website design software

The PRAGMA is an AVID FLYER HEAVY HAULER I built in 1999/2000 from a kit.

This kit came with a ROTAX 583 snow mobile engine. I considered this engine a no-option for various reasons. In search for an engine, I came across Tracy Crook’s book about his installation of a MAZDA rotary in his RV4. It turned out very quickly that this engine is about twice too big in power and in size/weight. I joint the Rotary enthusiast community  anyway. At the first gathering I attended, someone offered me a Subaru  engine, which ended up in the PRAGMA. This engine served me very well  with absolutely no problem during 10 years of operation.

The name PRAGMA for my airplane was the result of my attitude to building  the airplane. I wanted an airplane that can easily fly off my grass  strip, but does not have to be exceptionally “pretty”. This pragmatic approach gave me the idea for the name.

Even so the Subaru engine installation was a big effort, with redrive and  motorcycle carburetors and so forth, I always considered it a temporary  solution for giving me something to fly until I find a suitable rotary engine.


Conversion issues with the Subaru where no different from what a rotary installer  encounters. Unless you are willing to use an overkill radiator, one has to go through the process of optimizing everything from air flow to  coolant flow, and that is usually time consuming. It was for me too.

It still was worth every bit of the effort.


 What happened to the rotary idea? See One Rotor Wankel.



Installing the Single Rotor Wankel into the PRAGMA.

Toward the end of developing the Single Rotor Wankel named “RES12”, I started getting serious about the installation.

Considering an engine mount, the existing hard points were a real problem in regard to their location. One good point was, there are three hard points on each side, designed for a ROTAX 582 in a cylinder-down position. To make another long story short, I decided to add two more hard point at the  lower outside corners of the fire wall.

So here we go, removing everything from the seats forward and start cutting and welding. This ended with a new wind shield, new  instrument panel and everything that goes with it.

The lower two hard points were added to the frame.

Finally by mid February 11, the instrument panel wiring, as the last effort on the airframe modifications, was completed.


The new instrument panel lay out was, at least to some extent driven by RES12 engine specifics.

finally the engine installation is making some progress with the cooling air ducts taking shape.

Everything seems to be in place

Ready to go, though almost!

By later 2011 the engine was ready for a first run in the PRAGMA.  The engine configuration was identical to the engine in the test stand.

Recalling the configuration.

Fuel system AERO CARB with Holly pressure regulator set to 1.5psi fuel pressure.

Ignition with a StreetFire CDI, triggered with a reluctor type pickup, dual output coil.

Ok, everything checked, ready to go. Cranking-wa-wa-wa-wa-wa. Nothing happened. Checking again, fuel ok, ignition ok? hmmmm.

Unplugged the plugs and put two open observation plugs on. Crank- no spark. Taking the plugs out of the rotor housing, crank, spark. The difference in crank speed made a spark. Checked the trigger signal and it was about 25times better than the required thrash hold. now what?

One thing was certain, the difference between the two engines, test stand and airplane was the fact that the engine on the test stand is cranking faster because it has quite a bit of running time on it. BTW, the difference was approximately 100RPM, 350RPM on the test stand engine and 250RPM on the engine in the airplane.

After some unsuccessful tries to talk to MSD, I set the ignition up on the bench and tested it. It turned out to be a safety feature, new to me, preventing the system to fire if the second trigger pulse does not occur within a certain time after the first one. This totally killed this ignition module for me, after performing on the test stand really well.

Now what? I ended up using a set of ignition modules from a 1983 12A engine I had laying around, but I did not want to use those big antique coils. I found a real neat coil in my 1991 MAZDA truck. Now, I have an ignition system which fires realy relieable, however, for some reason it did not drive the tachometer I was using. My tachometer was one with the plug wire wrap pickup. What ever I tried, I could never get it to work again.

In addition to the tachometer problem, it turned out that the AEROCARB was more likely to flood with this new ignition system than it was with the CDI. I did not want to deal with either one in the airplane.

To make a long story short, I ended up building my own tachometer using a frequency-voltage converter chip and a 4digit panel meter. Works great.


For the fuel system, I went to the ROTEC throttle body injector, which is essentially an Ellison carburetor. Seems to work really good.

I am ready to do the first taxi runs before going into the cowling work.

Taxi testing, except for high speed was done. Everything working as planed.

However, it seems there is to much oil fumes comming out of the crank case breather. more on that later.

Now it is time to get into the cowling modification.

The cooling air inlets are in the prop wash for improving taxi and idle cooling performance. There is also an inlet for muffler cooling air.

                         Radiator outlet is down ward with no restriction.

With the cowling work out of the way, I was ready to go into the high speed taxi testing.

Weight and balance came out good with a 15lb lighter empty weight from the SOOB installation.


High speed taxi testing.

The engine is starting very reliable, and the throttle response is very good.

Cooling seems to be working properly. Water and oil temperature are stabilizing at 170F. OAT was 92F.

Acceleration is as expected, however, at close to take off air speed with WOT, the engine starts sputtering and loosing most of its power. After a few runs and some playing with the mixture control, I determined that the engine was running way too rich. Actually, the mixture control had to be turned halfway to full lean for the engine to run at full power. Now what?


Since  ROTEC had obviously not tested the TB on a similar engine, more accurate, a 1/2 RX7 engine, I had to start my own investigation into what is going on. I also wanted to use a tuned intake, which, in a P-Ported engine should be upstream from the throttle body in order to improve idle performance.

The result of the investigation is a system with an adjustable pressure pickup where the mixture at full rich mixture control is set. This is a fixed setting once it is done.

Tuned intake with reference pressure pickup for the fuel pressure regulator.

Hope this is the final fix for having a flyable airplane.

I have been wrong before!!!! And here I am again.

No matter what I did, the WOT sputtering did not go away. Finally I decided to investigate the problem on the ground test engine.

The first conclusion I reached after a lot of testing and analysis was that the problem was not mixture control but fuel flow between the pressure regulator and the throttle body. Since I did not have a detailed drawing of the interior of the throttle body, I had to do some guess work.

The first thing I did was trying to reduce some of the daed space which created airpockets. Did not make any difference.

Next I played some more with the pressure regulator location and the plumbing to the Throttle body, since this is mentioned by Rotec as a crittical area.

I reached a point were everything worked really nice. No more sputtering at WOT, and no hesitation at fast throttle opening. Full power seemed to be better I had with anything I had tried before. 

Ok so far. I copied the installation of the test stand exactly to the air plane. Started up the engine, warmed up to < 140F on oil and water. Now WOT! aarrggg!!! still sputtering.

What the****** is the difference between the test stand and the airplane installation? 

                                 Test stand installation.

Airplane installation (lowerd on the right from tail dragger effect)


I could not believe it!  The two pictures show the only difference.

The slide throttle on the test stand is oriented horizontal, in line with the e-shaft axis. On the airplane, the slide throttle is pointing up in the front, but is still in line with the e-shaft axis.

Back to the test stand.

Orienting the TB as it is on the airplane, the sputtering is back. Orienting horizontal or pointing down in front, running perfect.

This “Tail dragger effect really saved my day”.

The actual difference between running right and not running right is about 5deg. I leave it to the immagination of the reader what could happen, and probably would.

I totally fell for a statement in the Rotec description that “The TBI cam be mounted at any attitude and at any angle” with no adverse effect on its operation. Hahahaha!


Corrective meassures turned out to be a lot more work than I first anticipated.

The throttle linkage on the air plane where a stright forward solid push rod. This did not allow reorienting the TB without a lot of redesign and ending up with another inflexible situation. A flexible option with a Bowden cable and spring loaded turned out to be not acceptable in the airplane. I did this on the test stand.

I finally settled with a double action Bowden cable solution. The pictures should be self explanatory.

                                 Double action Bowden cable setup

After doing some cleanup work and making sure everything is secured and safe, Iam ready for doing some crow-hopping down the runway before taking her up to the first flight.


This is the status at the end of June 2013.

High speed taxi with crow hopping totally successful. Engine control is smooth and direct.

I am glad I stayed with the TB40 until the issues were corrected.

Now I have to go back to some of the cosmetic work I postponed for so long.