Q-talk 86 - Turbo Subaru Installation - Part 3
- Category: Q-Talk Articles
- Published: Wednesday, 23 December 2009 16:24
- Written by Jon Finley
- Hits: 864
Turbo Subaru Installation
(Part 3 of 3) Continued from Issue 85
I have performed several engine swaps and am adamant that the engine/airframe work properly on the ground before flight is considered. This includes cooling. I have heard many builders state, "it won't cool on the ground but maybe in-flight there will be enough airflow for it to cool." I feel that this is a great way to die. Since I have no desire to die, I ran the engine many, many hours on the ground. All sorts of runs were performed, everything from long idle sessions to long high power runs (tied down). When I finally began flying, I restricted my flights to orbiting the airport. We are a little jammed into the Class B Minneapolis airspace so I could only go up to 1500' AGL for my orbits. I spent about six hours over the airport before venturing away from the airport.
1 typically take off at full throttle with the wastegate open. Due to the configuration of the turbocharger and wastegate, this setting yields about 3" of boost. I normally see about 3200 rpm during initial climb. I maintain these settings for the first few thousand feet or until I am out of the airport area. As altitude increases, the amount of boost decreases, at 4000'-5000' MSL the boost has dropped to about zero (normalized). When additional performance is desired, the wastegate can be closed to achieve the desire level of boost. However, I found the additional workload that this introduces to be a safety issue. It is far too much work to manually control the boost level (there is a lag and then it changes VERY quickly) and do all the other things required. Typical cruise rpm is 3400-3500 rpm. The addition of boost during any phase of flight is quite noticeable. A high priority task is to create a mechanism to allow control of the boost level AND utilize overboost protection. This setup will allow the use of maximum boost during takeofC climb without increasing pilot workload.
I am still working on propeller tuning. I initially flew with 11.5 degrees of propeller pitch. While climb performance was exceptional, cruise speed was very poor. Next, I tried 12.5 degrees. This was too much pitch as climb suffered considerably. I then tried 11.75 degrees and have since changed to 12.25 degrees. With each flight I become more familiar with the engine and turbo and am better able to use the power. Obviously there are a number of linked factors that affect this effort and I expect to make steady gains as I learn how to use the turbo and engine.
I still don't have accurate numbers but will report them in the future. On my most recent flights, I have been cruising at approximately 140 MPH IAS and showing about 150 MPH TAS.
Many people have asked about the power output. The accurate answer is that I do not know. I believe that there are a huge number of variables that effect power output from any engine and the only way to truly know how much power is being delivered to the prop is to run the engine on a dyno exactly as it would be run on the airplane. So, with that said, the only thing that I can offer are calculated numbers, take them for what they are worth. The Subaru manual lists a normally aspirated EA-81 running at 3800 rpm at about 70HP and 92 ft/lbs of torque. CALCULATED HP and torque at 3800 rpm and 7 lbs of boost (45" manifold pressure) is 93 hp and 128 fit/lbs.
I have noticed that the three-bladed prop seems to produce considerably more drag than the two-blade wood prop during landing. I previously used my speed brake to reduce float and ground roll. Now, I find that I do not experience much float and believe the difference is the drag generated by the prop. I am pleased with the smoothness of the Subaru. When I relocated the radiator to the belly, I noticed a new "rough" sensation. I think the air hitting the radiator is very turbulent. I think that once the proper radiator ducting is installed this sensation will be eliminated.
I thought my research had uncovered the details of operating a turbocharged engine. Only now, after operating one, can I appreciate the complexity that the turbocharger adds. I am of the opinion that a turbocharged engine should use computer controlled fuel injection and ignition timing. The additional fuel requirements when operating under boost, the timing changes required when under boost, monitoring of detonation and control of the wastegate are all extremely difficult to monitor and adjust while flying the airplane. Do to the workload, I find that I am not taking advantage of all the power that the engine is willing to produce for fear of overstressing it or operating it with unsafe parameters. In addition, the turbo runs HOT-HOT-HOT (hotter than I realized). You must plan ahead for this fact.
The following list represents the major changes that I would like to make in the near future:
- Reduce cowling size. The bottom extension was added to enclose the radiator. Now that the radiator has been moved it is empty space. In addition, the amount of cooling air being allowed into the cowl can be reduced significantly.
- Intercooler. The engine would benefit greatly from a cooler intake air charge. I believe I can reduce the inlet air temp by at least 50 degrees (probably more) with an intercooler. This will yield a corresponding increase in performance while reducing the possibility of detonation. This will add an additional inlet/outlet to the cowl.
- Turbocharger over-boost protection. I would like to install the OEM wastegate control on a manual control lever such that I can control the boost level and have overboost protection.
- Radiator fairing/pressure recovery cowling.
- Knock sensor.
- Computer controlled fuel injection and ignition control. I am very excited about Tracy Crooks product (www.rotaryaviation.com) and hope to install it at some point.
- Aftermarket camshaft. Cams exist that make more power and torque in the rpm range that I am operating. I am searching for a proven camshaft for the EA-81 direct drive, turbocharged application before committing to this change.
- Prop tuning. I am told that the Warp Drive is not tailored to high-speed flight. Several people have optimized the WD prop. At some point it would be interesting to pursue this.
I have a long list of fuselage modification that I would like to make. I feel that I have a pretty dirty airplane and think there is speed to be gained through a clean up effort.
Would I do it again? Depends on which day you ask! Installing a non-standard engine is a tremendous amount of work. There are a lot of Subaru resources available now that did not exist when I started the process. I strongly recommend using these resources. The ability to purchase components will save hundreds of hours and possibly a great deal of frustration. If you really enjoy the building/ fabrication process, experimenting and are going to stick to it - the results are very rewarding. If you are not the mechanical sort and/or just want to fly, install an 0-200. If I were to do it again I would use the Subaru EJ-22 Legacy engine. It is 2200cc engine with overhead cams and four-valves per cylinder which produces 140 HP in the car (160 HP turbo). It is a heavier than the EA-81 but produces quite a bit more power. I believe, with close attention to weight savings, a 120 HP direct drive turbo EJ-22 could be built weighing about the same as an 0-200 and could cost significantly less. I believe the Subaru engines are very well engineered engines and can be very reliable aircraft engines.
Roger Enns has been of tremendous help to me throughout this project. He deserves the credit for many of the ideas that I have implemented and has been a very valuable resource - always willing to help. I appreciate his assistance more than I can state. Reg and Bud Clark were the pioneers. Bud has always been willing to help in any way possible and I appreciate his assistance.
You can order a printed copy of Q-talk #86 by using the Q-talk Back Issue Order Page.