Subject: Mistral engines Paul Janssens-Port Inserts-sound print reduction program?
From: rotaryeng
Date: 9/16/2013, 6:56 AM
To: AAA-rotaryeng





   Hi Paul.

   I think you'd like an update on what's happening with my Mistral engine,
   so here goes.

   I've been flying the Glasair since Dec 6 of last year. engine works fine
   and I'm very pleased with it.

   There are still some aircraft electrical signals interfering with the
   engine's sensors, so i'm getting the same warnings during flight, all of
   which I've come to understand quickly, and Mistral will be adapting the
   waring thresholds to suit my installation.

   I'm averaging 54 liters per Hour ( about 13 USG/H ) with mogas, and 0.2
   of a quart of oil. The 300 hp engine is just wonderful.

   My cooling is only just satisfactory. At OAT of 30°C, i do have to
   reduce immediately after gear retraction, and speed up to 140 KIAS so
   get more flow. About 2 minutes into the flight, the heat wave from the
   take-off reaches the sensors, and they'll set off the warning range of
   the system, but i still have enough margin to the alert threshold, and
   in 2 min the wave is cooled out of the system.

   Longest flight so far was 2h 45 min, at FL125 where I had about 53 %
   power remaining. I have yet to experience any trouble with the engine,
   but the ECU's sometimes bug up when i start them up, the remedy is
   simple, power down, wait a minute, and they'll do just fine.

   I have been doing aerobatics, and sustained G-turns to test the oil flow
   with Mistral always checking the recorded data afterwards. Up to 4
   sustained G's is all we did so far, but not a hint of any trouble. The
   Mistral has a wet sump, and we did see the early signs of oil pressure
   loss at values between 0 and 0.5 G, so we have chosen to stay away from
   the region < 1G with this engine configuration.

   I had a problem with MT propeller. they sold me what they said was the
   prop, specially made for this engine, which turns at 2350 rpm at
   take-off, and 2000 in the cruise. What they sold me was an off the shelf
   prop for a lycoming, which turns at 2700 and 2300 resp. so when i
   started doing my dive tests, the prop started overspeeding at around 210
   KIAS. I was a bit surprised, but did some test flights and calculations,
   and presented them with my findings. They initially tried to tell me i
   was wrong, but ended up sending me a mod that allows the prop to go to
   45° pitch, when the original setting was 35°. This clears the overspeed
   issue, and the prop is generally satisfactory, but my calculations still
   tell me I loose 17 % because the prop is turning at 300 rpm below which
   it was intended for, and that it's working at 5° higher angle of attack
   than it's ideal value. So far, MT has chosen to pretend I'm not there
   with regards to that issue. Time will show that to be a mistake.

   Noise....BIG problem.

   Mistral gave me a long absorption muffler from Chabord in France ( the
   one you see in the pictures ). Chabord sells these to be put on
   lycomings, and of course the result on my Glasair is catastrophic. we
   measured 93 DB at take-off, and 87 DB in a cruise setting overflight at
   1000 ft AGL, but that frequency....

   The extension pipe is my request. i didn't want the muffler to hang just
   under the fuel bay in the wing, after all, it's an experimental aircraft
   with retracts.

   Chabord took about a year to build another muffler ( bet ya the French
   take less time when it comes to cooking or making love ), and it was
   even louder than the original. Mistral then tried a tangential muffler
   from Rochhausen in Germany with a dovetail exhaust, on a 200 hp, 2-rotor
   engine, but no real improvement.

   I displayed the aicraft at the big Friedrischafen fair for Mistral
   engines ( you can say it's about like sun and fun, but mostly without
   the sun ). The reactions from the public and professionals were VERY
   positive. Mistral has now shifted their priority to sound reduction as
   they know that they won't be able to sell any more engines like this,
   and the way things stand now, we are rapidly making a bad reputation for
   the product. Clients are lining up for this engine, and there is some
   urgency if we don't want to loose them.

   We have measured frequency ranges on my Glasair, and a theoretical
   simulation will be ordered soon.

   Can you tell me if there has been any significant progress on noise
   surpression on your side of the Atlantic ? Mistral will provide me with
   a good solution in due time, but in the mean time I'm ashamed to fly my
   Glasair because it does disturb people on the ground. I haven't built
   this aircraft to give aviation a bad reputation.

   with kind regards,

   Paul janssens

   Belgium

   Rearrange the cooling system on the right side of the cowl. Put the rad
   core next to the cowl surface and the diffuser between the core and the
   hot engine to isolate the engine heat from affecting the rad core. Put
   vertical louvers on the right side of the cowl to let the hot air from
   the rad core escape.  They can be made adjustable to reduce the drag at
   cruise speed or on a cold day.


   This will improve the cooling.

   We have a new coaxial muffler design that works a little bit like a ram
   jet, ejector or thrust augmenter so it does not add too much to the
   aerodynamic drag. The tone of the rotary sound is much more pleasant.

   Peter G, I thought you might also be interested in this coaxial muffler
   design speaking of turbo compound.

   Paul Lamar


   Thanks for the update Paul. Glad to hear the 3 rotor is going well.
   What
   are your oil and coolant temps running at? From memory your main cooler
   is much larger than Tracy's is using on his 3 rotor. He has I believe
   540cu inches of cooler. See photos of his cooler and exhaust setup.

   Steve Izett

   Paul, check your rotor housing for the presence of these Inconel exhaust
   port inserts.  They were used in all normally aspirated 13B's to quiet
   the exhaust.  Mistral built a lot of engines with stock Mazda parts. The
   3 rotor was sold only as a turbo charged engine so it did not have them.
   Breaks up the supersonic shock waves coming out of the open port which
   is all of the time :)

   Paul Lamar

   Here is some more info on the subject.


   Paul Lamar --

   Might this be time to install a turbocharger to see if that could lower
   the sound (and provide other benefits, once the cooling is sorted out)?
   Marc



    hi Marc, mistral has tested the mazda turbo, but they weren't satisfied with it. Mistral is very safety minded. They have a lot of demand for turbo's, and will proceed to develop them at a later stage. I agree that that will help a lot, and for a Glasair, a turbo is a good idea.

   Paul janssens

   I've looked at a lot of turbo maps and have not found one that would work for a 3-rotor at both sea level and at 18,000 msl.

   Mark S.
   Lancair ES, N/A 20b

   Could you explain that a little more? How is the 3 rotor different in that regard than any aircraft engine?

   Paul Lamar

   The real question is why boost this engine at SL at all (if we want reliability)? All I've ever wanted to do is 'turbonormalize' it, IOW maintain SL pressure, hence engine output, at 25,000 or so-a typical certified application on many light aircraft. We only need to boost it when we begin loosing pressure as we ascend. Turbo maps have me a bit flummoxed as well because they are made mostly for cars and trucks operating at SL or close to it. I am sure an aircraft turbo spec would be much better than any Mazda turbo....... matched it to the output of the 3 rotor. An intercooler would be necessary as well. The controls to close the wastegate as you ascent could get bit tricky.....aircraft that are turbonormalized and FI (deck pressure injector shrouds, pressurized magnetos, etc) have many oil fed parts, but the new diesels (all turbo'd) and many never aircraft now have electronically controlled wastegates, PRVs, and deck compensators.
   Marc




   The last turbo aircraft that I worked on was a couple of months ago, and was a Seneca 2, with 6 cyl Cont TIO360s. The turbos were mounted behind the engines where the exhaust piping came together (about 2 feet behind the engine in the back of the cowling) and had manual waste gates, no pressurized mags, no intercooler, no extraneous parts. Flying the plane is a workout, especially during takeoff where you have to closely monitor temps, pressures and everything else normal pilots have to monitor. It proved very easy to over pressurize the engines during takeoff, even with the waste gate "closed". If the MP went over 30", you had to pull the throttle back a bit to keep it under control. The turbos were small Garretts, not much larger than the stock Mazda turbos. There was about 2 feet of piping between the closest cylinder and the turbos, more distance to the front cylinders from the turbo. The oil line came off the back of the engine, into the top of the turbo, and out the bottom just like one on a car and drained back into the oil pan. Nothing special. I did notice that the oil filter was twice as tall as one on a non turbo engine.

   Adding turbo normalization to a plane doesn't have to be that complicated. A manual waste gate that is left open for takeoff and gradually closed to maintain 30" MP as we climb or in cruise would be sufficient. I would argue that an inter cooler isnt mandatory, if you only plan to use the turbo at high altitudes where the air is considerably cooler than at SL. I have seen more than one turbocharged certified plane without inter coolers.

   Can we expect an altitude of 25k feet? Not realistically IMHO. I think a ceiling of 18k would be more than plenty. Just like in an automobile application, turbos are matched based on what you want performance to be. If you want a killer stop light to stop light response with little turbo lag, a low AR and small diameter impeller will get you there, but it will choke out at high RPM/pressures. For a diesel or continuous duty application, an AR closer to 1 would be better. A larger diameter turbine/impeller, larger gas chambers for better flow, and of course turbo lag. Match the flow of the turbo with the flow of the engine at the intended RPM and work load range.  I went with a modified Garrett T04, optimized for the flow of a 5 liter diesel at 3000 RPM. In this application, it produced 16 psi continuous. I plan to mount it like the Seneca, behind the engine with around 2 feet of exhaust pipe between the engine and the turbo. I will install a cooling cuff around the pipe at the extension piece that dumps overboard to cool the pipe before it gets to the turbo. This addresses these problems-

   1. Hot exhaust- the exhaust is cooled for a bit between the engine and the turbo-
   2. Noise- I still get my turbo, which is known to quieten the engine.
   3. Turbo-charging- I will set it at around 8psi to start with with a mechanical spring loaded waste gate actuator. I am expecting that 8psi spring pressure will remain fairly constant since it doesn't rely on differential air pressure to work.

   My NA viking had a ceiling of 15k, and with 300 hp would get there no problem. There are vikings with turbo IO540s and now 550s that are normalized with ceilings of 17k. My Viking cruised at 160 knots at 12.5 gph at 12500 ft, the turbo version gets 180 knots at 15000 on 15 gph. A little faster, but lots more fuel burn. Again, manual waste gates for the vikings. Procedure is to set the wastegate to open for takeoff at SL, and if you are at higher elevations do not exceed 30" MP. Again, one more thing to monitor during a busy time.

   With an EFI, automatic operation could be achieved, just read the outside pressure, and the MP, and make sure that the MP stayed within limits. And temps, and RPM, etc.
   I am working on this to be implemented in the EFI system.


   Kevin Alderman



-- 

The Rotary Engine News Letter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
You Tube http://tinyurl.com/beqqxas
Copyright 1998-2013 All world wide rights reserved.