|
-
POSTING RULES

-
Donate yearly (please).
-
Advertise in here!
-
Today's Posts
|
Insert Pics
|

07-21-2009, 08:16 AM
|
 |
|
|
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,766
|
|
This was discussed a year or so ago looking at NTSB data and various search strings. It appears you are quite a bit more likely to suffer a power loss event with an automotive conversion however it is hard to agree by what percentage.
From the same data, my conclusion was also that the core engines, at least in the case of Subarus were in fact more reliable than certified engines. Most events were triggered by supporting system failures- fuel, spark, electrical, cooling, lubrication, gearbox etc.
Bud Warren's previous Wheeler V8 used a carb and distributor type (non ECU) EI and worked well for about 500 hours.
Carb manifolds are available for the LS engines and frankly I'd feel better with that than an OE ECU which just plain scares me.
|

07-21-2009, 09:10 AM
|
 |
Senior Curmudgeon
|
|
Join Date: Sep 2005
Location: Dayton Airpark, NV A34
Posts: 15,420
|
|
Another option that I have only seen mentioned once before, is to throw away the EFI, and put on mechanical injection.
Airflow Performance makes units for auto engines, as well as aircraft engines, and the hardware seems to be identical other than the manifolds.
Or, if you really want to go EFI, there is this company called SDS that I suspect can help.
Ross, does your brain box speak Chevy??
__________________
Mike Starkey
VAF 909
Rv-10, N210LM.
Flying as of 12/4/2010
Phase 1 done, 2/4/2011 
Sold after 240+ wonderful hours of flight.
"Flying the airplane is more important than radioing your plight to a person on the ground incapable of understanding or doing anything about it."
|

07-21-2009, 10:23 AM
|
 |
|
|
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,766
|
|
We developed a system specifically for the LS engines mainly for desert racing and kit cars. Sales have been slow and a couple users have had odd issues with it. We have not accumulated enough hours in ground use to release the system for aviation use especially in light of the technical issues on this new board design. It uses a completely different board, microprocessor and software from the proven EM44/46 designs.
Many people have bugged us for the new system for V8 aviation apps including Geared but our policy is not to release anything for aviation until a good 10,000+ trouble free hours have been accumulated in automotive use first. To do otherwise is simply counterproductive to safety and our reputation. At present, there is no timetable for release to this market.
Mechanical injection would be another way to go or with something like Motec who has considerable experience with direct fire V8 apps.
Last edited by rv6ejguy : 07-21-2009 at 10:47 AM.
|

07-21-2009, 06:59 PM
|
|
|
|
Join Date: Jun 2009
Location: Clinton Township, MI
Posts: 4
|
|
ECMs in Light Airplanes...Not sure about that.
Hi, first post. I've been working in automotive powertrain controls for 9+ years now, first for an ECM supplier, Visteon, and now for GM Powertrain. My area is diagnostics calibration, so I've troubleshot just about every piece of an engine management system that can go wrong. Even in a simple system (engine only, no transmission), I wouldn't really trust flying in an ECU-equipped light plane with an automotive motor unless there was almost two of every key control component. Dual plugs, dual ECUs with fully independent power supplies, etc. Preferably dual engines...
Automotive ECUs are very durable, designed to operate -40 F to 150F, for 150000 miles. But the airplane 'drive cycle' is so different than a car, there's just too many unknown failure modes out there that can bite you. Temperature gradients for one, sustained high output from the output drivers on the circuit board, power supply fluctuations, etc. One obvious area is altitude compensation. Most automotive baro tables are limited to 12,000 feet, which is basically the highest road in the world. Anything above that is all error buildup.
The most aggravating issues to solve are the one-off cases that come out of nowhere; where it took a perfect alignment of multiple factors to reveal the issue. The thing is, no matter how robust the calibration and validation is, there's always one fault mode out there that doesn't show itself during development, no matter how many hours on the dyno, test vehicle miles, or development trip hours you have.
I'm glad to see yet another application of the venerable small block Chevy, though.
Rob Weiss
|

07-21-2009, 08:36 PM
|
|
|
|
Join Date: Aug 2007
Location: Granbury, TX
Posts: 103
|
|
RV-10 Gear Collapse
Read the string on the alternative engine forum. A couple things are clear and reinforce my beliefs on the auto engine vs Lycoming debate. First, auto engine converts are passionate about their endeavors. Second, they tend to put credulity aside (see for example the ludicrous speed and fuel flow numbers quoted for the accident airplane - and defended).
I'm going to get flamed for that last comment which I may deserve.
As the previous post notes, most of the failures come from the accessories. Regardless, the prop stops nevertheless. I have to tip my hat to the auto engine afficianados because they are true experimenters.
|

07-21-2009, 09:49 PM
|
 |
|
|
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,766
|
|
Quote:
Originally Posted by Rob Weiss
Hi, first post. I've been working in automotive powertrain controls for 9+ years now, first for an ECM supplier, Visteon, and now for GM Powertrain. My area is diagnostics calibration, so I've troubleshot just about every piece of an engine management system that can go wrong. Even in a simple system (engine only, no transmission), I wouldn't really trust flying in an ECU-equipped light plane with an automotive motor unless there was almost two of every key control component. Dual plugs, dual ECUs with fully independent power supplies, etc. Preferably dual engines...
Automotive ECUs are very durable, designed to operate -40 F to 150F, for 150000 miles. But the airplane 'drive cycle' is so different than a car, there's just too many unknown failure modes out there that can bite you. Temperature gradients for one, sustained high output from the output drivers on the circuit board, power supply fluctuations, etc. One obvious area is altitude compensation. Most automotive baro tables are limited to 12,000 feet, which is basically the highest road in the world. Anything above that is all error buildup.
The most aggravating issues to solve are the one-off cases that come out of nowhere; where it took a perfect alignment of multiple factors to reveal the issue. The thing is, no matter how robust the calibration and validation is, there's always one fault mode out there that doesn't show itself during development, no matter how many hours on the dyno, test vehicle miles, or development trip hours you have.
I'm glad to see yet another application of the venerable small block Chevy, though.
Rob Weiss
|
Nice post Rob. yes, I quite agree. As the complexity of control goes up, the chances of an IF, AND, AND, AND, THEN problem can surface. These are so much harder to see for the non OEM tuner. A simple IF coolant temp exceeds 235F THEN shut off injector outputs might be easy to see and alter. The problem is when several seemingly unrelated things come together to trigger the same response- well then Houston, we have a problem...
With most aftermarket ECUs, we don't need all the diagnostics and complexity. There are far less IFs and ANDs. The mission is far more simple- proper spark timing and mixtures to make the engine run well is job one and maybe the only job in fact. No chassis, emissions, self learning and minimal diagnostics tie-ins. Sometimes we don't even use hard code to give the response we want, we just alter a value in the user programmable lookup tables.
As an example for a MAP sensor failure, the lowest and highest values in the table (where the engine never actually operates) we enter a value equivalent to WOT at sea level. If the MAP sensor fails either open or shorted (usually one of these two), fueling reverts to near full power so by moving the throttle open, the engine will continue to run and produce good power despite the dead sensor. We use a board resistor to pull the failed input high even these days. The same goes for temperature sensors where open or shorted failures simply defaults fueling to a warm engine on a summer day. In these cases, there is no code operating in between.
|

07-21-2009, 10:05 PM
|
 |
|
|
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,766
|
|
Quote:
Originally Posted by Nukeflyboy
Read the string on the alternative engine forum. A couple things are clear and reinforce my beliefs on the auto engine vs Lycoming debate. First, auto engine converts are passionate about their endeavors. Second, they tend to put credulity aside (see for example the ludicrous speed and fuel flow numbers quoted for the accident airplane - and defended).
I'm going to get flamed for that last comment which I may deserve.
As the previous post notes, most of the failures come from the accessories. Regardless, the prop stops nevertheless. I have to tip my hat to the auto engine afficianados because they are true experimenters.
|
I think most auto engine experimenters (maybe not vendors though!) want to post accurate information about weight, fuel flow and performance. As many things are new, often fuel flow sensors and black boxes are not calibrated. I always encourage people to read up and use the best test methods they can to get a true picture of overall engine/ airframe performance.
Using the right/ left tank method and filling before and after the flight using one for taxi, takeoff, climb and the other for a timed cruise at a constant power setting combined with 4 way GPS data is pretty simple and accurate. Randy Crothers eventually used this method to confirm the ECU FF data. It is very hard to dispute results using this method after a few longer flights and getting similar results. Sometimes we are surprised, sometimes disappointed but the facts are the facts either way.
My mind is still open that the LS engines might produce similar or better FF vs. TAS figures compared to a Lycoming. Once Todd or Bill does a few runs as above, we will know one way or the other.
You are quite right in saying that certain accessory failures can often cause the prop to stop just as surely as an ECU issue so these are just as important as the engine itself. Careful design and testing must be carried out on the entire system.
Last edited by rv6ejguy : 07-21-2009 at 10:10 PM.
|

07-21-2009, 10:51 PM
|
|
|
|
Join Date: Aug 2007
Location: S21, Oregon
Posts: 161
|
|
Interesting.....
All the talk about the engine in this thread instead of the other one. I'm more curious about the weight of the plane at the time of the accident, and the approximate rate of descent that caused the gear to fail 
__________________
Kelly
RV-7 empennage done, wings done, fuselage to QB stage.
1973 Maule M4-220C flying
|

07-22-2009, 06:09 AM
|
|
|
|
Join Date: Jan 2007
Location: Conroe, Texas
Posts: 517
|
|
No dog in the Hunt...........
I'll never forget my aunt Alice, she was a very devote Baptist. Back in the early 60's when space exploration was really getting started, she made the following comment that stuck with me for life. I have used it to reflect & evaluate human opinion of future events.
Her comment was as follows: "All this foolishness on space exploration, If Man ever walks on the Moon, the Moon will turn to Blood."
Well, it didn't.
Maybe not today, maybe not tomorrow, but "This flag will rise!"
Bud / Bill stay after it. If we were all " Aunt Alice's", we would still be barking at the moon. We as "experimental builders" owe a lot to your trials & tribulations.
and Aunt Alice, God rest your soul, I loved you dearly.
__________________
Chuck Elsey
RV6 Start 7/06- Flying!
 N349CE
|

07-22-2009, 07:33 AM
|
 |
|
|
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,766
|
|
Quote:
Originally Posted by 1:1 Scale
All the talk about the engine in this thread instead of the other one. I'm more curious about the weight of the plane at the time of the accident, and the approximate rate of descent that caused the gear to fail 
|
In my forced landing a few years back on my 6A, the main gear splayed badly but not as much as on Bill's -10 here. The telltale on the G meter was pegged at 7 Gs. We were about 100 pounds under gross at the time.
Believe me, elevator authority in the flare, deadstick, below 65 knots with a forward C of G and full flaps... well there just isn't any. 
Hence my warning about using flaps on RVs in a power loss situation. Think of it this way, full flaps, drag is high, nose is way down, you let speed slip down to 65 knots, descent rate around 25 fps. At 100 feet, the ground is rushing up at you at an alarming rate and you will be on the ground in 4 seconds. In that last 4 seconds, the instinctive thing is to actually pull back to arrest the scary descent rate but there is no energy left and this will actually increase descent rate or make you stall. You are out of options at this point and are going to hit hard. This has hurt and killed a number of RV pilots faced with a power loss.
Keep it clean and keep your 80-85 knots right down to your flare.
Last edited by rv6ejguy : 07-22-2009 at 07:36 AM.
|
| Thread Tools |
Search this Thread |
|
|
|
| Display Modes |
Linear Mode
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -6. The time now is 02:34 AM.
|