. I want to understand this plenum area more and play with guides, deflectors and internal baffling down the road. The NASA sponsored info was never really finished. It just ended. By the way, my old 1958 Piper Apache Twin used a dedicated dog house plenum. hummmm 1958, so its not a new idea. It also had exhaust augmentor tubes! state of the art 1958 style. It was suppose to do 172 mph cruise, top 182 mph. I don't know about that, but I could do a 160 mph econ cruise (about 16 gal/hr). Not bad for a big cabin 5 seater. You can get up and walk to the back seats (provided you have a copilot of course).

 

[zkvii]

Hi,

Well I can agree that getting the dog house is fairly simple, the transition from cowl to plenum is NOT. We have gone for a custom version of the cowl and have 2 AL .025 rings mounted - not too hard. The fibreglass work to get between the two is proving difficult however. Any ideas would be good.....

WRT to tufting plenum - interesting. Does this mean the prop doesn't need to be moving for testing - just high speed air going into the inlets? How fast does the air need to be moving? Are tufts just pieces of wool?

Carl

 

[Walter Atkinson]

**have you expiremented with vertical inlets,tall and narrow, mounted as far outboard as the cowling shape will tollerate??

Have you tested annular, or "Smile" inlets??

What about inlet guide vanes such as my Stinson has??**


No, we have not.

BTW, one plenum is better than two left-right sections. It allows fro more expansion of the air and better slowing of the velocity. That is advantageous.

Walter

 

[RV8RIVETER]

yarn

Great discussion! My "brain hurts" already!

Hopefully the experts will chime in, but would think fine knitting yarn or even that annoying thin plastic fake grass would work.

While not the same as flight (hard to find a 160mph blower) , high speed blown air preferrably a few inches before the inlet, would give a good approximation of conditions inside the plenum. You could also change the angle of the blower to test climb conditions as well. Of course Walters camera post is an idea. I would think ground testing would be more detailed and easier.

 

[hevansrv7a]

While not the same as flight (hard to find a 160mph blower) , high speed blown air preferrably a few inches before the inlet, would give a good approximation of conditions inside the plenum. You could also change the angle of the blower to test climb conditions as well.

My Sears shop vac claims 170 mph blower speed.
h

 

[gmcjetpilot]

It's the shape

**have you expiremented with vertical inlets,tall and narrow, mounted as far outboard as the cowling shape will tollerate??

Have you tested annular, or "Smile" inlets??

What about inlet guide vanes such as my Stinson has??**


No, we have not.

BTW, one plenum is better than two left-right sections. It allows fro more expansion of the air and better slowing of the velocity. That is advantageous.

Walter

I think vertical inlets would look funny, personally and not fit into the natural cowl shape, while providing the needed area. Sealing that shape to a plenum would also be a challenge. Round happens to fit within the boudary of the cowl. I don't think shape is a huge critical factor; other shapes will work, like an oval, but round has the obvious advantages of easier clamping and sealing to a round flex duct. How many rectangle or oval hoses are there? (Actually there are a few.)

Don't forget that round inlets (proper ones) have the surface profile of a thin low camber laminar airflow airfoil (wing). There is way more to it than punching a hole in your cowl.

"Annular smiley lip" - Walter, Richard VanGrunsven tried it on the #1 RV-6 prototype in the mid/late 80's. It had an annular inlet. He flew it quite a bit. I flew the plane, but when the kit came out, he went back to the RV-4 style cowl. The small spinner and blunt prop blades near the spinner beating the living heck out of the air was still a problem. The other issue is ducting air up from a lower smiley inlet above the cylinders for down draft cooling. It's a bit of a mess. Also you have to get induction air out of it. It turns out three inlets: left cylinders, right cylinders and induction works is simpler. The Cafe foundation tried an annular inlet and huge spinner on their Mooney w/ some success. It looks odd but it has some advantages (report below).

The P-51 is an inspiration to many, including the lower lip induction scoop. There is a huge differnce in the spinner and prop between a RV and P-51. The Mustangs gigantic spinner and prop blades (airfoil adjacent to the spinner) where no accident. Also behind that annular scoop is room for a long diffuser, running the whole length of the engine to the firewall.

FACTOID: Did you know a jetliner turbofan engine has a fan spinner, that if damaged or removed the engine could not run. With out a spinner the engine shuts down from disturbed air flow into the compressor. The flow around that fan spinner is so critical design and manufacturing calls for extream alignment and smoothness. When GE, P&W, RR do impact test they are as worried about the spinner as much as the fan blades. When the fail a blade (and they have to by regs with an explosive charge) it is spectacular.

"inlet guide vanes such as my Stinson" - I know what you are talking about, but aren't those just a fashion statement, aka, like car grills of the late 40's and 50's. The Globe swift had them. I could be totally off base, my friend has a 108 he's rebuilding. I don't recall the grill / guide vanes. However there's no reason one could not put guides in a RV inlet, to reduce transverse flow. You see these on wings: stall fence, VG's and other aerodynamic helpers. Aerodynamics of a wing is not a lot different than a cowl, it's just inside out.

(Meet my little airflow helpers)
http://cafefoundation.org/v1/aprs/localflow1.pdf
(P-51 spinner and annular inlet on Mooney)
http://cafefoundation.org/v1/aprs/localflow2.pdf
(Globe swift - highly modified)
http://cafefoundation.org/v1/aprs/localflow3.pdf

 

[RV8RIVETER]

My Sears shop vac claims 170 mph blower speed.
h

Well what do ya know, I never would have thought that.

I believe volume would be important to. So as long as there is a relative high volume of air, why not?

 

[penguin]

How about a little more exit area?

To encourage the air to go through the cylinders would some gills on the bottom of the cowl help? I'm thinking that a reduction in pressure on the underside of the engine would encourage to higher pressure air in the plenum to flow through the cooling fins. By fitting some gills/vents/louvers to the bottom of the cowl (say a pair of 3"x3" vents for around 18 sq in of additional exit area), sticking down into the airflow to create a reduced pressure area behind each louver, more air might be drawn through. I don't really want to hack holes in my cowl on the off chance that it might work so does any one have any experience of such a mod?

Pete

 

[gmcjetpilot]

Cool idea

To encourage the air to go through the cylinders would some gills on the bottom of the cowl help? I'm thinking that a reduction in pressure on the underside of the engine would encourage to higher pressure air in the plenum to flow through the cooling fins. Pete

Vents will help cooling, but at high speed cruise drag is too much. For me its not MORE cooling its less cooling drag at high or cruise speed. I think most RV's are adequately cooled, except for the IO360 angle valve.

I wounder why there are not more cowl flaps on RV's. I guess they are hard to fit? Here is on guy that tried it, looks nice.
Cool idea. Appears he did it to lower oil temps in climb (15F), not for speed:
http://www.nwacaptain.com/ (RV-8 web site/cowl flap)

I guess Van's keep it simple philosophy is anti-cowl flaps; It is hard to argue with success. I just can't see installing a cowl flap easily on a RV-7 and how much improvement in speed will be gained. One inventive RV'er with a rotary had variable INLET's:. http://www3.telus.net/haywire/RV-9/C-FSTB.htm (towards the bottom). It does not say but are the inlets are computer CONTROLLED? What about the auto throttle? FL250! Cool. Not a lot of info but some interesting pictures.

 

[Ben Beaird]

To encourage the air to go through the cylinders would some gills on the bottom of the cowl help? I'm thinking that a reduction in pressure on the underside of the engine would encourage to higher pressure air in the plenum to flow through the cooling fins. By fitting some gills/vents/louvers to the bottom of the cowl (say a pair of 3"x3" vents for around 18 sq in of additional exit area), sticking down into the airflow to create a reduced pressure area behind each louver, more air might be drawn through. I don't really want to hack holes in my cowl on the off chance that it might work so does any one have any experience of such a mod?

Pete

I installed a set of louvers in the lower cowl of my -6A. They lowered the CHT's by 30F. At the same time I also 'glassed a lip to the outlet of the lower cowl in order to try & draw more low pressure air out. I think that helped as well. But the louvers had the most effect.

I bought the louvers from Vans. I believe they were originally intended for the RV-10.

Ben Beaird
Plymouth, WI
-6A 143 hrs

 

[RV8RIVETER]

Pumping

To encourage the air to go through the cylinders would some gills on the bottom of the cowl help? I'm thinking that a reduction in pressure on the underside of the engine would encourage to higher pressure air in the plenum to flow through the cooling fins. By fitting some gills/vents/louvers to the bottom of the cowl (say a pair of 3"x3" vents for around 18 sq in of additional exit area), sticking down into the airflow to create a reduced pressure area behind each louver, more air might be drawn through. I don't really want to hack holes in my cowl on the off chance that it might work so does any one have any experience of such a mod?

Pete

You also could use the free energy source you have. The exhaust. You could use the exhaust to "pump" the air from the lower portion of the cowl. Also it would be a good idea to help direct the air to your exit, that 90deg of firewall before the exit is probably a big hinderance. Dave Anders installed a 45deg ramp on his firewall to help smooth that transition.

 

[penguin]

You also could use the free energy source you have. The exhaust. You could use the exhaust to "pump" the air from the lower portion of the cowl.

Wade,

I don't really follow what you are suggesting. Are you saying that the exhaust pipe be cut to end just inside the cowl? I think that would induce a lot of vibration in the floor boards - I lengthened my pipes back to the stock length a could of years ago to greatly reduce the floor vibration. Or are you suggesting some kind of inductor just upstream of the cowl exit? Do you have any pictures?

Pete

 

[chuck]

Wade,

I don't really follow what you are suggesting. Are you saying that the exhaust pipe be cut to end just inside the cowl? I think that would induce a lot of vibration in the floor boards - I lengthened my pipes back to the stock length a could of years ago to greatly reduce the floor vibration. Or are you suggesting some kind of inductor just upstream of the cowl exit? Do you have any pictures?

Pete

Lot's of people talk about this (e.g. google on 'Exhaust augmentor' and Charlie Airesman, and NACA 818). Many examples shown on varieze and other fiberglass AC. As I've mentioned before, look at Kent Paser's, Speed with Economy. He details how he did exhaust pumping.

Lot's of RV folks talk about it but I have not found a single picture of the setup. Presumably this is because these modifications are made by racers and unfortunately they tend to be vague on what they are doing. If you read Dave Anders notes about going fast (and reducing cooling drag) he gives a ton of information about the changes he made, why he made them and how much faster he went. There is zero detail about how he implimented it - no pictures, sketches, dimensions etc. I suppose that would be like looking at the answers in the back of the book.

George (aka gmcjetpilot) has posted some links.

 

[Mike S]

Walter, George, thanks for addressing my comments about inlet shape etc.

When I read the info about air comming in at the outer edges of the inlet, and going back out at the inner edge, it made me think of the shapes I mentioned, as a way to control the airflow.

Seems that is there is air comming back out of the inlet, there is either TOO much comming in in the first place, I.E. inlet size too big, or there is actually a lower pressure area toward the inboard side, turbulance from the spinner/prop root airfoil??

A pair or anular inlets, oriented verticaly, located at the outer corners of the cowl is what I thinking-------maybe even add some guide vanes to help smooth the airflow.

Anyone done testing to see the pressures inside the cowl in flight, right at the inlet, back in the plenium, and the lower cowl/exit side?? You could have lots of pressure in the cold side of the cowl, but unless there is sufficient differential in pressure drop in the hot side, there aint gona be no cooling.
IMHO, air volume is the determining factor more than pressure.

Mike

 

[walter]

It's been mentioned that the airflow coming off the spinner root and entering the
cowl is quite disturbed. Some have mentioned that the air may actually be
exiting the cowl in that area. I wonder if a VG placed along the inside of the
cowl inlet just behind the spinner root might help. Heck, if that helps how about
some VGs placed on the top of the cowl or sides of the baffling or on the upper
and lower surfaces of the inlet. Why stop there, how about the cowl exhaust
area? Just thinking out loud.

Another idea that I've been tossing about with Dan is the potential for the lower
baffles, on the bottoms of the cylinders, being too close to the cylinder fins,
potentially limiting the amount of airflow. Akin to the thought of spacing the
rear baffle further away (1/8") from the backside of the #3 cylinder with washers.

Another idea (probably not a good one) is the use of vertical runners on the
bottom of the cowl (inside of course) directing airflow towards the cowl exit,
maybe an inch or two tall.

Time for some experiments.

 

[Lorin Dueck]

Wade & "Yukon" -

I work with Chris Zavatson (Lancair 360) N91CZ - AKA your "scientist".
He has a wealth of knowledge, and computer skills.
BS in Mechanical Engineering & Aeronautical Engineering.
Brilliant at composites....
... and VERY careful to make sure his web site posts (www.N91CZ.com) are correct & accurate.
One of the best guys you'll ever meet.

As you can imagine - the "Spam Can vs. Plastic airplane" jabs run rampant at work!!!

Lorin D
9A Wings (Tanks)
(N194LD reserved)

 

[RV8RIVETER]

Pump

Wade,

I don't really follow what you are suggesting. Are you saying that the exhaust pipe be cut to end just inside the cowl? I think that would induce a lot of vibration in the floor boards - I lengthened my pipes back to the stock length a could of years ago to greatly reduce the floor vibration. Or are you suggesting some kind of inductor just upstream of the cowl exit? Do you have any pictures?

Pete

Pete

Keep in mind I am still building my -8, so the particulars may not apply, BUT. The exhaust pipe will end BEFORE the exit of the lower cowl. I am not saying to cut the exhaust pipe. I am more eluding to extending the exit past the pipe exit. Basically we are building a venturi. This configuration would probably work better on a 4 into 1 exhaust (which I have) , but you should still be able to use it on a 2 pipe exhaust. In that case it would probably be better to build two ejectors and outlet tunnels. There is a NACA paper on exhaust ejector pumping of piston cooling. If I have distilled the data correctly, the recommended length for our applications is a mixing tube (straight pipe past the exhaust outlet) is 18 inches and a diffuser of approx 6 inches (outlet diameter larger than mixer). I am sorry, but I do not have the inlet/outlet ratio with me or the exhaust/mixer ratio. I can email the paper to anyone who can't find it. Now I know this is a long distance. The Varieze listed in the first post had a much shorter length, and he was able to feel air draw when at idle (nice the prop is out of the way eh). I am sure we could get a positive effect from a shorter length, just not as much as the optimum (which is 24" mixer + 8" diffuser). How much shorter will work, ?? . Again work/reward ratio is personal, but to me it is better to improve cooling while decreasing drag. Have not been able to assemble my engine yet to see what I can get away with.

Chuck: Yes, info on these issues/modifications is very hard to find, and concrete graphic examples even harder. Another reason for starting this thread. I don't know why, I guess people are reluctant to give away info they have worked so hard to obtain. ?? The NACA paper is pretty helpful though.

Mike S.: Chris Zavatson took detailed pressure readings on his plenum design for his Lancair.

Lorin: Lucky you. Chris's work is indeed very thorough and detailed. He was also very kind enough to answer some of my questions. Would love to be able to watch him do some of his composite work.

 

[dan]

Another idea that I've been tossing about with Dan is the potential for the lower
baffles, on the bottoms of the cylinders, being too close to the cylinder fins,
potentially limiting the amount of airflow. Akin to the thought of spacing the
rear baffle further away (1/8") from the backside of the #3 cylinder with washers.

Actually, what I was talkin' bout is widening the actual opening in the outboard lower cylinder baffles. Not spacing them away from the cylinder, but making the opening itself bigger. If it's 2" now try 3" and close in on it from there...just my theory.

 

[walter]

Actually, what I was talkin' bout is widening the actual opening in the outboard lower cylinder baffles. Not spacing them away from the cylinder, but making the opening itself bigger. If it's 2" now try 3" and close in on it from there...just my theory.

I agree with Dan. But I'm also thinking that the "nearness" of the lower
baffles to the fins may have an effect as well. Since most of the Baffling kits
sold by Vans are the same, for this engine, perhaps I have mine a bit too
tight. I did make them very snug in my installation. I plan to do
a lot of testing as I get to bring the plane closer to home this weekend.

And as Dan mentions, the "gap" may not be wide enough. It may also not be
very efficient as an exhaust as well, given that it's openings have 90 degree
bends on the exhaust lip.

Well all this is just speculation, I'm no scientist. But it will be fun to play with.

 

[rv8ch]

Theoretic Calculations

This is a great thread. Given that it takes so much work to make changes to
the cowling and baffling, I wonder if anyone has done some theoretical
calculations to determine what the minimum cowling inlet size should be
for a particular engine generating a specific amount of power?

All the variables are known - air mass, heat generated,
desired delta Ts, air velocity - just need to solve for volume
and thus cowl inlet size.

Once you have this number, then you will know how efficient your
setup is, and you can gradually work towards reducing your inlet
size until you get the optimimum efficiency.

For example, if the calculation for your engine/power config says
that a minimum cowl opening of 400cm2, and yours is 800cm2, then
you know that your cooling system is 50% efficient.

 

[RV8RIVETER]

Open Gap?

I agree with Dan. But I'm also thinking that the "nearness" of the lower
baffles to the fins may have an effect as well. Since most of the Baffling kits
sold by Vans are the same, for this engine, perhaps I have mine a bit too
tight. I did make them very snug in my installation. I plan to do
a lot of testing as I get to bring the plane closer to home this weekend.

And as Dan mentions, the "gap" may not be wide enough. It may also not be
very efficient as an exhaust as well, given that it's openings have 90 degree
bends on the exhaust lip.

Well all this is just speculation, I'm no scientist. But it will be fun to play with.

WALTER

I am not sure I am following you. You want to open the space between the baffling and the fins?

But, we want the air to go thru the fins. If it does not go thru the fins it is not cooling anything. And if it does not go thru the fins, the pressure drop is not sufficient.

 

[RV8RIVETER]

Lycoming chart

This is a great thread. Given that it takes so much work to make changes to
the cowling and baffling, I wonder if anyone has done some theoretical
calculations to determine what the minimum cowling inlet size should be
for a particular engine generating a specific amount of power?

All the variables are known - air mass, heat generated,
desired delta Ts, air velocity - just need to solve for volume
and thus cowl inlet size.

Once you have this number, then you will know how efficient your
setup is, and you can gradually work towards reducing your inlet
size until you get the optimimum efficiency.

For example, if the calculation for your engine/power config says
that a minimum cowl opening of 400cm2, and yours is 800cm2, then
you know that your cooling system is 50% efficient.

Mickey

The air requirement for the Lycoming O IO-360 engines is here,
http://www.n91cz.com/Pressure/PlenumPressure.pdf
Figure 15.

 

[rv8ch]

The air requirement for the Lycoming O IO-360 engines is here,
http://www.n91cz.com/Pressure/PlenumPressure.pdf
Figure 15.

Thanks Wade. That's a very interesting paper.

 

[BruceMe]

Chillin like a

I agree that it's interesting, but unfortunately, it doesn't answer your question.

Minimum inlet size is not a simple hp * constant equation. There are many factorys:

HP, Cruising Speed, climb speed, climb power, desired maximum temp.

Many of those very sexy and skinny inlet high-permance planes on the line at OSH can't climb indefinately at full (or even half) power. RV's that aren't really tightly cowelled have the same problem in varying degrees. Pilots avoid the problem with a "cruise-climbing" which typically fixes the issue on marginal installations.

There's a lot of "cool stuff" happening around engine installations right now that will increase flow and decrease drag and this is a great thread!

I wonder if anyone has head-to-head comparisons of a newer plenum systems vs. the classic rubber-seal systems in termps of real-world cooling improvements and drag?

Also... What would be involved in adapting a rubber-seal system to a plenum? Looking at pictures, I could shape some foam/layup some glass and be pretty close.

-Bruce

 

[gmcjetpilot]

Baffle gap picture I drew up

click me once or twice, watch me grow

Most find noticeable changes with a washer thickness change.

 

[RV8RIVETER]

baffled cowl

click me once or twice, watch me grow

Most find noticeable changes with a washer thickness change.

OK George, thanks for answering my question. By opening the gap more air flow thru that area, with the extra air being dragged along spilling thru the fins.

But this senario really only pertains to non-plenum configurations. Would you agree? In a tight well executed plenum the delta p in top/bottom pressure would push air thru anywhere it could.

 

[gmcjetpilot]

Baffles, Plenums, inlets and diffusers

But this scenario really only pertains to non-plenum configurations. Would you agree? In a tight well executed plenum the delta p in top/bottom pressure would push air thru anywhere it could.

No I don't think so, regardless of whether you have a "hard top" on your baffles or use soft seals against the cowl, you can play with gaps. I agree a good plenum should be better, but it can't push air thru solid aluminum.

The idea behind gaps is just balance the CHT's. Look at those fins in the front of #2 or back of #3 jugs adjacent to the baffles. There's just no area with the baffle tight against the short fins for air to go. If your CHT's are acceptable, than may be this is moot, but we are talking about balancing the CHT's. This is also why we cover up almost the whole front of #1 jug (front right), it's over cooled (deep fins direct into the breeze). There are plenty of RV's flying around just fine with less than perfect baffle gaps and unbalance CHT's. However in some cases people see very high #3 CHT's, adding the gap is goodness and needed. 2nd hand info, I have heard of 30F drops in CHT with a small increase in gap. Of course there's a limit to how much you can do with gap alone. If the rest of the cooling system sucks not even gap adjustments will help. Its just one small piece of the puzzle.

I don't look at the plenum as increasing cooling but reducing cooling drag, acceptable cooling with less cooling drag, aka speed. Also one change in part of the system requires other changes. A plenum that does not leak can only do so much. No one part can make a differnce if there's a choke point or horrible detail in other area. You are only as good as the worst part in the "system".

Most folks start with Van's baffle kit and just add the top anyway, so most RV's share the same sheet metal bottom baffles. Those "choked" gaps on jug #2 and #3 are still "adjustable" regardless. With the hard top attached to the baffle you might think its too stiff and restrictive to be able to adjust the gaps. That's why we have hammers and wedges. Little spacers can make out of silicone might work as well.

Think of the cooling system as four parts.

The top plenum, is either a piece of aluminum or fiberglass you bolt onto the baffles or it's a bunch of silicone strips resting against the cowl. Either way a top plenum is formed.

The bottom is what I call the baffles. This includes the side, back and bottom, e.g. Vans baffle kit. These parts rest, rub and bolt directly to the engine. We can adjust and play with gaps, independant of the other parts of the cooling system. The job of the bottom baffles is guide air around, into and through the fins. There is some SIDE leaks I would like to reduce. I know a guy that left the standard center baffle between the cylinders off his newly overhauled engine (O235 on a Grumman). He went out to break-in the engine and proceed to melt his piston rings. Ouch

The third part of the "system" are the inlet's and diffusers. This part takes inlet air and leads it to the top plenum. This is a little tricky since there's no way around joining the fixed cowl with a moving engine. With soft seals, sealing the front part of the engine and cowl is a challenge.

The last part the air must travel is the lower plenum and exit.

The principles are simple. For a top plenum, no leaks. The principle for the baffles is guide the air where it is needed with NO (side) leaks or stagnation (choke) points. The inlet diffuser's job is a smooth turbulent and leak free path for air to get from the inlet to the plenum, while slowing down. The lower plenum and exit does not get the attention it deserves, but the comments made so far are right on. The lower plenum needs to re-accelerate the air and discharge it parallel to the free airstream, with minimal turbulence.

 

[Walter Atkinson]

We have significantly improved cooling on some models with cooling louvers on the lower cowling. If positioned optimally, they have little drag penalty. We have taken this success to other airframes and in some cases with dismal results. In one case I recall, the cooling results were WORSE with the louvers. This stuff is not intuitive and requires in-flight testing.

BTW, after you find out that the leaf-blowers do not work for this project, you'll at least have a very clean driveway. (don't ask me how I know. I have a bizillion mph leaf blower, now used for my driveway...) <g> Short of a complete wind tunnel, this needs in-flight testing!

Arrrrgh.

Walter

 

[RV8RIVETER]

No leaf blowers

Walter

I am sorry I used the term "blower", I was not referring to a leaf blower in my earlier posts. I do not think, and your experience seems to verify, that it will not put out sufficient volume.

The NASA paper clearly states that ground testing is the best method to determine internal flow dynamics. And their ground test set-up looks fairly strait forward. I think since we are trying to convert fast air to air pressure, and air pressure delta is what is driving the cooling flow that a blower with adequate volume and pressure output is all that is needed. As well as some method to lower outlet pressure. Since I will be using an ejector I am planning on using my shop air to replicate exhaust flow and drive the cooling air outlet pressure lower. Although maybe a shop vac would work as well?

Can you tell us what your experiences were and why they did not prove adequate? Did you have airflow past the cowling outlet or a vacuum source, to help lower pressure?