Mazda Rotary Engine

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Armilite

Well-Known Member
Joined
Sep 5, 2011
Messages
3,517
Location
AMES, IA USA
Actually, I am building a Long-EZ with rotary power..
=================

What are you using for a Redrive? Which Motor 13B or 20B? What's your Target hp your after? Years ago, Finding a Redrive to Handle the hp made at high rpms was the biggest problem. Also Dealing with the Exhaust Heat.

Today, you have better Synthetic Oils, Engine Coatings, and Billet Gear Drives to use.
 

Attachments

dwalker

Well-Known Member
Joined
Mar 6, 2021
Messages
145
=================

What are you using for a Redrive? Which Motor 13B or 20B? What's your Target hp your after? Years ago, Finding a Redrive to Handle the hp made at high rpms was the biggest problem. Also Dealing with the Exhaust Heat.
I'm using a 13B, most likely semi-P ported with RX8 rotors and a balanced rotating assembly.

Today, you have better Synthetic Oils, Engine Coatings, and Billet Gear Drives to use.
I'm looking at the Marcotte redrive as it sends the most robust currently available. I've spoken to the fellow in New Zealand about his Redrive, which is a version of Tracy Crooks, and I like it enough I may buy some parts from him and machine the rest myself.

I'm going to use a turbocharged 13B Semi-P ported motor with a balanced Renesis rotating assembly.
I will use AEM Infinity engine management. The goal is to make 200hp at about 4500 to 5000 Rpm.
Not concerned about exhaust heat and will be using a similar cooling arrangement as the Pro Formula Mazda cars.
 
Last edited:

Armilite

Well-Known Member
Joined
Sep 5, 2011
Messages
3,517
Location
AMES, IA USA
I'm looking at the Marquette redrive as it sends the most robust currently available. I've spoken to the fellow in New Zealand about his Redrive, which is a version of Tracy Crooks, and I like it enough I may buy some parts from him and machine the rest myself.

I'm going to use a turbocharged 13B Semi-P ported motor with a balanced Renesis rotating assembly.
I will use AEM Infinity engine management. The goal is to make 200hp at about 4500 to 5000 Rpm.
Not concerned about exhaust heat and will be using a similar cooling arrangement as the Pro Formula Mazda cars.
==================

The output of late 13B-T was 185 hp (138 kW) at 6500 rpm! To get 200hp at 4500-5000rpm your probably looking at:

1300cc at 4500rpm 4 lbs of Boost = 175hp
1300cc at 4500rpm 5 lbs of Boost = 185hp
1300cc at 4500rpm 6 lbs of Boost = 194hp
1300cc at 4500rpm 7 lbs of Boost = 203hp

1300cc at 5000rpm 4 lbs of Boost = 195hp
1300cc at 5000rpm 5 lbs of Boost = 205hp
1300cc at 5000rpm 6 lbs of Boost = 215hp
1300cc at 5000rpm 7 lbs of Boost = 226hp

Good Luck on your Build!
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,183
Location
US
==================

The output of late 13B-T was 185 hp (138 kW) at 6500 rpm!
With even mild peripheral porting and boost, turbo 13Bs in aircraft have no trouble exceeding 200hp at 5500 rpm, and they'll happily turn faster than that and make more power.
 

dwalker

Well-Known Member
Joined
Mar 6, 2021
Messages
145
==================

The output of late 13B-T was 185 hp (138 kW) at 6500 rpm! To get 200hp at 4500-5000rpm your probably looking at:

1300cc at 4500rpm 4 lbs of Boost = 175hp
1300cc at 4500rpm 5 lbs of Boost = 185hp
1300cc at 4500rpm 6 lbs of Boost = 194hp
1300cc at 4500rpm 7 lbs of Boost = 203hp

1300cc at 5000rpm 4 lbs of Boost = 195hp
1300cc at 5000rpm 5 lbs of Boost = 205hp
1300cc at 5000rpm 6 lbs of Boost = 215hp
1300cc at 5000rpm 7 lbs of Boost = 226hp

Good Luck on your Build!
My current street car motor makes 390whp (not at the flywheel, but at the wheels) at 17psi by 4000rpm. Boost is largely an irrational way of talking about power output, rather we should always talk in lbs/min when discussing power output potential. You can run 30psi on a small turbo that maxes out at 30 lb/min, or 10psi with a turbo that puts out 70lbmin. Obviously this is grossly oversimplified, but as an example-

The 13B FD made 220-ish hp (probably flywheel) at roughly 9psi. However those turbos were very small and working very hard to make even 9lbs, and making about 30 lb/min at best. Change the twins for single in the 50 or so lb/min range and while boost pressure stays pretty much the same, power will go up. The average 13B REW will gain 100whp at the same boost level going from twins to a single.

I was going to write out a very long explanation of the how and why and the development steps that have lead to the current engine outline, but it's beyond the scope of this discussion. The motor for my plane will have all the power it needs between the working rpm of 4000-6000, based on a 2.4 to 2.8 reduction box.
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
1,540
Location
Jackson
I don't know your airframe or its prop diameter limits, but with typical homebuilt airframes' diameter limits, you might want to take a long hard look at ratios if you're only going to turn the engine at 4000 rpm. There are diameters where the prop can turn too *slowly* for efficient operation.

Charlie
 

dwalker

Well-Known Member
Joined
Mar 6, 2021
Messages
145
I don't know your airframe or its prop diameter limits, but with typical homebuilt airframes' diameter limits, you might want to take a long hard look at ratios if you're only going to turn the engine at 4000 rpm. There are diameters where the prop can turn too *slowly* for efficient operation.

Charlie
To be honest, l'll be more concerned with engine operating rpm and psru ratios as I nail down what is and is not available and will work. Currently the only for sure available unit is the 2.4ratio, which if I use the same prop rpm limit as I would with have with an 0-320, puts the engine at 6200rpm.
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
1,540
Location
Jackson
I have a 2.85 for my Renesis, but would have chosen 2.4 if it had been an option. As you say, perfect for 6k rpm & Lyc prop speeds. As you probably know, 6k rpm
is supposed to load a rotary's bearings & gears the least. The NA Renesis is good for about 180 hp at 6k rpm, which is quite adequate for my RV7.

Your mention of 4k rpm is what got my attention. ;-)

Stay in touch; we'll be hosting anothe rotary/ alt engine gathering once we're truly ' post-Covid'.
 

dwalker

Well-Known Member
Joined
Mar 6, 2021
Messages
145
I have a 2.85 for my Renesis, but would have chosen 2.4 if it had been an option. As you say, perfect for 6k rpm & Lyc prop speeds. As you probably know, 6k rpm
is supposed to load a rotary's bearings & gears the least. The NA Renesis is good for about 180 hp at 6k rpm, which is quite adequate for my RV7.

Your mention of 4k rpm is what got my attention. ;-)

Stay in touch; we'll be hosting anothe rotary/ alt engine gathering once we're truly ' post-Covid'.
I ran the Renesis in the RX8 in the Pirelli World Challenge, along with the Pikes Peak Hillclimb (pictured, driver withdrew before the race) etc. and know way more about those motors than I ever intended to.
 

Attachments

EzyBuildWing

Well-Known Member
Joined
Sep 23, 2009
Messages
246
Location
Sydney NSW Australia
Bill Joengbloed Aviation, Box 40, Chino Airport, Ca 91710 converted a 13B (turbo'd) for aviation....here's what he said, many years ago:

First a bit of an introduction as the name Jongbloed is actually pronounced "Youngblood". I'm Bill Jongbloed and my background is in automobile racing and spans 23 years. The last 12 years have been devoted to the manufacturing of modular racing car wheels, however, prior to this I was an active racing car mechanic and fabricator. Since aviation is where I've wanted to be since childhood, entry into the market seemed a natural. Also, having witnessed first hand the development of both Chevy V-8 and Mazda rotary engines made me feel that they were both quite suitable for use in experimental aircraft.

Once the decision was made to develop a reduction drive unit, a decision had to be made between cog belts, Hi Vo chains or spur gears. Belts seemed to be the simplest and I had lots of experience with them as accessory drives on race cars.

We found, however, that to harness 200 BMP required a 6.0 inch wide belt. That wasn't too bad, however, my intent was to eventually harness up to 750 BMP and this would require about 22.0 inches of belt width! Chains were next up and weren't as bulky as belts but it was becoming apparent that gears would still be the best all round solution. So gears it would be and thought was initially given to using gears from a Muncie 4- speed automotive transmission.

That became a fleeting thought as it became apparent that, short of bearings and seals, the gear reduction unit would be best if all component parts were designed and built from scratch. My father is a retired aerospace mechanical engineer, so I joined forces with him and asked a friend of mine from racing named Pete Weissman to look over our shoulders. He agreed and the design was begun. Pete, by the way, designs and manufactures probably the best and most exotic racing car transmissions in the world, from TransAm racing right up to Formula One.

Our layout was to be a very simple, strong and compact design using straight cut spur gears with a quillshaft to isolate the harmonics of the crankshaft (eccentric in Mazda) from the gears and prop shaft. The quillshaft is very critical as, for instance, any change in the crankshaft will also require a change in the quillshaft. This simple approach also enabled us to complete the program in a reasonable length of time (our spare time).

Our gearbox layout raises the propshaft up 5.0 inches vertically and offsets it to the left 1.0 inch giving extra room around the exhaust system. This 5.0 inch vertical offset also allows all of the Mazda accessories, such as water pump, carburetion, alternator, oil filter and foil fill neck, to remain in the stock location. This goes a long way towards minimizing the homebuilders already large task Jongbloed Reducer at hand. Joining the reduction unit to the Mazda engine is a bellhousing of our design. Incorporated in the bellhousing are Lycoming type compression mounts that act in concert with another pair at the rear of the engine (front in the car) to complete the mounting of the powerplant and gearbox assembly in the airframe. A stock Mazda starter motor is used along with the automatic transmission flywheel.

Now, let's look at the powerplant itself:

The Mazda 13B engine was chosen because of its turbine smoothness and its compact design. It is also an expandable design and, in fact, a three rotor version is currently being raced by the Mazda factory. My guess is that within 2 years these engines will see introduction in the passenger car market and aircraft are only a natural at that point. This version is called the 13G, more on this later. The Mazda engine is very simple mechanically, however, it can't be run backwards so one important note is that you will need a counterclockwise propeller. Fred Griffith of the Great American Propeller Company assures us that this causes no hardship other than simply ordering the right prop. Our conversion of the Mazda utilizes turbocharging. The reasoning is twofold as we not only wanted to offer an engine with superior performance at altitude, but also to muffle a non-turbocharged Mazda requires an act of God plus about ten feet of tubing and mufflers.

The turbocharger does a commendable job while boosting performance. This noise business is really important and seems to be often neglected. Since noise is a byproduct of fuel, let's discuss fuel consumption. A properly tuned Mazda will have fuel specifics of .55 to .60 Ib./hp/hr. and since it is liquid cooled, won't need to run rich for climb.

That sounds high, however, in his forum last year at Oshkosh, the AVCO-Lycoming man stated that while an air cooled motor may cruise at .45 Ib./hp/hr. if you checked consumption on the entire flight or "mission", as he put it, the total fuel burn would average .55 to .60 Ib./hp/hr.

Doesn't that sound a lot like the supposedly thirsty Mazda? The next subject is how fast will it turn? Mazda's are quite happy to run for long spans of time at 5000 to 6500 rpm but they get tired really quickly if they are spun faster than that; I don't care what the guy in the bar tells you, he probably has the wrong tachometer!

A Mazda set up such as ours will produce 170 BH at 5000 rpm and 200 BMP at 6500 rpm. Remember that there is no magic, so cast a wary eye towards those claims of 300 BMP and 250 pounds. To reach those HP levels you must spin 9000 to 9500 rpm and that only happens on racing tracks in racing cars. My engine consultant is Clayton Cunningham of CCR racing whose motors are about the best in the business and he says 6500 rpm.

As far as 250 pounds goes, it's true but that's a race engine without an aircraft gear reduction. What dwells under our turbocharger is a basically stock 13B motor with a mild port job, plus one mandatory modification. There are some planetary gears inside the motor which must be exchanged for their hardened counterpart. These gears are available from a variety of sources but they are a must.

Carburetion is at this time experimental, however, an automotive Holley 2 barrel seems just fine unless aerobatics are your desire with negative GS.

Now, let's return to the gear reduction unit as we've just talked about rpm. Our unit is geared at 2.29 to 1.00 which means that you will have a propeller turning 2875 rpm for takeoff (6500 engine rpm). Cruising at 5000 engine rpm will give you a propeller turning 2183 rpm.

If you ever wondered why gear ratios seem odd, such as 2.29 or 4.56, the reason is this. The term "prime numbers" means that each tooth of, say, this pinion gear will, with each revolution of the gear, move over one tooth so that each gear tooth will contact every tooth on the opposing gear.

This greatly reduces the wear of the gears and is why gear ratios are always odd, such as our 2.29 to 1.0 ratio. Speaking of gears, ours are machined from 8620 steel and carburized. Our shafts are machined from 4130 and 4340 steel then heat treated and ground to finish size. Oiling is by splash, however, provision is made for pressure oiling should that be desired. Hydraulics can also be provided for constant speed propellers although great hope is placed on the new electric constant speed propellers being developed by the Great American Propeller Co. of Oceano, CA.

A prototype can be seen on our test engine and considerable testing will be carried out in cooperation with Great American. Cooling, radiators and plumbing have for sure turned many homebuilders off to automotive engines, and for good reason. Designing a proper cooling system is a major task and is another neglected area. Towards this end, we embarked on a universal cooling pod assembly that looks like the belly scoop from a P-51 Mustang. This was no mistake as the Mustang cooling system actually developed thrust from its hot exhaust air. This is called thrust recovery and in the case of the P-51 was reputedly worth 150 BHP!

So much for having to design your own, and remember our pod assembly can be shifted fore or aft to achieve the best weight balance prior to final installation. Our pod is also narrow enough to fit under the fuselage of a Long-EZ.

Continuing on the subject of weight, let's now look at powerplant weights. According to Jane's, a Lycoming 0-360 weighs 268 pounds and produces 180 BHP. That's fine but put that engine on a scale with oil, exhaust and acessories and it will read 325 pounds.

Besides, once you are up to 10,000 feet, your available power is 125 BHP, giving a powerplant power-to-weight ratio of 2.6 lb. per hp. Now, by installing our Mazda conversion, your engine-gear reduction weight, including oil, water, turbo system and accessories, will be 330 pounds forward of the firewall and 50 pounds aft in the cooling pod.

You will have turbine smoothness, better aerodynamics up front where it counts and 200 BMP at 10,000 feet. Total package weight is 380 pounds, giving a power to weight ratio of 1.9 lb. per hp. Remember, too, that the cooling pod will be virtually drag free due to thrust recovery. Mounting of any powerplant requires a proper motor mount and cowling.
 

dwalker

Well-Known Member
Joined
Mar 6, 2021
Messages
145
I like Joengbloeds wheels, used many on formula cars and smallbore racers.
Given that this was likely published in the late 80s or maybe earlier, it's an excellent treatise on the rotary in Aviation.
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
1,540
Location
Jackson
It's worth pointing out that no one I'm aware of over the past 25 years of following rotary conversions is making any of the internal mechanical changes he details. The only internal failure that comes to mind happened when someone used part of the car's intake manifold and left some of the valving in place (unneeded in a/c). Vibration later loosened one of the valve's positioning pins, which entered a combustion chamber & trashed the apex seals.

The other detail I'd disagree with is his projected difference in takeoff vs cruise rpm. With a constant speed prop and a 'boosted' engine, you can operate as he described. But the practical reality is that very few alt engine guys invest in constant speed props, so takeoff rpm and cruise rpm are typically going to be very close to the same number (unless you take off and then pull power *way* back to 'loiter'). Big spreads in prop RPM with a fixed pitch prop will likely result in poor efficiency at one end or the other.

Oh, he's a bit high on installed weight of a rotary, too. My Renesis FWF, complete with dual full size alternators, relatively heavy muffler, redrive, coolers, etc weighs 335 lbs dry. I'm pretty confident I could swap one alternator for a turbo and still be well under 350.
 

dwalker

Well-Known Member
Joined
Mar 6, 2021
Messages
145
It's worth pointing out that no one I'm aware of over the past 25 years of following rotary conversions is making any of the internal mechanical changes he details. The only internal failure that comes to mind happened when someone used part of the car's intake manifold and left some of the valving in place (unneeded in a/c). Vibration later loosened one of the valve's positioning pins, which entered a combustion chamber & trashed the apex seals.

The other detail I'd disagree with is his projected difference in takeoff vs cruise rpm. With a constant speed prop and a 'boosted' engine, you can operate as he described. But the practical reality is that very few alt engine guys invest in constant speed props, so takeoff rpm and cruise rpm are typically going to be very close to the same number (unless you take off and then pull power *way* back to 'loiter'). Big spreads in prop RPM with a fixed pitch prop will likely result in poor efficiency at one end or the other.

Oh, he's a bit high on installed weight of a rotary, too. My Renesis FWF, complete with dual full size alternators, relatively heavy muffler, redrive, coolers, etc weighs 335 lbs dry. I'm pretty confident I could swap one alternator for a turbo and still be well under 350.
Just have to remember he is a few decades behind on technology
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
1,540
Location
Jackson
I wonder if he's flying what he described. I don't remember his name on either of the rotary email lists, but a lot of alt engine guys never participate in internet groups; they just get stuff built and fly it.
 

Lendo

Well-Known Member
Joined
Feb 6, 2013
Messages
711
Location
Brisbane
rv7charlie, I would be interested to know what size prop would you run with that 2.4 ratio PSRU? Certainly keeping the rpm up at 2,500 prop rpm.
George
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
1,540
Location
Jackson
My logic: Lyc design rpm is 2700. Lyc powered RVs expect prop diameters in the 68"-72" range for 2700 rpm full output, or 75% power at ~8K' altitude. 2700 * 2.4 = 6480, which is a nice conservative rpm for a ~180 HP aviation rotary at sea level, or 75% of 180 HP at altitude.

The 2.85 at the same rpm begs for a significantly larger diameter prop, but even the later, longer legged RVs are limited to a max diameter of 76". That's the dia I purchased.

Charlie
 

AIRCAB

Well-Known Member
Joined
Feb 6, 2015
Messages
117
Location
Vancouver Island
Kyle is selling a service that does not provide nitriding, which is in fact a very necessary thing to have in a rotary engine for it to live a long life, so I am not surprised to hear him say that. Chip sends his side housings to Tennessee (not to far from where I sit, actually) to have them nitrided. Pick a longtime rotary engine builder- Rob at Pineapple, Dan at Atkins, Chip at Chip's Motorsports, Darrel Drummond, etc. and they will all wantplated to be nitrided after having been ground. And they are right.

Mazda has experimented themselves with Nikasil rotor housings, which some of us have duplicated with a US company, and this shows promise as far as wear and reducing friction/improving sealing. Goopy Performance in New York now able to rechrome the rotor housings to like new condition. IF I were still experimenting with the rotary I would have a set of side housings ground .010 then Nikasil lined and built back up .015 and lapped/resurfaced back to true, the rotor housings stripped of the chrome lining down to the steel insert then plasma sprayed, nikasil lined and lapped surfaced. This would result in a very durable and wear resistant surface that would also seal exceptionally well.
The reason these processes have not gotten the attention or development in any of the rotary applications, other than drone engines, is the availability of new parts at reasonable cost from Mazda and options in the aftermarket. If that changes, we may see much more technology applied in the Wankel rotary.
You seem to have missed what he is doing after lapping !
 

harrisonaero

Well-Known Member
Joined
Oct 31, 2009
Messages
580
Location
Coeur d'Alene, ID
A Mazda rotary on a V6 STOL (Bushmaster, Javelin, etc.) is about the biggest bang for the buck in a good performing aircraft that can be built cheaply, cruises reasonably fast, can haul a lot, has a "safety cage" airframe vs aluminum or composite, and can get into most any strip that's out there for fun camping and adventure.

I owned a flying V6 STOL sans engine and was putting a rotary on it (Bruce Turrentine built, using Tracy Crook's products, etc) when my job changed and I sold the airframe. Only thing left from that project is a professionally built engine mount made by Snowline that made mounts and other 4130 weldments for Aviat Aircraft and Columbia.
 

Lendo

Well-Known Member
Joined
Feb 6, 2013
Messages
711
Location
Brisbane
rv7charlie, I like what dwalker said about technology, the Mazda Rotary is best with P-port and lightened housings, it's a pity Mistral had to give up their dream of Certification, personally I would have liked them to start smaller with Experimental, they may have survived longer. I run your numbers through the JC Prop Program for 72", I will look again at 68" and offer the results, just for general builder information. You probably have all the data you need.

Mr. Youngblood's approach was for a standard engine and it was a pleasant surprise to see accurate information on weights, hence my comments of lighted housings.

I have been researching a 2.73 ration PSRU for a friend, 6000 for cruise and 6580 for max power, 180 hp cruise to 200 hp at sea level and losing 3hp per 1000 ft for altitude at 100% VE, considering increases for PP and losses for PSRU etc. However no numbers are perfect until their tested, but I believe this could be conservative. I have done this for the 72" prop and getting Pitch of 102.49 at 75% of blade (pretty good numbers). JC program gives 214.4 mph max and a 1.3 P/D (Pitch/Diameter), where 1+ is good and 1.3 is excellent. the 72" was my friends selected Diameter, which compares to well with Jack Norris's book on props.
I don't know if he has sought a prop builders professional advice on this Diameter.
George
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
1,540
Location
Jackson
My practical, real world experience shows that below 2700 rpm (Lyc 'standard'), and until you get well north of 200 mph, the limit to diameter is gear leg length; not tip speed. I know from personal experience that a 72" prop is just as fast (or faster) than Van's traditional standard of 68" on a short leg RV4, and beats the 68" in climb like a bad dog. The later long-legged RVs can swing slightly longer props; Van sells a 76" Hartzell for the bigger 4 cyl engines. The 'standard fixed pitch metal they now sell is 72" (up from the old 68" they used to recommend), reinforcing the theory that diameter is gear leg limited.

I'll bet that if you run numbers for a clean airframe like an RV, with fixed HP of around 180-200 @ 2700 rpm, and don't ask for more than around 220 mph, you'll get up to at least 76-78" before high speed efficiency starts to degrade, and you'll see climb rate continue to improve as diameter grows. (This assumes pitch is optimized for each diameter.)

edit: The corollary is that tuning the prop slower than the Lyc standard 2700 at full power demands larger and larger diameters to preserve efficiency. So if we're limited to, say 72" by gear leg length (ground clearance), then a too-wide ratio turning the prop down at say, 2000 rpm will be giving up thrust efficiency. I'm no aero expert, but my gut feeling is that it's akin to span/aspect ratio issues with wings. Within reason, more span/less chord improves performance.

Lighter is always nicer, until it causes your wallet to float away....
 
Last edited:
Top