Direct Drive Rotary?

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

PTAirco

Well-Known Member
Joined
Sep 20, 2003
Messages
3,507
Location
Corona CA
Yes, thank you, I'm well aware of that.

The fact is that what it REALLY boils down to is that how much air and fuel you move through the engine is what determines the power of the engine. You can make more power by revving it up more or you can make more power by giving it more boost. Either way, if you pump the same amount of air and fuel through it, EITHER way, PSRU or boost, it will make exactly the same amount of power, either way, same power, OK?

That's the fourth and last time I'm going to attempt to explain that one, sorry for my failure the first three times.

Maybe one of you engineers can help me out here, or maybe we should just keep it a secret :)
It's a valid approach to increasing the useful power of a direct drive, but I don't think you can make a straight comparison. Twice the revs or twice the amount of air/fuel going through it on paper may make the same power, but to force twice the normal amount of air into an engine takes considerable power, more than a turbo alone can provide, I would think. Remember also that a turbo will only work as hard as the engine - at low RPMs the available exhaust energy is not that great. And if you resort to supercharging, well, that also takes power away from the engine.

I am playing devils advocate here - I intend to run a direct drive engine myself and am seriously looking at turbocharging too, having just been across the Rockies twice last week, it makes almost no sense not to have a turbo on any aircraft engine.
 

Topaz

Super Moderator
Staff member
Log Member
Joined
Jul 29, 2005
Messages
13,989
Location
Orange County, California
Yes, thank you, I'm well aware of that.

The fact is that what it REALLY boils down to is that how much air and fuel you move through the engine is what determines the power of the engine. You can make more power by revving it up more or you can make more power by giving it more boost. Either way, if you pump the same amount of air and fuel through it, EITHER way, PSRU or boost, it will make exactly the same amount of power, either way, same power, OK?
Errr... no. A PSRU doesn't change the power of the engine at all. It simply allows you to use a larger, more efficient propeller to turn whatever power you've got into thrust. Adding a turbo may increase the power output of the engine, but if that power is shoved into a sub-optimal propeller, you've gained nothing.

Ultimately, it's how much thrust the propeller produces that matters. Turning a relatively larger prop more slowly is the way to go, if you can manage it. Once the tips go sonic, it doesn't matter how much power you pump into your propeller. The efficiency drops so low turning that power into thrust that you end up with a net loss compared to a better setup. It's surprisingly easy to get propeller tips up into the transonic speed range. Many airplanes do, which is one reason you can often hear more propeller noise than engine noise.

A turbo isn't a direct substitute for a PSRU, and you have to look at the entire engine-propeller combination to determine the amount of thrust produced. In the end, the thrust coming off the prop is the only thing that matters.
 

Dana

Super Moderator
Staff member
Joined
Apr 3, 2007
Messages
8,952
Location
CT, USA
Errr... no. A PSRU doesn't change the power of the engine at all. It simply allows you to use a larger, more efficient propeller to turn whatever power you've got into thrust. Adding a turbo may increase the power output of the engine, but if that power is shoved into a sub-optimal propeller, you've gained nothing.
I think what he means was to use a turbo to boost the HP of the engine at a lower rpm to the same levels that it normally produces at its design rpm, so the PSRU becomes unnecessary. However, the drawback is that the engine is now putting out much more torque than it's designed to produce, causing a lot more stress on the internal components, while the rpm is down so cooling may suffer. For a reliability standpoint, the PSRU is probably the better approach... rotaries are suited to high rpm operation, and everything has been optimized for that.

-Dana

"A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away." -Antoine de Saint-Exupéry
 

Starman

Well-Known Member
Joined
May 16, 2009
Messages
2,011
Location
High in the Andes Mountains
It's a valid approach to increasing the useful power of a direct drive, but I don't think you can make a straight comparison. Twice the revs or twice the amount of air/fuel going through it on paper may make the same power, but to force twice the normal amount of air into an engine takes considerable power, more than a turbo alone can provide, I would think. Remember also that a turbo will only work as hard as the engine - at low RPMs the available exhaust energy is not that great. And if you resort to supercharging, well, that also takes power away from the engine.
Sorry about the delay, I've been out of town.

Your concerns are reasonable but relatively unfounded, I think. First off, running at my target RPM of around 4000 is not low RPMs, it's at the top of the torque curve and running at maximum volumetric efficiency for a Mazda rotary, therefore the boost will be added right at the best possible "sweet spot" on that engine to get the most for the least so to speak.

The turbo creates some back pressure but the power to run the compressor is essentially free if you use a turbo. Also, how much boost you get out of a given RPM all depends on turbo sizing, you have to size the turbo for the specific operating range you intend to use and it will work just fine. There are some very small turbos made to run on very small engines, (like lawn tractors?) Theoretically you could use one of those tiny ones to turbocharge and boost a big V8 just coming off an idle, I say theoretical because even though you could boost it obviously it wouldn't be practical or useful.

I am playing devils advocate here - I intend to run a direct drive engine myself and am seriously looking at turbocharging too, having just been across the Rockies twice last week, it makes almost no sense not to have a turbo on any aircraft engine.
Devils advocate I like, bring it on. You will experience diffeent issues supercharging a piston engine than I'm looking at using a rotary. Keep in mind the stock Mazda turbo and it's oiling system is not designed for continuous use and will not last long.

A wankel is, as "they" say, half way between a piston engine and a turbine. You could link it with a very high capacity turbo and boost it way beyond what a normal piston engine could take and make huge amounts of power ... briefly, to keep the engine from melting.
 

Starman

Well-Known Member
Joined
May 16, 2009
Messages
2,011
Location
High in the Andes Mountains
I think what he means was to use a turbo to boost the HP of the engine at a lower rpm to the same levels that it normally produces at its design rpm, so the PSRU becomes unnecessary. However, the drawback is that the engine is now putting out much more torque than it's designed to produce, causing a lot more stress on the internal components, while the rpm is down so cooling may suffer. For a reliability standpoint, the PSRU is probably the better approach... rotaries are suited to high rpm operation, and everything has been optimized for that.
Like I mentioned before, 4000 RPM is plenty high and the cooling correction (if needed) is running the water pump a little faster than stock.

Concerning the reliability standpoint, a rotary engine can not 'break' and suffer a catastrophic failure like a piston engine can because there are no parts to break (except for the tip seals). The weak link is the tip seals, and I think high RPMs will cause more wear on them than high torque will and the seconds parts I get (after the aftermarket dual ignition for it), will be racing tip seals.
 

Starman

Well-Known Member
Joined
May 16, 2009
Messages
2,011
Location
High in the Andes Mountains
Errr... no. A PSRU doesn't change the power of the engine at all. It simply allows you to use a larger, more efficient propeller to turn whatever power you've got into thrust. Adding a turbo may increase the power output of the engine, but if that power is shoved into a sub-optimal propeller, you've gained nothing.

Ultimately, it's how much thrust the propeller produces that matters. Turning a relatively larger prop more slowly is the way to go, if you can manage it. Once the tips go sonic, it doesn't matter how much power you pump into your propeller. The efficiency drops so low turning that power into thrust that you end up with a net loss compared to a better setup. It's surprisingly easy to get propeller tips up into the transonic speed range. Many airplanes do, which is one reason you can often hear more propeller noise than engine noise.

A turbo isn't a direct substitute for a PSRU, and you have to look at the entire engine-propeller combination to determine the amount of thrust produced. In the end, the thrust coming off the prop is the only thing that matters.
I agree, and I know I'm suffering an efficiency loss with a smaller diameter propeller. I'm too lazy to have done the math on it yet because my mind she is already made up and it doesn't really matter. It doesn't really matter because I will be using the great equalizeer called boost.

I say it will suffer a loss of efficiency rather than a loss of thrust because I can also make up for this loss of thrust with more turbo boost. In other words, if a smaller diameter prop will loose me say 25% in thrust then I just crank up the boost by 25% and I get my thrust back.

The main penalty is worse fuel consumption and more cooling required.

So, I can make up for both the lack of torque multiplication of a psru AND the loss of prop efficiency due to small diameter with more boost. Also, the Mazda rotary is no slug at 4000 RPM even without boost.

Another thing is that the efficiency loss due to small diameter is greater at low air speeds but becomes quite small at high speeds.

Next!
 

Starman

Well-Known Member
Joined
May 16, 2009
Messages
2,011
Location
High in the Andes Mountains
Go to the engine threads for rotaries and look at what is used. Some of them have tried spring dampers from clutches and others use rubber isolators, all in the name of tuning the resonant frequency away from firing frequencies.

I don't remember anyone making a Wankel rotary engine with direct drive - they all use some sort of reduction drive.
I don't think it has been done before because the world was waiting for ... Starman ... TAADAA :)

I'm going to try to avoid a PSRU for several reasons. 1, is because I don't like complicated machinery, I want things to be as simple as a hammer. 2, they appear to be expensive, and cost almost as much as I want to spend on my whole plane initially. 3, according to some experienced people here none of the PSRUs are either good designs or are proven for the rotary, in other words, they are all potential failures. I have noticed that there are people flying behind them now though, but there are more negatives for me. 4, I'm making a pusher so want a smaller diameter prop (60" - 4000RPM), and the reduction on available PSRUs is too much, it will make the prop RPM too slow for a small diameter prop. I could go as big as a 72" prop. 5, I don't like the idea of constantly running the engine at redline. 6, PSRUs weigh a lot more than nothing :) . There may be some other reasons but I forgot them.

So this is the initial plan. At 4000RPM a non turbo wankel will put out 100hp, which is acceptable but anemic. I'm getting a turbo, so it should do better in stock condition, but later I'm going to play with the turbo size and boost the fork out of it so I can basically double and possibly even triple the torque at 4000RPM. That way I get as much horsepower as someone with a reduction gear but without a reduction gear.

I realize that the smaller diameter prop looses some efficiency, but not too much at higher speeds so that doesn't bother me. I will consider going with an even smaller diameter prop and reving the engine even more when the time comes to order a prop. I'll loose more efficiency on the prop but get more power from the engine so I'll get more thrust as a result. The penalty for that is worse fuel economy.
 

Lucrum

Well-Known Member
Log Member
Joined
Jun 10, 2008
Messages
956
Location
Canton, GA
So this is the initial plan. At 4000RPM a non turbo wankel will put out 100hp,...
According to a spreadsheet I found on the web, you can expect 121HP from a normally aspirated 2 rotor at 4K RPM.
At the same RPM and 10psi boost(approximately 50 in of manifold pressure I think) expect about 164 HP.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,289
Location
Saline Michigan
I will debunk a couple of things and talk Mazdas here.

I knew Everitt Hatch, and loved what he did to make the Powersport engine and reduction drive work. We talked a bunch of times about how to make the system robust. His friend Alan Tolle has done more dead stick landings than about anyone around while he and Ev were figuring out how to make a PSRU live on a Mazda rotary. The Powersport is a stiff system.

There are several designs of PSRU's on Mazda rotary engines flying on airplanes. How close to the edge they are, I do not know, but since they have quite a few engines flying now without failures, perhaps they are OK. All of the outfits I know about are mentioned on this website: engines There might be others, but Powersport, Atkins, and Crook have quite a few engines flying with their PSRU's.

Next, if the engine is a 200 hp engine making 200 hp at 7000 rpm, or a 350 hp engine making 200 hp at 4000 rpm, it is still making the same 200 hp, and will probably need about the same amount of cooling, glycol flow rates, oil flow rates, etc, because the rejected heat is primarily a function of the power. Yeah, you may need to change the pulley ratio for the water pump and install a bigger oil pump (roughly one-third of the cooling is oil based), but the cooling load, cooler sizes, and flow rates will all need to be about the same if the power is about the same.

Next, O-200's tuned for Fomula One air racing all run about 4000 rpm. So there are even people making efficient props to run around those speeds... Yeah, a 4000 rpm O-200 is special, but they know how to build good props for that rpm.

There are Mazda 13B engines running great big Roots style superchargers that make monster torque and horsepower, and can do incredible wheel stands during launch, and then after a day at the strip, these guys drive them home on the street. Other folks build big horsepower road race engines that run all season and get freshened up over the winter not because the power is down, but just because it is a good idea... So, with proper components I suspect that Starman's goal of a 200 hp 4000 rpm engine is probably within reach with the right pieces. I do not know which pieces need to be race parts and which are stock, but there are folks out there who may have an opinion...

You can size an exhaust turbine and compressor to do about anything you want in terms of boost if you are willing to give it a little time to spool up. Starman's budget may get blown by having to use a special turbo assembly, or maybe there is an appropriate one already in the boneyards...

The biggest question is: Will the choices made and the pieces selected hold together? They can fail. Remember, you still have to pump and regulate fuel, trip the spark at the right time, pump glycol and oil, have the rotor seals, gears and bearings all survive, etc. And then you have a shaft, prop bearing, custom prop, all parts that are unique with nobody else running one, etc... Plenty of opportunities to find your self at reduced power, or even without power entirely.

So, let's not become cocky, or Murphy may have to remind us all of who is really in charge...

Billski
 

Starman

Well-Known Member
Joined
May 16, 2009
Messages
2,011
Location
High in the Andes Mountains
According to a spreadsheet I found on the web, you can expect 121HP from a normally aspirated 2 rotor at 4K RPM.
At the same RPM and 10psi boost(approximately 50 in of manifold pressure I think) expect about 164 HP.
That's definitely no slouch even at 4000rpm and it will be plenty of power for starters. After seeing that I was just dyin' to know how much power it put out at higher revs so I looked up some dyno charts.

I Goggled for turbo kits and found one that puts out 225hp at 4000rpm, and the power doesn't even go up much higher at higher revs if I'm reading it right.

Dyno chart: CorkSport Mazda Performance » 2010 MazdaSpeed3 First Dyno Results

Here's one that appears to be a stock turbo wankel and the power jumps up amazingly after 4000rpm, it goes up to around 250hp at 5000rpm.

chart: 1994 Mazda RX-7 RX7 stock seq turbos Dyno Results Graphs Hosepower - DragTimes.com
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,289
Location
Saline Michigan
Careful. In performance cars, they know that the road user can only be on the turbo for short chunks of time and for a pretty small fraction of the vehicle life. Since they are more concerned with turbo lag than we are, compromises in turbo size that may adversely affect turbine life and efficiency are frequently made that you may not want in your airplane. There are books on the topic of turbocharger selection and installation that you should probably get into.

For an airplane, I suspect that you want to size your boost hardware for continuous duty at your range of power settings, and within the controller's range of control. If you find boneyard engines with turbos that fit that fine, but if they don't you may not be happy with the results...

Billski
 

Starman

Well-Known Member
Joined
May 16, 2009
Messages
2,011
Location
High in the Andes Mountains
I agree with all that, but forgot I read about the stock turbo oiling system being inadequate for extended high boost. At first I want to just use a completely stock junkyard engine with the stock turbo. I don't really care about getting a lot of power in the beginning, and if and when the turbo seizes the engine will still make non turbo power, which is sufficient.

That's the original low cost plan to get in the air. Later, when I want to play with turbo sizing I'll probably buy an aftermarket item and won't be going to a wrecking yard and I realize it will start getting spendy at that point. However, once I determine the size turbo I need there just may be one the right size on some other car, and I will look into that.

As far as using high power with very high boost is concerned, I envision using these very high power levels for only short duration, like sixty seconds or so, for quick climbs. Therefore the stock oil and water cooling systems won't be overwhelmed, particulalry because the rotary is so insensitive to slight overheating. If I want to use high power for extended duration then I will be needing to boost the cooling systems also.
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,077
As far as using high power with very high boost is concerned, I envision using these very high power levels for only short duration, like sixty seconds or so, for quick climbs. Therefore the stock oil and water cooling systems won't be overwhelmed, particulalry because the rotary is so insensitive to slight overheating. If I want to use high power for extended duration then I will be needing to boost the cooling systems also.
The biggest hassle with any auto conversion, besides PSRU or other drive issues, is cooling. The stock systems are designed to cope with around 30% power, with very brief pulls at full throttle or slightly longer at maybe 60-70%. And maintaining adequate airflow to a rad without causing a whole lot of drag is an art in itself. I managed to do it with a Subaru conversion using the stock car radiator, but I had to come up with a fancy plenum to get enough pressure differential. Even then, it got up to 240°F or more on a long climbout, and that wan't anywhere near sea level so it wasn't producing full power. It was clear that simply hanging a rad in the breeze would never keep it cool.

Everything I ever read about Mazda conversions indicated cooling hassles. That engine burns a lot of fuel in a small package so there's a lot of waste heat to dump. In most engines, 50% of the heat goes out the exhaust and another 25% or so via cooling systems (oil and coolant). The remaining 25% is turned into useful power. Tha Mazda, IIRC, has a habit of warping its side plates when they get hot, and the thing loses power big time.

Dan
 

Starman

Well-Known Member
Joined
May 16, 2009
Messages
2,011
Location
High in the Andes Mountains
I agree, that all makes sense. I'm hoping to copy the P-51 system, which is good for 1500hp, and reducing it. Also, the P-51 system was optimized so that the rejected heat from the radiator actually produces some thrust, or so I read somewhere. It sounds like I will need to speed up the water and oil pumps after I put on the dual ignition.

So, let's not become cocky, or Murphy may have to remind us all of who is really in charge...
Don't worry, if anyone can find a way to screw up it's me, and I see the drive system as the main big challenge in my project.
 
Last edited:

MKIV

Active Member
Joined
Aug 19, 2009
Messages
30
Location
New Zealand
It sounds like I will need to speed up the water and oil pumps after I put on the dual ignition.
Don't worry, if anyone can find a way to screw up it's me, and I see the drive system as the main big challenge in my project.
Many factory automotive water pumps are not efficient & become even less when sped up. I think its "howard stewart or maybe just stewart' who make V8 Nascar pumps that we actually can reduce the pump RPM by a large amount, yet still move sufficient coolant to keep things together @ 900HP & 9000 RPM. Looking at crank pulley dia of around 4" & pump of approx 6". Mind you in the AC life is a bit better- clean cool air & lower temps at altitude- but you gotta come down sometime & sitting in line waiting to depart after an airshow will see temps rise.
Some oil pump designs can have the same issues.
 

Starman

Well-Known Member
Joined
May 16, 2009
Messages
2,011
Location
High in the Andes Mountains
Thanks Jac, I think speeding up the stock water pump will be sufficient for the stock turbo engine because of the lower engine rpms at direct drive. For example, lets assume that the stock pump is good for up to 8000 engine rpm and faster than that it starts to cavitate. OK, if I'm running at 4000rpm then I can double the water pump speed and still it will not actually be exceeding it's design limits.

I'm sure I will be needing to look into more aftermarket goodies as I start to stuff more fuel and air into it.
 

EzyBuildWing

Well-Known Member
Joined
Sep 23, 2009
Messages
48
Location
Sydney NSW Australia
Duncan Aviation(Comanche, Okla.) many years ago listed in their brochure a direct-drive single rotor Mazda RX-7 bridge-port aircraft engine. They machined the crank to accept a new main-bearing located on the front of the engine. The main-bearing was housed in an easily-machined circular housing. Propeller-boss was attached to the front of the re-machined crank.

Engine was rev-limited to 60hp at 4000 rpm. Engine could be fitted with a fixed-pitch prop, or cockpit-controlled variable-pitch prop( Warpdriveprops makes a carbon-fibre one, and their website has a good tip-speed calculator). Duncan photo of this single-rotor engine shows it fitted with a 2-blade in-flight adjustable prop.

WEIGHTS:

Basic engine with oil-pump, fuel-pump, water-pump, Posa 32 mm Super-carby, ignition, engine mounts and prop-flange was 110 lbs.

To this must be added: alternator (4lbs), starter(5.5lbs), 9" x 14"engine coolant radiator(5 lbs), and 7" x 8" oil cooler(1.8lbs)

This gives a DRY INSTALLED WEIGHT of 126.3 lbs.

Engine-coolant( 10.4 lbs) and oil (6lbs) must be added...which brings wet-weight up to about 143 lbs..... not bad for 60hp continuous with 3.5 gph fuel-burn(regular or unleaded) and a 3000 hr TBO.

How much static-thrust could be obtained from such an engine reving at 4000rpm?

A McCulloch 72 HP ( 68 actual) two-stroke in a gyrocopter swinging a 48" or 50" diameter prop delivers about 260 lbs of thrust at around 4000 rpm.

Igor Bensen said that for "respectable performance" a gyro requires 1 lb of static-thrust for every 2 lbs of gross weight, so a 520 lb gross gyro looks feasible. A small 4-stroke chain-saw engine(with a hand-pull recoil-starter) could replace the heavy starter-motor, and could also pre-spin the rotors. A smaller and lighter battery and alternator could thus be used. Maybe EFI and a turbo would give more power (thrust), for little nett additional weight.

Duncan were also offering an "Aerodyne Model 2.66 VATN turbo(Variable Area Turbine Nozzle)" , which weighed 15 lbs and which would increase power and quieten exhaust noise.

Duncan also offered a single-rotor fited with a Uniroyal HTD cogbelt rated at 300hp. Reduction ratio was 3:1 and output was 120hp at 9,000rpm for 5gph, and a DRY INSTALLED WEIGHT of 165 lbs. This power output agrees with the output of a Racing Beat single-rotor bridge-ported engine which NASA tested in 1982 which produced 128hp at 9000 rpm.

Does anyone else think a single-rotor direct-drive looks attractive as a gyro-engine?

I can post a 3-view drawing of the engine if anyone's interested.
 
2
Group Builder
Top