Quantcast

Industrial engine electronic management system development - HBA style

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,355
Location
Rocky Mountains

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
4,076
Location
capital district NY
It seems some of the COP can be had for around 20 bucks apiece, for un-cowled engines it may be ok if adequately supported, it is for 4.3 Toyota so should be reliable?
Other smart coils seem to start around 60 bucks and up.
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,355
Location
Rocky Mountains
I have zero experience with generic COP but genuine Toyota COP should be cheap and easy to find. The engines they were attached to, in both the Toyota and Prism, died early deaths due to oil usage.

And as rvejguy noted: Who knows what is really inside the cheap stuff!?
 

spaschke

Well-Known Member
Joined
Oct 24, 2012
Messages
244
Location
Denver, CO
I had a 2002 passat with the similar set up. The cheap $20 COP would last a 1.5 - 2 years of daily driving. The $70 COP would last 1.5-3 years. Eventually they sold the cheap ones with a lifetime warranty for $35. After 2 or so years I never paid for another. I think the problem is engine heat. If they could be in a cooler place, they would last longer.
I would change them every year during inspection.
 

rv6ejguy

Well-Known Member
Joined
Jun 26, 2012
Messages
3,899
Location
Calgary, Alberta, Canada
Many OEM COP units had terrible reliability when they first came out. VW, Audi, Mazda, Ford being at the top of the list. The electronics better be robust when subjected to the levels of heat they see buried down inside a valve cover and these weren't, nor could they have been well tested by these companies.

Waste spark coil packs, mounted externally, are far more reliable in our experience which is one reason we use these instead of COP on all our systems- automotive and aviation.

While offshore coil and parts makers often say they meet or exceed the reliability of brands like Denso, Delphi etc., experience has shown that they simply don't follow the same attention to process control, use the same quality of raw materials, electronics or have the same levels of testing to validate the design. You get what you pay for here and I wouldn't use any of this junk on an aircraft.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
5,321
Location
US
If it proves practical to adapt an existing industrial engine EFI unit for aircraft use on these small engines (e.g. the Kohler or Briggs system using Delphi hardware and code), is there any appetite for a dirt-simple backup system? That would seem appealing if we can't determine and test all the interdependencies and failure modes in the built-for-another-purpose commercial EFI system.
A very simple backup system might be something like the Redneck EFI mentioned earlier in this thread. Or, it might be an open-loop system that uses just one sensor (MAF) and an in-cockpit mixture trim knob. No, it would not be especially fuel efficient, but it would be designed to keep the engine running, be cheap, be simple enough to be easily developed and tested.
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,355
Location
Rocky Mountains
A very simple backup system might be something like the Redneck EFI mentioned earlier in this thread.
This project didn't quite develop the interest I expected. I't on semi permanent hold pending getting my hands on a B+S EFI unit to dissect. This kind of all depends on what is inside the B+S Delphi EFI. If it can be hacked to make it aircraft rated then this project is likely dead. If not then a simple TBI style backup, as mentioned, may be practical.
 

jbiplane

Well-Known Member
Joined
Jan 29, 2014
Messages
245
Location
Krasnoyarsk, Russia

One good company from Spain give me their ECU for tests.
Small, but powerfull. I continue design my own ECU and making it completelly compatible with Speeduino to simplify support.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
5,321
Location
US
Interesting project based on ARM processor:
Student 8 cylinder ECU with parts for 64USD
https://skemman.is/bitstream/1946/29570/1/LokaskyrslaRikardur.pdf
Thanks, that is an interesting project. It is for a high RPM racing motorcycle, so we might make some different choices for an airplane (batch vs sequential injection, etc) and we'd go much farther to develop fault-tolerant modes, shielding/noise control, etc. He ran out of time to do the full testing (so he judged the effort "unsuccessful"), but I hope his paper was still well received by his thesis panel--the project seemed well thought out and his thesis did a good job of explaining what he'd done.
 

AdrianS

Well-Known Member
Joined
Jul 5, 2014
Messages
571
Location
Australia
While I like the concept, I am not keen on making it Arduino based. They are not designed for the heat, vibration or electrical noise you will see in an aircraft system.

I design the odd bit of electronics for automotive use : TI's guidelines* (attached) state
The governing specification for tail light electronics is that they be able to
withstand transients of +100 V and -300 V
And to assume any wiring may be shorted to power or ground by the installer...

We use NXP processors for our industrial control systems : 125C rated, built-in watchdog timer, CAN and SPI on-chip, etc.
However they are not as simple to start playing with as an Arduino, and require more skill/experience to program.
 

Attachments

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
8,801
Location
North Carolina
I'll be rooting around for some COP modules to use on my Briggs in the local pull-a-part junkyard. I did have some BMW ones. +12v, GND, signal. They have removeable extensions to allow them to plug onto plugs. Longer than I need, so maybe I can find some shorter ones, or make shorter extensions, or find some wires that will work. I do like the idea of using car coil/driver units. They have had a lot of testing and are easily available. If they located somewhere a bit cooler, they should last even longer. Junkyard ones are for initial test.
For the driver, I'm thinking a very simple advance system. A magnetic pickup sensing a sector. For low revs, it triggers the ignition on the trailing edge. It steadily advances up to the leading edge. We don't need a very complex system and the simpler it is, the simpler it is to debug and make reliable. Parts where someone else has already done the hard work don't count there ;)
My humming and hawing about carbs vs EFI was recently settled after I found a pair of $50 33mm CV carbs. that'll do for starters.
 

AdrianS

Well-Known Member
Joined
Jul 5, 2014
Messages
571
Location
Australia
For the driver, I'm thinking a very simple advance system. A magnetic pickup sensing a sector. For low revs, it triggers the ignition on the trailing edge. It steadily advances up to the leading edge. We don't need a very complex system and the simpler it is, the simpler it is to debug and make reliable.
My 1976 Ducati had that approach to electronic ignition. A 2 step trigger on the flywheel meant that at low RPM it triggered on the 2nd step (smaller air gap), then started triggering on the earlier small step at about 3000 RPM, advancing the spark timing by about 10-15 degrees or so.
That's probably all you need for aircraft use : retard for starting/idle, then full advance at working revs.
 

jbiplane

Well-Known Member
Joined
Jan 29, 2014
Messages
245
Location
Krasnoyarsk, Russia
Great news. Open source RUSEFI finally made standalone 4 injection + 4 ignition channels ECU.
Seems this is only ARM based real time ECU where you can test all hardware connecting to desctop PC :)
 

jbiplane

Well-Known Member
Joined
Jan 29, 2014
Messages
245
Location
Krasnoyarsk, Russia
I finally succeed with injection and ignition boards with functionality close to 3 time more expensive Megasquirt 2 Extra
1...4 cylinder for 150 USD
1...8 cylinders for 250 USD
At moment tested at -18...+80C on 4-stroke car engines
For aircraft use will use better connector or will make bunch of wires goes from ECU board
Some pins layout written on Russian, will make English docs soon
 
Last edited:
Top