VansAirForceForums  
Home > VansAirForceForums

- POSTING RULES
- Donate yearly (please).
- Advertise in here!

- Today's Posts | Insert Pics


Go Back   VAF Forums > Regional Forums > Northwest
Register FAQ Members List Calendar Search Today's Posts Mark Forums Read

Reply
 
Thread Tools Search this Thread Display Modes
  #11  
Old 12-12-2019, 06:19 AM
rv6ejguy's Avatar
rv6ejguy rv6ejguy is offline
 
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,546
Default

Quote:
Originally Posted by keitht View Post
Ross,
It wasn't my intent to get into a fight about EFI or any other technology that we have the choice to put on an experimentally certified airplane or not.
When comparing hydromechanical with EFI it should be pretty obvious when looking at the number of parts in a hydromechanical system, their failure rates and failure modes and comparing that with the number of electronic components, mechanical components, solder joints, software and dependence on dual independent electrical power that there is a need for a alternate channel with a reliable switch over for the EFI. There is a lack of published data comparing predicted failure rates with actual demonstrated failure rates for either hydromechanical or EFI. If you have data comparing predicted failure rates with actual failure rates and the same for failure modes and effects and are prepared to share that data that would be an industry first in the experimental world. The experimental hydromechanical FI systems are very similar to the certified hydromechanical FI so it would seem reasonable to expect similar failure modes and availability with no need for a second alternate channel.
As with the discussion on alternators if the volume of product is sufficient then the very high cost of doing the design engineering, analysis, development, testing and continuous improvement and support to meet predicted performance is justified and necessary to compete and stay in business. If DO178 software and DO254 hardware standards are not being used then the question is "what standards are being used and how do they match up".
Pmag have a following just as your remote mounted EI system does. I am not going to discuss the process I used to determine that using dual Pmags would work for me other than to say it satisfied my concerns about having a total loss of engine power and bounded the risk with numbers I could live with.
As far as experience with EFI for piston engine aircraft the answer is no I don't. My experience is with gas turbine engines for turboprop and helicopters and fly by wire systems for commercial and military aircraft. When I see an EFI for piston engines that is certified, affordable and cloned in the experimental world then maybe I will take another look at doing a retrofit to reap the benefits.

KT
Sorry, it does seem like your intent was to cast aspersions on EFI with no admitted real world experience with them. The OP asked people to comment on their experiences and you have stated you have none- only opinions- which you said yourself, don't count for much.

Now you conclude here there is a need to have a backup system with EFI but no need on mechanical systems. This was why I offered that we have far more single systems flying than dual systems and in 500,000 hours they have demonstrated very good reliability.

I HAVE published a list of known failures on SDS to date a couple years back here should you choose to use the search button.

DO178 and 254 standards are guidelines which may help electronics be more reliable. They don't guarantee reliability. Reliability is only proven in field use as we have done over 25 years in business now. Theory is great but only practice proves it.

It doesn't surprise me that you're not willing to share your analysis of your EI choice here. Others have to share data to be validated but you don't? Makes sense to me...

You probably won't see "affordable" certified, EFI migrating over to Experimentals. Lycoming's IE2 probably satisfies your standards but isn't affordable. Certification takes tremendous time and money to satisfy FAA standards as Lycoming IE2 engineers told me at Reno years back when they were developing it.

Aerosance also had a certified EFI system but had issues despite being certified. As I've said before, certification is no guarantee of reliability.
__________________

Ross Farnham, Calgary, Alberta
Turbo Subaru EJ22, SDS EFI, Marcotte M-300, IVO, Shorai- RV6A C-GVZX flying from CYBW since 2003- 436.9 hrs. on the Hobbs,
RV10 95% built- Sold 2016
http://www.sdsefi.com/aircraft.html
http://sdsefi.com/cpi2.htm



Last edited by rv6ejguy : 12-16-2019 at 09:01 AM.
Reply With Quote
  #12  
Old 12-12-2019, 09:57 AM
Toobuilder's Avatar
Toobuilder Toobuilder is offline
 
Join Date: Jul 2009
Location: Mojave
Posts: 4,471
Default

Quote:
Originally Posted by keitht View Post
...When comparing hydromechanical with EFI it should be pretty obvious when looking at the number of parts in a hydromechanical system, their failure rates and failure modes and comparing that with the number of electronic components, mechanical components, solder joints, software and dependence on dual independent electrical power that there is a need for a alternate channel with a reliable switch over for the EFI. There is a lack of published data comparing predicted failure rates with actual demonstrated failure rates for either hydromechanical or EFI. If you have data comparing predicted failure rates with actual failure rates and the same for failure modes and effects and are prepared to share that data that would be an industry first in the experimental world. The experimental hydromechanical FI systems are very similar to the certified hydromechanical FI so it would seem reasonable to expect similar failure modes and availability with no need for a second alternate channel.
As with the discussion on alternators if the volume of product is sufficient then the very high cost of doing the design engineering, analysis, development, testing and continuous improvement and support to meet predicted performance is justified and necessary to compete and stay in business. If DO178 software and DO254 hardware standards are not being used then the question is "what standards are being used and how do they match up".
Pmag have a following just as your remote mounted EI system does. I am not going to discuss the process I used to determine that using dual Pmags would work for me other than to say it satisfied my concerns about having a total loss of engine power and bounded the risk with numbers I could live with.
As far as experience with EFI for piston engine aircraft the answer is no I don't. My experience is with gas turbine engines for turboprop and helicopters and fly by wire systems for commercial and military aircraft. When I see an EFI for piston engines that is certified, affordable and cloned in the experimental world then maybe I will take another look at doing a retrofit to reap the benefits.

KT
It strikes me that you are using theoretical data to support a preconceived bias. Your comparison between the relative complexity of hydromechanical and electronic systems for example seems "obvious" to you, yet I draw the exact opposite conclusion. True, we don't have a lot of light aircraft history to draw from, but we certainly have plenty in the automotive industry. And in practice, cars have become orders of magnitude more reliable with the advent of electronic controls. That's a tough fact to ignore.

It's also interesting that despite your analysis, you picked the EI with by far the most spotty service record of any available. The one that has seen multiple catastrophic failure modes, including ones where the failed ignition continues to fire the plugs, but will do so at radically inappropriate timing. And after multiple hardware redesigns and software versions trying to beat back the demons you still have to remove and inspect the unit far more frequently than the 100 year old magneto.

This thread is about experience with EFI. Im looking forward to contributions to this knowledge pool as an EFI user myself. I don't feel qualified to contribute just yet as I have less than 5 hours logged, but the system was flawless as well as a game changer.

I'd have a bunch more time logged, but the "reliable" hydromechanical prop governor failed and the engine and prop are now junk.
__________________
WARNING! Incorrect design and/or fabrication of aircraft and/or components may result in injury or death. Information presented in this post is based on my own experience - Reader has sole responsibility for determining accuracy or suitability for use.

Michael Robinson
______________
Harmon Rocket II -SDS EFI - Flying
RV-8 - Flying
1940 Taylorcraft BL-65 -flying
1984 L39C - flyable, available for sale
Reply With Quote
  #13  
Old 12-12-2019, 11:02 AM
rv6ejguy's Avatar
rv6ejguy rv6ejguy is offline
 
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,546
Default Stats

For transparency and clarity, I dug up my post from 10/11/17 in another EI reliability thread to save people the trouble of searching:

"Well, I'll break the mold here and publish our stats if you care to believe them. I published this (with some new additions) on another aircraft forum a while back. Note, most of our systems for aircraft, outside of military usage, are integrated fuel and ignition systems so more complex than ignition-only products. I've presented the whole history here to the best of our records and my recollections. If you're a direct SDS customer, feel free to ad any comments, negative or positive.

The stats on our electronics, compiled over the last 20+ years (reported to us or observed personally, attributed to actual electronic component failure):

550,000 + flight hours on over 1850 systems
High time ECU- 145,000 hours (bench test one)
High time fliers of our systems- 1500 hours (Rover V8), 1700 hrs (Lycoming O-360 and Subaru), 2 at 2000+ hours (Rotax 912 and Jabiru used for flight training)
High time automotive system (EM-1, circa 1994 used in 5 different cars- estimated 12,000+ hours).
ECU failures on aircraft- zero
Crank sensor failures- zero
EJ25 coil pack failures- zero
Fuel pump failures- zero (2 failed when improperly mounted)
Wiring harness failures- zero
Injector failures- 2 unconfirmed (claimed but never sent back for our inspection)
Temp sensor failures- 2 in aircraft (non-critical system failure and new fix applied 5 years ago)
MAP sensor failures- 5 in aircraft (non-critical failure, cause on 4 found to be improper mounting with vacuum port facing up- moisture ingress and these sensors were 3rd party supplied as well)
TPS failures- 2 in aircraft (1 over 20 years old, moisture ingress. Sealed design used for last 10 years)

Incidents/ Accidents

We've heard of 1 forced landing which was the result of poor wiring by the installer with an injector wire pinched to ground. Same customer, some time before had intermittent engine running, traced to a very scary wiring connection to injector power.

One V6 where the user installed an aluminum exhaust system which collapsed under the heat load and melted through the Hall sensor cable. Forced landed in a tomato field. Afterwards, we supplied with fire sleeve over the cables and now have Tefzel cables.

One plane limped around the pattern after takeoff on 3 of 4 cylinders with an apparently dead injector (low EGT and CHT indication). Landed safely. Injector was checked extensively, worked fine in testing. ECU changed out and old one sent back. No fault found in old one and it has been running perfectly in our shop car for nearly 4 months now. We suspect a wiring issue in the aircraft as no fault was found with the ECU or other components. I should mention that this aircraft has a brand X ignition system on it, not ours. This one remains a mystery.

One fatal accident locally here where I helped the TSB find that the installer had mis-wired circuits to the wrong breaker value. Breaker tripped at about 300 feet AGL, pilot tried to land back on a crossing runway, stall/spin.

We've had reports of three, 3rd party ignition drivers attached to our systems which failed from poor/cold solder joints, fortunately all on the ground. Lesson: don't attach 3rd party devices to SDS. We sent new parts to people so they could remove the defective 3rd party components and continue flying safely.

At least a dozen reports of rough running/engine shutdowns in flight. Most of these eventually linked to wrong plug wires or non- resistor plugs used or wired contrary to our recommendations. Bad grounds are the #1 cause of odd running issues. Some ECUs have been found wet inside or had flood damage at some point. Third party parts added to our electronics in several cases. All aircraft down safely although probably some soiled underwear in a couple cases.

Several ECUs sent back by customers saying they didn't work. Bench tested, installed on test engine here. Worked fine, sent back, customers later found issues with their wiring or installation.

Of course, we've had a number of people screw up wiring on installation and fry things, usually when they cut into the harness and make their own connections. Others have programmed things inappropriately, mounted magnets wrong, fuel pumps with reverse polarity, forgot power and ground connections etc.

We've had some vexing, weird issues on occasion but have eventually solved almost all of them for our customers. Almost always wiring installations.

I can think of two software screwups on release of EM-5 6 cylinder systems. Discovered prior to any engines being run in one case and a rough running issue which was not a safety concern, new boards shipped to our commercial customer. New hardware/ software sent out to 4 affected customers immediately at our expense in the second case.

We supplied some components to a vendor about 10 years ago which, through insufficient testing and rushing to meet a deadline, were causing hot running coils COP. Lesson learned, test, test, test and the design is only ready when it's ready.

Zero CPI failures in aircraft or automotive usage to date.

The bottom line, based on all this experience, is that you're unlikely to suffer a critical electronic failure with our products in your flying career if not modified in some way and are properly installed and operated as we recommend.

Of course, not all EFI/EI brands have our record of reliability and some are really quite poor from reports we get from their ex-clients who are replacing other brands with our stuff.

We encourage everyone to report their problems so we can see if there is a pattern and develop a fix. I have no problem with people reporting their problems here. We won't cut off support to you.

If you've had an issue/ failure with SDS that we haven't resolved, contact me directly or post it here. We don't have anything to hide.

We're not perfect but we try to fix anything wrong and make it right. We build on evolutionary improvements which we try to test thoroughly before product release."

************************************************** ******

I'll add some more data we have now from the last 2 years:

I've seen one actual Walbro fuel pump failure now. I got the pump back from the customer and dissected it. The motor on it ran normally but it did not pump anything. Despite taking it down to the last part, I could find no reason why it wasn't working.

The one mystery event above with the engine running on 3 of 4 cylinders happened again. This aircraft had a third party ignition system on it, not SDS. This supplied our ECUs with the RPM signal instead of using our usual Hall Effect sensor. Changed over to our sensor and the problem has never re-occurred a couple of hundred flight hours down the road.

One more fatal accident in 2018 with a plane having SDS EFI on it. After 10 years trouble free flying with the previous owner. The new owner, a very low time pilot, took off after 2 hours of dual on type with questionable weight and C of G. Stalled and spun in on his first solo flight. Engine running at impact according to the accident report.

One customer fried 2 ECUs in quick succession this year, made it down safely on the backup ECU in each case. After 5 months of investigation, photos revealed the cause- installer had tie wrapped the air temp sensor cables to the spark plug wires! Rerouted, problems gone and flying well now.
__________________

Ross Farnham, Calgary, Alberta
Turbo Subaru EJ22, SDS EFI, Marcotte M-300, IVO, Shorai- RV6A C-GVZX flying from CYBW since 2003- 436.9 hrs. on the Hobbs,
RV10 95% built- Sold 2016
http://www.sdsefi.com/aircraft.html
http://sdsefi.com/cpi2.htm


Reply With Quote
  #14  
Old 12-13-2019, 12:08 AM
keitht keitht is offline
 
Join Date: Jun 2016
Location: coupeville wa
Posts: 103
Default

Quote:
Originally Posted by Toobuilder View Post
It strikes me that you are using theoretical data to support a preconceived bias. Your comparison between the relative complexity of hydromechanical and electronic systems for example seems "obvious" to you, yet I draw the exact opposite conclusion. True, we don't have a lot of light aircraft history to draw from, but we certainly have plenty in the automotive industry. And in practice, cars have become orders of magnitude more reliable with the advent of electronic controls. That's a tough fact to ignore.

It's also interesting that despite your analysis, you picked the EI with by far the most spotty service record of any available. The one that has seen multiple catastrophic failure modes, including ones where the failed ignition continues to fire the plugs, but will do so at radically inappropriate timing. And after multiple hardware redesigns and software versions trying to beat back the demons you still have to remove and inspect the unit far more frequently than the 100 year old magneto.

This thread is about experience with EFI. Im looking forward to contributions to this knowledge pool as an EFI user myself. I don't feel qualified to contribute just yet as I have less than 5 hours logged, but the system was flawless as well as a game changer.

I'd have a bunch more time logged, but the "reliable" hydromechanical prop governor failed and the engine and prop are now junk.
I am stunned at your post. Just consider the thousands of fuel injected certified piston driven aircraft flying around the world with single string hydromechanical fuel servos. I repeat- single string - no redundancy- no backup. What is it that you dont think that says about reliability and availability. It is true that there is a higher level of routine preventative maintenance with a hydromechanical system compared to an EFI but the overall predicted failure rate of the EFI is higher. If each path of the EFI has a failure rate of something10E-4 per hour then a dual channel system will have an overall failure rate of twice something 10E-4 with the probability of loss of function in the range of 10E-7 per hour - comparable to the demonstrated failure rate of a hydromechanical system for loss of function.
Lets not confuse the system and functional requirements for automobiles. Single point catastrophic loss of function on a first failure are an acceptable failure mode for the automobile fuel system and ignition system and the system monitoring features are all focussed on diagnostics and fault isolation and not preservation of function or redundancy management. Indeed automobile electrical and electronic systems have become very reliable due to better components, better packaging, better design support tools for electrical, thermal and mechanical analysis and derating. Not least is automated manufacturing tools that allow 6 sigma variability to become a reality. Lets not fool ourselves into thinking what is OK for an automobile is also OK for aviation. The design requirements are fundamentally different in terms of failure modes and effects and the consequences of failures. There is a lot that can be taken from high end automotive electronics and applied to aviation but the fundamental system engineering requirements are not part of that set. BTW I am truly sorry to hear that the failure of the prop govenor resulted in destruction of your engine and propeller - truly a mind numbing experience.

KT
Reply With Quote
  #15  
Old 12-13-2019, 01:43 AM
McStevens McStevens is offline
 
Join Date: Sep 2019
Location: Edmonds, WA
Posts: 73
Default

I too am stunned at Ross's post about the fleet history. That's really quite impressive. Nice to see that level of transparency too.
Reply With Quote
  #16  
Old 12-13-2019, 07:31 AM
rv6ejguy's Avatar
rv6ejguy rv6ejguy is offline
 
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,546
Default

Quote:
Originally Posted by keitht View Post
I am stunned at your post. Just consider the thousands of fuel injected certified piston driven aircraft flying around the world with single string hydromechanical fuel servos. I repeat- single string - no redundancy- no backup. What is it that you dont think that says about reliability and availability. It is true that there is a higher level of routine preventative maintenance with a hydromechanical system compared to an EFI but the overall predicted failure rate of the EFI is higher. If each path of the EFI has a failure rate of something10E-4 per hour then a dual channel system will have an overall failure rate of twice something 10E-4 with the probability of loss of function in the range of 10E-7 per hour - comparable to the demonstrated failure rate of a hydromechanical system for loss of function.
Lets not confuse the system and functional requirements for automobiles. Single point catastrophic loss of function on a first failure are an acceptable failure mode for the automobile fuel system and ignition system and the system monitoring features are all focussed on diagnostics and fault isolation and not preservation of function or redundancy management. Indeed automobile electrical and electronic systems have become very reliable due to better components, better packaging, better design support tools for electrical, thermal and mechanical analysis and derating. Not least is automated manufacturing tools that allow 6 sigma variability to become a reality. Lets not fool ourselves into thinking what is OK for an automobile is also OK for aviation. The design requirements are fundamentally different in terms of failure modes and effects and the consequences of failures. There is a lot that can be taken from high end automotive electronics and applied to aviation but the fundamental system engineering requirements are not part of that set. BTW I am truly sorry to hear that the failure of the prop govenor resulted in destruction of your engine and propeller - truly a mind numbing experience.

KT
Keith, where is your real-world data to back up your theoretical calcs and claims here? How do you calculate failure rates without data?

It seems you think chips, resistors and caps are falling off the boards with regularity. They clearly aren't, otherwise no ECUs would run 100 hours before failure, let alone 145,000 as in one case. Any servos or mags run that long with no maintenance? You seem to think that a static electronic part is intrinsically less reliable than spinning, rubbing, meshing mechanical parts.

While the result of an ECU failure in a car is far less likely to cause a safety issue, the auto OEMs can't afford ECU failures as a result of systemic design problems, it could cost them hundreds of millions because they can build millions of that one model. Just consider the hundreds of millions of EFI cars running around today. Do you see them littering the roads from ECU failures?

I came from the automotive repair world and worked on carbs and points when I was a young buck. EFI/EI changed that world completely and everyone except you it seems, would concede that EFI/EI was the single most important development in improving automotive durability and reliability. Automotive EFI/EI doesn't need a backup primarily because this stuff is so reliable these days. Far more reliable than a Lycoming engine as far as MTBF.

Nothing theoretical, we've PROVEN over 24 years and 700,000 flight hours that our automotive type ECU designs can be applied to aircraft and be very reliable and even offer redundant ECUs in the event that the primary ECU fails if that is still a worry. I've discussed here in other threads how we test and validate our designs. We do a lot of thermal testing and running ignition and fuel drivers way past spec for hours on end. We never use marginal components which could cause heat related failures. We won't try to save money if it will have an effect on durability. Our track record in the real world seems to support that we're doing something right.

Mechanical systems like servos, carbs and mags have failures, we've seen lots of them posted here on VAF and from experiences of my flying friends and customers.

Do you think maybe we could have learned how to build reliable electronics after 25 years and 10,000 controllers delivered?

Your bias is overwhelming here. You have no experience and no hard data in this field yet you feel we should listen to you (even though you posted a quote that opinions without evidence should be ignored). That doesn't make any sense. It would be like me posting advice on how to build a composite airframe. I know squat about that topic since I have no experience in the field. That would just be ridiculous.
__________________

Ross Farnham, Calgary, Alberta
Turbo Subaru EJ22, SDS EFI, Marcotte M-300, IVO, Shorai- RV6A C-GVZX flying from CYBW since 2003- 436.9 hrs. on the Hobbs,
RV10 95% built- Sold 2016
http://www.sdsefi.com/aircraft.html
http://sdsefi.com/cpi2.htm



Last edited by rv6ejguy : 12-13-2019 at 06:53 PM.
Reply With Quote
  #17  
Old 12-13-2019, 10:23 PM
johnbright's Avatar
johnbright johnbright is offline
 
Join Date: Aug 2010
Location: Newport News, Va
Posts: 286
Default

Quote:
Originally Posted by Toobuilder View Post
... I'd have a bunch more time logged, but the "reliable" hydromechanical prop governor failed and the engine and prop are now junk.
Would EFI rev limit feature have prevented damage to the engine and prop?
__________________
John Bright, RV-6A 25088, at FWF
O-360, 8.5:1, vert sump, dual SDSEFI EM-5-F
Schematic, Dual battery, dual alternator
Construction Photos
Newport News, Va
Reply With Quote
  #18  
Old 12-14-2019, 12:05 AM
Mike S's Avatar
Mike S Mike S is offline
Senior Curmudgeon
 
Join Date: Sep 2005
Location: Dayton Airpark, NV A34
Posts: 14,927
Default

Quote:
Originally Posted by johnbright View Post
Would EFI rev limit feature have prevented damage to the engine and prop?
Probably not------it is my understanding that the speed he was going, the flat prop drove the engine to the excessive RPMs.
__________________
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."
Reply With Quote
  #19  
Old 12-14-2019, 12:16 AM
Toobuilder's Avatar
Toobuilder Toobuilder is offline
 
Join Date: Jul 2009
Location: Mojave
Posts: 4,471
Default

Quote:
Originally Posted by Mike S View Post
Probably not------it is my understanding that the speed he was going, the flat prop drove the engine to the excessive RPMs.
Yes, the rev limiter was programmed, but if you jam your car into 1st gear at freeway speed and drop the clutch, the engine RPM is going to the Moon even with key "OFF".
__________________
WARNING! Incorrect design and/or fabrication of aircraft and/or components may result in injury or death. Information presented in this post is based on my own experience - Reader has sole responsibility for determining accuracy or suitability for use.

Michael Robinson
______________
Harmon Rocket II -SDS EFI - Flying
RV-8 - Flying
1940 Taylorcraft BL-65 -flying
1984 L39C - flyable, available for sale

Last edited by Toobuilder : 12-14-2019 at 12:20 AM.
Reply With Quote
  #20  
Old 12-14-2019, 09:05 AM
rv6ejguy's Avatar
rv6ejguy rv6ejguy is offline
 
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,546
Default

Correct, the rev limiter may be of use with a completely lost prop or a partial governor failure as we've seen a couple times at Reno. Won't do much in this case.
__________________

Ross Farnham, Calgary, Alberta
Turbo Subaru EJ22, SDS EFI, Marcotte M-300, IVO, Shorai- RV6A C-GVZX flying from CYBW since 2003- 436.9 hrs. on the Hobbs,
RV10 95% built- Sold 2016
http://www.sdsefi.com/aircraft.html
http://sdsefi.com/cpi2.htm


Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -6. The time now is 04:43 PM.


The VAFForums come to you courtesy Delta Romeo, LLC. By viewing and participating in them you agree to build your plane using standardized methods and practices and to fly it safely and in accordance with the laws governing the country you are located in.