Turbo Subie Passes 250 Hours on the Hobbs

My Dad and I just finished a 2.1 hour cross country flight up North yesterday and on shutdown, the Hobbs clicked over to 250.0 hours- a small but important milestone.

How has the experience been?

The first two years seemed to be spent making countless modifications to the cooling system. Last year all I really changed was the type of coolant going from Evans NPG+ back to EGW (70% water, 30% Prestone EG and a touch of Redline Water Wetter). Since that last mod, the cooling system performance has very good. Temperatures in the climb rarely exceeds 180F making it one of the coolest running Subes flying.

I did add a new turbocharger scavenge pump this spring going from a diaphragm type to a gear type which also saved 3 pounds.

The core of the engine has never been touched in this time despite a forced landing due to my fault in not noticing a bum alternator and one time losing all the coolant and flying for 11 minutes. Compression and leakdown were good so I continued to fly it.

The engine was hard on spark plugs initially and I had to go to platinum types to solve this.

I had to remove the oil pan last year to fix a persistent seep from an unused turbo drain fitting.

A couple years ago, I removed the turbocharger to install a larger compressor wheel and housing for better altitude performance. The hot section parts and bearings are original.

I've had zero problems with the Marcotte M-300 PSRU and zero problems with the IVO Magnum prop. I get about 50 hours between brush changes on the prop ($7.50/ set).

Lately, oil consumption has been creeping up and compression creeping down. I'm attributing this to a steady diet of 100LL which has been slowly sticking up the rings in the grooves (much tighter tolerances than a Lycoming). I've lately been feeding the engine a dose of Marvel Mystery Oil at each fuel fill as this has been shown by others to free up lead bound rings on Subarus. I hope this works.

Performance wise, we see around 1000 fpm climb at 1750 lbs. gross at +20C and 4000 MSL and 2000+ fpm solo/ 1/3rd fuel and at around 0C, again at 4000 MSL. Due to fuel prices today, I generally cruise only at 25 inches now and 4400-4600 rpm. Leaned to 1400F EGT, this gives us a fuel flow of just over 7 US Gal./hr. Speeds range from about 135 knots TAS at 6000-7500 MSL to around 140 knots TAS at 9-10000 MSL. The airplane will cruise at 160-165 knots true above 14,000 feet using 30 inches and about 8.8- 9 gal./hr.

The engine package has been remarkably reliable in the last two years and we are having a lot of fun flying it.

I plan to revise the radiator system in the future to reduce drag, weight and complexity over the present system.

I've learned a lot, especially in the early years and it has been satisfying (and sometimes frustrating) developing this unique propulsion package from scratch.

Ross, thanks for the report.

Even though I have no intent to put a Subie in my plane, I really do enjoy the learning experience from following the various threads on this subject.

Thanks again,

Good report, Ross.

I'm all out of ideas with the H6 and the original small EGG radiators designed for the 2.5 engine. Yesterday after the second take off coolant temp hit 220 before reaching 2500' so it was level off and increase prop pitch as usual. The cowl has more than adequate exit area so the only thing left to do is modify the front of the cowl to provide for direct ram air to the radiator surface. At present I am guessing the Van's cowl inlet is about 60% of the boxed radiator area. Air has to flow around the inlet opening and find its way to the radiator and there probably is much turbulence. Straight in would be better as Jan has found out with a new cowl. I can make a cold engine take off and climb OK but with a heat soaked engine, forget it. The H6 needs more radiator area and that's why current engines have it. The EGG factory recommends cowl flaps but with the small rads there is inadequate cooling with a fixed opening and closing it down would only make it worse.

I could change from NPG to water and EG anti freeze as per your set up. But that would necessitate increasing coolant system pressure. At present it is 7. At what pressure are you running the cooling system? Will changing to water and EG net 10 degrees better cooling? I guess I could try it and find out.

Cruise speeds relative to fuel flow are similar. I have been using 8 gph and 143 KTAS in the past but it may be a couple knots less since opening the cowl for more air flow plus the new internal muffler may have a bit excess back pressure. There is an EGG recommendation to drill out some of the internal baffles to correct it.

The cowl exit was at 5x17 but I closed it down to 4x17 after initial cooling results were not much better and there was some added drag. The cowl exit has been modified three times and I hope to fill and finish it this summer so it can be painted again. The new muffler has not affected oil temperature one bit even though it is within .75" of the pan. It is quieter, the 2 torpedo external mufflers are gone, and the bottom skin no longer vibrates.

David, I run a 13 psi cap and saw a 10-12F reduction in temps just switching the coolant with no other changes. My opinion is that the 2.5 rad setup is not adequate on the H6 for extended climbs on a hot day.

I wanted to be able to climb from my field elevation (4000 MSL) directly to 10,000 at 85 knots (my best climb speed) and +25C OAT without the coolant temps exceeding 200F. This is no problem now.

The fun or trial and error and educated guesses!

At present I am guessing the Van's cowl inlet is about 60% of the boxed radiator area. Air has to flow around the inlet opening and find its way to the radiator and there probably is much turbulence.

David: It doesn't sound like you have attempted to place a "trumpet" shaped fiberglass fairning from the cowling inlet to the radiator face without leakage? Our rotary guys have found over the years that this is very beneficial to improved cooling.

Doug Lomheim
RV-9A, 13B FWF

Cool

Thanks Ross,

This power plant has obviously been the tinkerer's dream...And you don't have to worry about a potential $3600 hit to your pocket at 350 hours plus as us ECi drivers now do..

Good job. I had an EA81 soob conversion in my Zodiac and despite the fact it tried to kill me three times I really enjoyed the proces of finally turning it into a reliable package.

Cheers

Frank

Here I've been telling people who don't want a Subaru to get a Lycoming clone and then these cylinder things come along. Looks like a "real" Lycoming might have been a better choice. Arrrgghh. I feel for anyone who has to lay out money for things like this.

I guess we all have our burdens to bear. If my rings don't unstick in the next 50 hours or so, I may be pulling some stuff apart too.

ROSS

Congratulations on your milestone!


DAVE

Exit flow and shape would be my guess as the best way to improve your cooling. I say this because I assume you are just dumping the outlet of the box radiators into the lower cowl. If I am wrong my apologies.

I don't know if you have read the posts on RV Cooling from a traditional engine perspective, but there is good info and links for reading that would pertain to your water cool engine. You also have an -A which has caused some exit area cooling flow difficulties. A picture has been posted which shows flow into the exit.
Completely controlling the outflow of the radiators all the way to the cowl exit and smoothing the cowl exit would provide a substantial improvement. A good "trumpet" inlet is also recommended as non optimal shapes have been shown to leave the edges of the radiator devoid of airflow.

My 0.02 anyway.

RV8RIVETER said:

ROSS

Congratulations on your milestone!


DAVE

Exit flow and shape would be my guess as the best way to improve your cooling. I say this because I assume you are just dumping the outlet of the box radiators into the lower cowl. If I am wrong my apologies.

I don't know if you have read the posts on RV Cooling from a traditional engine perspective, but there is good info and links for reading that would pertain to your water cool engine. You also have an -A which has caused some exit area cooling flow difficulties. A picture has been posted which shows flow into the exit.
Completely controlling the outflow of the radiators all the way to the cowl exit and smoothing the cowl exit would provide a substantial improvement. A good "trumpet" inlet is also recommended as non optimal shapes have been shown to leave the edges of the radiator devoid of airflow.

My 0.02 anyway.

Click to expand...

Thanks, Wade, what you suggest is very true.

A smooth plenum would probably make a world of difference as was well established as long ago as WWII. I think the diffuser angle is something like 12-14 degrees to and from the radiator to provide for no separated air flow.

The problem creating a plenum with this installation is the radiators are located quite close to the engine block, there isn't room for a plenum. The air coming through the radiator hits the engine and has to make a 90 degree turn to escape. The right radiator is about .75" from the block, the left a bit more. Air does find its way out of the compartment, but it is not a smooth flow.

Eggs biggest compromise

David-aviator said:

Thanks, Wade, what you suggest is very true.

A smooth plenum would probably make a world of difference as was well established as long ago as WWII. I think the diffuser angle is something like 12-14 degrees to and from the radiator to provide for no separated air flow.

The problem creating a plenum with this installation is the radiators are located quite close to the engine block, there isn't room for a plenum. The air coming through the radiator hits the engine and has to make a 90 degree turn to escape. The right radiator is about .75" from the block, the left a bit more. Air does find its way out of the compartment, but it is not a smooth flow.

Click to expand...

David,
I have seen and worked on several Egg Soob conversions. The lack of exit space on the radiators is the biggest compromise Jan has made with respect to cooling. The testing that has been done in the past and more recently shows that the exit of the radiator can be even more important than the entry. Some people question the radiators on the side of the engine, or below. These locations will usually allow for better exits than the front cheek radiators. Egg has chosen to go with the cheek location for packageing reasons mostly. A better radiator location would help cooling a lot. IMO
Bill Jepson

Update.

I was doing an oil change and some checks on VZX this weekend. As I mentioned previously, the compression has been slowly falling for some time and oil consumption has risen to Lycoming levels lately. The latest compression tests have shown a reversal in the dropping trend and they are now starting to improve. I've been lacing the fuel with Marvel Mystery Oil for the last 20 hours or so in hopes of freeing up suspected stuck rings. The theory was that the 100LL was not agreeing with tight Subaru tolerances here.

The second item of note was that the NGK platinum spark plugs are not showing the high wear of the conventional plugs fitted earlier.

I changed the oil in the PSRU also. This looked nice and clear and did not have anything nasty (metallic) in it.

I also changed the IVO propeller brushes and did a good inspection of all hoses, engine mounts, exhaust etc.

The new scavenge pump is nice and dry which is really nice after the earlier one which leaked like a sieve.

I found a problem with the alternator belt idler. The bearing was loose again and upon further examination, it was evident that the bearing had spun in the pulley. I'll be boring the pulley to accept a larger bearing this time in hopes of making this part last longer this time around.

I hope to have it back together mid week and start piling on the hours again.

I need to get back to my RV10 doors soon.

At Almost 300 Hours Now

I'm just completing the annual on VZX this weekend.

The Hobbs is now showing 297 hours and I did a dry and wet compression test plus pumped some high pressure air into the cylinders to listen for leaks. Bad news is that compression has dropped a bit more in most cylinders both dry and wet. There is a slight hiss out the breather pipe which is no surprise and a bit from the exhaust pipe as well. So it seems like the engine is just getting a bit tired and perhaps the 100LL is making valve seal less than great.

Oil consumption has seemingly stabilized (I hope) at around a quart every 6 hours or so. When new, it didn't need anything between 40 hour changes. I switched to Aeroshell about 20 hours ago in hopes that the sticky rings might get freed up. Appears not to have done anything so I'll switch back to Mobil 1 at this change.

I'm going to try running mogas instead of 100LL over the next year and monitor compression every 20 hours or so. Maybe if I get all the lead out of the system, something good will happen.

Some reports from NZ have indicated problems with compression loss due to exhaust valve erosion on Subes running 100LL at the 200-300 hour mark. Just wondering if anyone else has accumulated these sort of hours on a Sube running only 100LL and what problems if any they have had? Engines running mogas appear not to suffer these issues with one guy down under reporting 3800+ hours on an EJ22 without ever being opened up!

The PSRU oil is still coming out clean with no signs of distress. No problems with the turbo, exhaust system or the dreaded IVO propeller. Cooling is great and the NGK platinum plugs are holding up well.

I have to remove the left fuel tank to fix a persistent seep from the sending unit but other than that, things appear pretty routine with no big issues so far.

Rotary10-RV said:

David,
I have seen and worked on several Egg Soob conversions. The lack of exit space on the radiators is the biggest compromise Jan has made with respect to cooling. The testing that has been done in the past and more recently shows that the exit of the radiator can be even more important than the entry. Some people question the radiators on the side of the engine, or below. These locations will usually allow for better exits than the front cheek radiators. Egg has chosen to go with the cheek location for packageing reasons mostly. A better radiator location would help cooling a lot. IMO
Bill Jepson

Click to expand...

The next biggest blunder that Egg did aside from the "no room for exit air" is that he took some standard Oil coolers and "grafted" on the 1" large H2o hoses to make them look like radiators. When you look inside those nicely welded on hose ports you see a 3/8" hole actually entering and exiting the cooler, not exactly what the coolant wants to see.... Then there is the debt of the cooler, about 2" too deep to be efficient.

I think many of the STI guys have now realized that there are far more efficient coolant radiators available and are fitting them.

Cheek mounted rad setups however are always going to be relatively high drag and inefficient compared to proper, duct mounted setups. Cheek mount is convenient to be bolt on out of the box with minimal hose runs and cowling modifications which is why Jan is using them.

Hi Ross,
Why do you say "dreaded IVO propeller"? From previous posts I thought you liked the IVO...

-Dj

I was sarcastically referring to all the "experts" who told me how terrible the thing was going to be. It has worked flawlessly for me so far...

Ah! Good to know. Thanks! My intent is to use an IVO on my Sportsman 2+2 with the Eggenfellner E6 (Arriving next month!!!).

Do you have any issues with the prop in really cold weather? I'm in Maine, and I think you are in Canada, and are likely to see below zero temps like we do here.

-Dj

rv6ejguy said:

Oil consumption has seemingly stabilized (I hope) at around a quart every 6 hours or so. When new, it didn't need anything between 40 hour changes. I switched to Aeroshell about 20 hours ago in hopes that the sticky rings might get freed up. Appears not to have done anything so I'll switch back to Mobil 1 at this change.

I'm going to try running mogas instead of 100LL over the next year and monitor compression every 20 hours or so. Maybe if I get all the lead out of the system, something good will happen.

Click to expand...

Subies are not made for leaded fuel. Running synthetic oil with 100LL is not recommended as synthetics cannot disburse the lead. AeroShell 15w-50 is semi-synthetic so it cannot help in an "over leaded" situation. Try Phillips 20-50 for a few oil changes. Phillips is 100% dino oil with good lead disbursing qualities.

There is an additive called TCP that you can add to the fuel (and maybe oil) to help disbursement of lead deposits. We had to use in it Rotax 912's to "get the lead out". http://www.aircraftspruce.com/search/search.php

I'm not sure if TCP will help, but it is the only thing I can think of that scavenges lead.

I've been using Decalin lead scavenger for several years and internal inspection of the engine has showed no signs of lead deposits in the lower end, probably due to frequent oil changes. For the last 50 hours or so, have also been adding MMO to the fuel as many people swear by it to avoid lead problems.

I've just filled the tanks with Mogas and will be doing a test flight this afternoon. I'll stick with the Mogas for the next year and see what transpires.

We are on the learning curve here with what does and does not work. Experimental aviation at its best.

IVO Hub details

Ross-
Great to hear of a successful Subie-IVO combination. Were they able to deliver a prop that could be bolted directly to the PSRU hub? Or, did you need to manufacture/purchase any bushings, spacers, etc? Did you go with 3-blade? Ground adjust or inflight adjust? I welcome all details.
Thanks,
Greg

GregClawson said:

Ross-
Great to hear of a successful Subie-IVO combination. Were they able to deliver a prop that could be bolted directly to the PSRU hub? Or, did you need to manufacture/purchase any bushings, spacers, etc? Did you go with 3-blade? Ground adjust or inflight adjust? I welcome all details.
Thanks,
Greg

Click to expand...

Sorry I missed your post from a while back.

IVO supplied the prop and adapter to SAE 2. The gearbox manufacturer machined the drive lugs to suit the IVO adapter. I did have an issue with incompatible radii on the prop flange and adapter which took a little lathe work to rectify: http://www.sdsefi.com/air18.html

I have a 76 inch Magnum 3 blade with the high pitched blades, in-flight adjustable.

Interesting to look back what I wrote 10 years ago now- I was hoping for 3000 fpm solo and 200 knots... I was young and foolish back then. I ended up with 2000 and 180.

rv6ejguy said:

Interesting to look back what I wrote 10 years ago now- I was hoping for 3000 fpm solo and 200 knots... I was young and foolish back then. I ended up with 2000 and 180.

Click to expand...

It never hurts to dream!

HI Ross, I started with a NSI 2.5 installation. I was buring 100LL and made it to 75hrs before removing the highly modified engine. It had burnt exhaust valves. I replaced it with a STI SOHC engine, all stock. I ran a turbo for the first ten hours and after that I took the turbo off. Running 100LL with an additive this time, I got to 80hrs before burning the exhaust valves. I rebuilt the heads and this time I got 75hrs before burning the exhaust valves. Max EGT's 1425 or less. I have just rebuilt the engine. New SS valves, new pistons etc. It's running better than ever and I am running 94 car gas this time. I don't know what is burning the valves, I give it lots of gas and the timing is max at 31degree's under high power. A friend is running a NSI engine on a Glass Star. He had problems at 250hrs with burnt valves. Once that was fixed, all he does is fly.........over 1000hrs at this point. He burns car gas. His timing and fuel tables are close to mine. So??? I am going to burn car gas this time around. Time will tell)
Steve rv-6 2.5 sti 9.5 to 1 compression 27"/4400rpm/1950rpm prop 1600' 162kts

Hmmm. Very interesting. I've clicked over 300 hours now and the valves are not burnt yet but it does not seem entirely happy on 100LL either. I'm burning 80/20 91 unleaded/ 100LL now and checking the vapor pressure with a gauge from Peterson. The winter gas up here is apparently kinda scary so I need that Avgas in there.

One other guy in Oz had a reported 3800 hours on an EJ22 in a gyro trainer, always run on unleaded. There seems to be a common thread here.

Out of curiosity, how wide were the exhaust seats ground?

I'm really impressed with the speed you are getting from yours.

Ross, Steve and or anybody. What would be the mechanism at play here when we see 100LL as the commmon thread for burnt valves? How would the 100LL contribute to that type of problem?

I can understand the lead getting the rings gummed up or stuck. Does the lead get the valve guides stuck enough to prevent proper valve closing?

Buildup on the valve faces?

I am more than a little curious as you can imagine

Randy C

I never see any evidence of lead buildup in the guides or stems. In the old days of leaded auto fuel, burnt exhaust valves were somewhat common on certain engines. This seemed to be due to lead buildup between exhaust seat and valve face.

lead problems?

rv6ejguy said:

I never see any evidence of lead buildup in the guides or stems. In the old days of leaded auto fuel, burnt exhaust valves were somewhat common on certain engines. This seemed to be due to lead buildup between exhaust seat and valve face.

Click to expand...

Ross,
I have seen several examples of burned exhaust valves caused by leading of the valve stems in the guides. These were not subies, rather motorcycles, and VW's used in migets, and both in racing applications. Not being a subie expert I'd ask you are they using sinstered iron or typical bronze guides in the late-model Subarus?
Bill Jepson

All the Subes I've worked on have cast iron guides. Not sure the guide material would make any difference in this respect. I've never seen a stuck valve on on a Sube either. Would be interesting to borescope a Sube with a few hundred hours of 100LL operation to see if lead deposits are building up on the stems.

100ll

I am not sure of the valve grind. I will have to check. Each time I take the heads off I find a layer of something on the piston and heads?? I don't know what it is, but I would think it would be clean after only 75hrs. I was guessing that maybe some lead gets under the valve seat and starts the whole problem. Don't know.

Steve

That's my guess. 4 valve OHC engines have very light seat pressures due to their low mass. Could be a contributing factor. I wanted to mention that I've now passed 300 hours on the Hobbs a couple weeks ago.

Ross, this is not a criticism but why run leaded fuel in an engine not designed for it? Also, why use oil not designed for leaded fuel? I'm very interested in why you're running into problems so early. Who can afford to rebuild a motor every 200 hours?

It's a pain to haul mogas to the airport and tank up and it's not available for cross country trips.

The oil change interval was only 25-30 hours. This with the addition of a lead scavenger in the fuel (Decalin) was supposed to keep the evil spirits away.

I'm only speculating what might be happening. A rebuild would be under $1000 total so I can afford to rebuild every 3-400 hours- not that I want to but it's taken about 5 years to get 300 hours on it. I'm going to keep flying it and doing compression and leakdown tests every 40 hours or so as long as the oil consumption does not get too high. I've switched over now to 80/20 mogas/ 100LL.

I'm looking forward to the fabled unleaded avgas that is being tested now.