Do it
N941WR said:
I do realize that I will have to transition from the round opening to the cooling plenum. Then again, maybe not as I think the smaller opening might work out fine w/o having to worry about pressure drop due to the reduced size. Thoughts on that?
You can modify your stock cowl with round inlets, just keep in mind the secret starts at the lip and continues through the throat, duct to plenum and plenum.
I would buy some rings from Dave Anders of RV-4, 265mph cafe foundation winner (by 80% over previous cafe foundation high scorer, Harmon's Rocket HRII). SJ rings are fine but not quite as well contoured per the nasa report. You can also make your own by fiber glassing them in. Just keep in mind it is 3-D flow and the throat is critical, e.g., no big area discontinuities.
You'll have to glass in the rings, and make the transition from the rings to the plenum. The flex part is usually made of silicon or neoprene. You can have that made or actually cut out from an old wet suite. The plenum can be made from a stock Van's baffle kit and covered with aluminum or fiberglass. Stiffen the top of the plenum in the middle and make a bracket to one of the center case bolts.
It is more work but the results will be worth it. Now for size of rings you can get by with less than the 25-33 in-sq I have. In general you can cut inlet down by 40%. I don't know what area Van uses for the O-235 cowl but measure it. Than note what temps people are getting out in the field on a 100F day. If they are seeing acceptable temps with the stock cowl, you can scale down the inlet area by approx 35-40%. Why? Less leakage in the plenum, more efficient inlets (higher pressure and no reverse flow) to name two big reasons.
bumblebee said:
Most cooling drag is due to the difficulty in achieving proper exit-air management.
Go to
http://www.sdsefi.com/air41.htm for a very concise explanation of the problem and difficulty in finding solutions. This particular installation is liquid-cooled, but the fundamentals are the same.
I have to dis-agree most cooling drag is NOT from the exit. You make it sound like this is not a worth while mod at all, and all the improvement is in the exit. The point air leaves the free air stream and is "captured" by the cooling system till it exits and rejoins the free airstream is important. No one area is more important and one poor detail can negate the gain in another. If you improve the efficency of the air you use so you can reduce the inlet (and exit) area you will reduce cooling drag.
The SJ cowl reduces the inlet area by about 40% and gains up to 8 mph cruise. That is like finding an extra 10-15 hp! Drag reduction rules. The inlet design tends to be more critical than the exit because of the higher velocities and need for efficient pressure recovery. I said more critical (design) not more important. The exit is important, but it's not as critical or hard to achieve a good exit. We know a reverse scoop with parallel walls at a 30 degree angle and "tricks" like exhaust augmentation (accelerating the cooling air with the exhaust gas) work. Also Variable exit geometry (aka cowl flaps) work. However the problem is the aircraft structure is fixed. You have to do major structural modification on a RV to adapt some of these exit principles. The inlet is fairly easy and has known benifit for our air cooled engines. As well we know soft seals and the cowl to baffle leakage is high with the stock Van's set up. Some builders do a better job than others, but the "SJ" style system takes out the guess work of, "is it leaking".
The nasa report did research exit design, but they emphasised the inlet and high pressure side (plenum) of the cooling system more. The exit is not un-important but it is fairly well understood. It is the inlet that was miss understood. No matter how great your exit is, it does not matter if you have a leaky system with poor pressure covery inlets with weird reverse and turbulent flow.
www.sdsefi.com does great work on his water cooled Subaru powered RV and his efforts do have similarities to air cooling in that we're talking fluid dynamics and aerodynamics. However they're very differnt in principle and heat transfer. Just look at the link and see the cowl sprouting a bunch of scoops. Does not look like any air cooled RV cowl. That should give you a clue, it is a differnt animal. With air cooled we go right from engine (metal) to air. Going from engine to liquid to air is a bit different, you have a middle man. Yes water cooling is great in a car, but in an "AIR-plane" we have plenty of air. The challenge is getting water cooled engines cooling drag down. The P51 did it with very thick radiators in a belly scoop, and it also had cooling problems as does the Subaru. However the P51 was going 400 mph and that is a differnt world and thick radiators do not work. Water cooling is great for getting the heat out of the the engine but getting it into the air is the issue for those guys. sdsefi is doing good work but it has limited benifit to us Lycoming flyers.
The good news is we have +80 years of research in air cooled engines and are the beneficiary of this research. NACA (now NASA) has studied air cooled engines extensively from the 30's thru the 40's until jets came into their own. Of course most work was on radials and inline but there is good info there. Later in the early 80's the NASA funded report done by Miley/Owens/Cross of MSU and TA&MU investigated Horz opposed aircraft air cooled engine cooling efficency. This is where the wide spread round inlet with internal airfoils and sealed plenum data came from. Keep in mind this is not if it will work, it does work. This is not a debate.
The secondary benifit with a "dog house" plenum and the round inlet for us, is you can seal it with a round duct (and clamps), achieving a near leak free system. Just reducing leaks is a 38-55% gain in efficency. If you improve efficiency you can reduce inlet area. Reduced inlet area equals less cooling drag. This is independant of the exit. However a poor exit can ruin the gains above. It all has to work together, a weak link is a weak link.
The bottom line if you are going to do it, do yourself a favor, beg, borrow and steal the ideas of LoPresti, Sam James (designed by Barnard) and Dave Anders. You don't have to buy the rings; you could form them from fiberglass. No doubt you can make your inlets a little smaller than those for a O320.
You can also control or
throttle the flow through the cooling system with the exit as Bumblebee alluded to. The exit is a way to control the flow. However you have to start with efficient inlets and a leak free plenum. That is what the round inlets with leak free connection to the leak free upper plenum do. This area represents 55% loss in a typical GA plane installation from leakage alone! A well built RV with a stock cowl and soft seals may be better than a typical GA plane, but you would be surprised how the air flow is working in the inlets, backwards. No matter how nice your exit is, poor inlet design will kill the efficiency of the system, resulting in high temps or excessive cooling drag or both.