Thanks for the compliment, although its not difficult to do better than Ford
(some of their stuff (eg throttle pots.) leaves a lot to be desired.
Similarly, a professionally built custom loom is going to be a lot better
than a proportion of homebuilders would have the patience or experience to
put together. With most auto manufactures (including BMW) there is a
delicate balance between quality and cost. As such its fairly
straightforward to improve upon the reliability prospects of the OEM wiring
by means of,for example, better architecture (the custom loom needs only to
be routed for the EFI and not a miriad of other functions), better
connectors (putting all earth wires onto a single flat ring terminal is
cheap for the OEM but not ideal), improved strain relief and support,
aircraft spec. wire, etc.
How such a custom loom is installed and checked in accordance with
instructions is a matter for the local regulatory/inspection system (eg PFA)
to police. It is however a fact that most 2-stroke failures experienced in
the UK were in the early years largely a consequence of poor installation
rather than, necessarily, a flaw in their design.
Following recent events, Rotax might feel they are not up to the challenge;
maybe their problems have been a consequence of to much "car stuff".
Carbs. and dual ignition might represent a comfortable alternative for
technophobes! But carbs aren`t without their problems and figure
significantly in the accident ratings for the effects of icing alone.
Similarly, an EFI system can be backed up with a simple "tap and injector",
but I remain to be convinced that the additional complexity is worth the
trouble.
Apologies to the forum if this thread is becoming overworked.
Duncan McFadyean
-----Original Message-----
From: Wilksch Airmotive <mark@wilksch.com>
<milesm@avnet.co.uk>; europa@avnet.co.uk <europa@avnet.co.uk>
Date: Thursday, August 26, 1999 10:29 PM
Subject: The great fuel injection debate - round 3
>Dear Duncan
>
>These figures are most helpful - brings the whole scene into perspective.
>
>I think that the 1:500000 sounds about right for a whole system integrated
>by one manufacturer eg. a Ford engine in a Ford car with a wiring loom by
>Ford. With your knowledge and the TLC you will lavish on your aircraft
>would will probably do better.
>
>Now consider a Company who wants to put the engine into production and send
>it to all corners of the world to the kit plane market where the generic
>engine will be installed on dozens of different airframes by all sorts of
>different builders who all think that they know best and probably want to
>customise looms etc. Even with well documented installation guidelines etc
>I think that the likely outcome will be failure rates 10 or 100 times
higher
>than your the 1:500000 that we might agree is applicable to
(non-duplicated)
>car systems. To stay out of court the manufacturer will either need
control
>of the whole system eg a Toyota engine/plane (we almost saw one) or will
>want do build in mechanical back-up or electrical redundancy channels. The
>cost of this compared to carbs is going to be much higher. .......it
won't
>be a business proposition for the likes of Rotax. Don't forget, carbs are
>very simple cheap devices which work rather well.
>
>I think that the first mechanical gasoline injection system to hit cost
>targets will win. Anyone else agree with me?
>
>Regards............MARK WILKSCH
>
>
>PS: Thanks for your stimulating response.
>
>
>-----Original Message-----
From: ami mcfadyean <ami@mcfadyean.freeserve.co.uk>
><milesm@avnet.co.uk>; europa@avnet.co.uk <europa@avnet.co.uk>
>Date: Thursday, August 26, 1999 09:48
>Subject: Re: Flying: Fuel consumption in 'Classic' with Rotax
>912.
>
>
>>NTSB Figures:
>>Over a 5-year period in the early 90`s there were 93 accidents in the US
>>attributable to magneto failure. Annual piston airtime in the States is
>>about 20,000,000 hours per year. Some simple sums tells you that the
>failure
>>rate of what must be mostly twin mag. systems is 1 in 1,000,000 (the same
>as
>>your figure for dropping dead at the controls!). This doesn`t include all
>>the other probable failures which didn`t end with an accident.
>>
>>Bosch Figures (published in an SAE paper):
>>Beyond the stage where infant mortality is at a high rate, the failure
rate
>>of the electronics alone is about 1in 5,000,000 hours. Electronics win.
>>However, (also from Bosch figures) 90% of all SYSTEM failures are caused
by
>>something other than the electronics (eg wiring (mostly),
>>transducers,actuators etc). On that basis, whole-system failure rate comes
>>down to 1 in 500,000. Magnetos win.
>>These figures probably include a large number of units which were killed
by
>>"maintenance" (ie non-operational failures) stated by the aftermarket
>>companies as being the largest cause of returns on electronic units.
>>
>>The risk we face in the UK of dying in a GA accident is 1 in 150,000hours
>>flown (based on what I surmise from GASIL figures; someone please correct
>me
>>if they know different). So, that level of risk is presumably acceptable
to
>>all of us who choose to go flying, although CAA are always making efforts
>to
>>reduce this (pun intended). 1 in 500,000 would therefore seem to be of an
>>acceptable order of magnitude, especially if one is not guaranteed to die
>>during such an occurrence.
>>
>>Duncan McFadyean
>>
>>-----Original Message-----in the
>From: Wilksch Airmotive <mark@wilksch.com>
>><milesm@avnet.co.uk>; europa@avnet.co.uk <europa@avnet.co.uk>
>>Date: Wednesday, August 25, 1999 11:01 PM
>>Subject: Re: Flying: Fuel consumption in 'Classic' with Rotax
>>912.
>>
>>
>>>Dear wiggly-amps technofiles
>>>
>>>Re EFI for GA:
>>>
>>>I would assert that the failure rate of the whole system ie black box,
>>>looms, power source etc is significantly worse than duel mag systems.
The
>>>duel mag always wins because it is positively checked at every flight so
>we
>>>have a dual system with a relatively low MTBF requirement.
>>>
>>>Since we can't compete with the dual mag system we need to ask what is
the
>>>real reliability requirement. A good benchmark is the probability that
>the
>>>pilot will spontaneously drop dead at the controls. My friends in the
>CAA
>>>tell that this is about 1 in 1000000 per flying hour. My anecdotal
>>>experience with cars with black boxes tells me that EFI is not that
>>>reliable, particularly if subjected to a relatively low level of TLC.
>>>
>>>Making an EFI duel channel with genuine redundancy is not so easy as it
>>>first appears (as we found at Cosworth). In any case times sequential
>>>injection is completely unnecessary for an aircraft engine which spends
>>most
>>>of its life running pseudo-steady state.
>>>
>>>This whole area is under continuing discussion within a working group in
>>the
>>>JAA so clearer (and I believe sensible) guidelines will emerge. We then
>>>need viable technical solutions to emerge!
>>>
>>>Regards......MARK WILKSCH
>>>
>>>
>>>-----Original Message-----
>>From: ami mcfadyean <ami@mcfadyean.freeserve.co.uk>
>>><milesm@avnet.co.uk>; europa@avnet.co.uk <europa@avnet.co.uk>
>>>Date: Wednesday, August 25, 1999 06:06
>>>Subject: Re: Flying: Fuel consumption in 'Classic' with
Rotax
>>>912.
>>>
>>>
>>>>OK, OK,so mechanically injected 2-stroke charged diesels are superior to
>>>>everything else!! but the failure rate of modern automotive FI
>electronics
>>>>compares favourably with the rate at which TWIN magneto failures kill
and
>>>>injure people in the US.
>>>>Just to put things in perspective.
>>>>
>>>>However the reliability of the wiring is another matter. I agree that in
>>>>this respect some car based solutions are less than ideal.
>>>>
>>>> Bosch, nevertheless (the manufacturer of the K and subsequent
electronic
>>>>systems) consider the latter to be more reliable. Maybe they have an axe
>>to
>>>>grind (and they don`t make the wiring).
>>>>.
>>>>
>>>>Duncan McFadyean
>>>>
>>>>-----Original Message-----
>>>From: Wilksch Airmotive <mark@wilksch.com>
>>>><europa@avnet.co.uk>
>>>>Date: Monday, August 23, 1999 9:49 PM
>>>>Subject: Re: Flying: Fuel consumption in 'Classic' with
>Rotax
>>>>912.
>>>>
>>>>
>>>>>Dear Miles
>>>>>
>>>>>
>>>>>I have a better way of doing it. Pure machanical with electronic
>>>trimming.
>>>>>
>>>>>There is no use in messing about with car based solutions. The only
one
>>>>>which will succeed in the long run is one designed to do the right job
>>for
>>>>>aviation - no
>>>>>compromises. I would love to get one up and running on a 912! My
>>>>>solution will also work on 200,000 Lycontinentals.
>>>>>
>>>>>Development cost is the problem - will Nigel help us with the budget?
>>>How
>>>>>many orders can you get me?
>>>>>
>>>>>I can show you some of the components next time we meet - you will be
>>>>>impressed!
>>>>>
>>>>>Regards...........MARK WILKSCH
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>-----Original Message-----
>>>>From: Miles McCallum <milesm@avnet.co.uk>
>>>>>Date: Monday, August 23, 1999 11:55
>>>>>Subject: Re: Flying: Fuel consumption in 'Classic' with
>>Rotax
>>>>>912.
>>>>>
>>>>>
>>>>>>Carbs.. CV carbs... I don't want carbs at all: I want fuel injection.
>>>>Nigel
>>>>>>has the kit (not Rotax - they won't do it) but he won't get around to
>it
>>>>>>until enough people show an interest......
>>>>>>
>>>>>>M
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>
>
|