What's the problem with electronic ignition?
Newps
Frank, Henderson wrote:
>
> Fellow aviation conversationalists:
>
> Since the benefits of electronic ignition are well known, and since
> the engineering effort required would be fundamental, I would like to
> know why this presents a special problem for us in the aviation
> community.
>
> I know, I know.....with electronic ignition, what would you do when
> the electrical system went out? This is the excuse I hear for sticking
> with magneto ignition. But wait a minute. There is an obvious solution
> to this problem.
>
> Simply keep the current two-plug, dual-magneto ignition system that's
> currently on all existing aircraft engines and incorporate a fail-safe
> switch, which would house an electronic ignition system tapped into
> the existing wiring. That way, electronic ignition would operate under
> all circumstances except upon electrical failure, when the ol'
> magnetos would kick-in.
>
> So what's the problem?
None as far as I'm concerned. Get rid of the stupid mags altogether.
And while you're in there tear out my mixture control. Why for
Chrissakes I got to dick with the mixture in the 1990's is beyond me.
I'm probably pissing away 1-2 gph for just this reason.
>
> Frank Henderson
> E-Mail: Frankie66 at mindspring dot com
> Web site: http://frankie66.home.mindspring.com
old cfii a&p
Frank Henderson wrote in message <37c741de...@news.mindspring.com>...
>Fellow aviation conversationalists:
>
>Since the benefits of electronic ignition are well known, and since
>the engineering effort required would be fundamental, I would like to
>know why this presents a special problem for us in the aviation
>community.
>
>I know, I know.....with electronic ignition, what would you do when
>the electrical system went out? This is the excuse I hear for sticking
>with magneto ignition. But wait a minute. There is an obvious solution
>to this problem.
>
>Simply keep the current two-plug, dual-magneto ignition system that's
>currently on all existing aircraft engines and incorporate a fail-safe
>switch, which would house an electronic ignition system tapped into
>the existing wiring. That way, electronic ignition would operate under
>all circumstances except upon electrical failure, when the ol'
>magnetos would kick-in.
>
>So what's the problem?
>
J. Neteler
Well, actually it is already available.
Some of the early pioneering work was done on homebuilts. Klaus Savier used
an advanced ignition system is his Varieze to set some very impressive
efficiency marks. He has a company that caters to the homebuilt end of the
market. See http://www.lsecorp.com for details.
On the certified end see http://www.unisonindustries.com/lasar.html for an
advanced system for Lycoming engines. Continental is making some rumblings
is this area as well, but I haven't yet seen anything that can be bought.
I have not yet seen any independent test results from the users of the LASAR
system that indicate whether or not it is worth the money.
A more efficient ignition system would be worthwhile from a safety
perspective as well. The increased efficiency is more time in the tanks.
On our Debonair, Gamijectors dropped the fuel flow from an average of 13.5
gph to 12.5 gph and thereby added about 30 minutes to the endurance. It
would be interesting to see if an improved ignition system could result in a
similar gain.
J. Neteler
frank_henderson
Bob Noel
> None as far as I'm concerned. Get rid of the stupid mags altogether.
> And while you're in there tear out my mixture control. Why for
> Chrissakes I got to dick with the mixture in the 1990's is beyond me.
> I'm probably pissing away 1-2 gph for just this reason.
Are there not scenarios where the ability to adjust the mixture
would be useful, e.g., a hot cylinder(s) and maybe even high altitude
operations?
(fwiw) I agree that it would be really nice to have a single
engine control rather than 2 or 3 per engine. But having a
human in the loop can reduce the cost of control systems because
the human can be assigned all or most of the control logic.
--
Bob
(I think people can figure out how to email me...)
(replace ihatessppaamm with my name (rnoel) and hw1 with mediaone)
Newps
Bob Noel wrote:
>
> In article <37C759B5...@mcn.net>, Newps <ne...@mcn.net> wrote:
>
> > None as far as I'm concerned. Get rid of the stupid mags altogether.
> > And while you're in there tear out my mixture control. Why for
> > Chrissakes I got to dick with the mixture in the 1990's is beyond me.
> > I'm probably pissing away 1-2 gph for just this reason.
>
> Are there not scenarios where the ability to adjust the mixture
> would be useful, e.g., a hot cylinder(s) and maybe even high altitude
> operations?
>
> (fwiw) I agree that it would be really nice to have a single
> engine control rather than 2 or 3 per engine. But having a
> human in the loop can reduce the cost of control systems because
> the human can be assigned all or most of the control logic.
A computer can do it so much better than you or I.
Bob Noel
> > (fwiw) I agree that it would be really nice to have a single
> > engine control rather than 2 or 3 per engine. But having a
> > human in the loop can reduce the cost of control systems because
> > the human can be assigned all or most of the control logic.
>
> A computer can do it so much better than you or I.
If and only if it is programmed correctly. I know it's
a tired old cliche, but the reality is that computers
can and do screwup in truly spectacular ways.
It can be hugely expensive to have a computer-controlled
system that can handle every situation correctly.
The NRE makes developing such a system prohibitive.
Roy Smith
> Are there not scenarios where the ability to adjust the mixture
> would be useful, e.g., a hot cylinder(s) and maybe even high altitude
> operations?
Presumably, an electronic engine control system would monitor engine
temps, density altitude, and a whole slew of other variables and adjust
the mixture accordingly.
> (fwiw) I agree that it would be really nice to have a single
> engine control rather than 2 or 3 per engine. But having a
> human in the loop can reduce the cost of control systems because
> the human can be assigned all or most of the control logic.
People are better than machines at making judgement calls. Machines are
much better than people at monitoring systems and continiously making
minor adjustments to optimize operation. The best of both worlds is an
automatic system which runs things under routine situations, but informs
and alerts a person who then has the ability to over-ride the defaults or
take over completely in an emergency.
The automatic system also needs a fail-safe, or "limp home" mode, where
even if all the fancy gizmos to up in smoke, the engine can keep turning
and producing enough power to keep the plane flying well enough to get to
a controlled landing at an airport.
James M. Knox
>Are there not scenarios where the ability to adjust the mixture
>would be useful, e.g., a hot cylinder(s) and maybe even high altitude
>operations?
Absolutely. But a well designed system should always fall back as the
situation dictates. And the final fallback is manual control. The newest
designs I have seen from folks like GAMI all have a "limp home" mode that
should keep then engine turning if anything will.
jmk
Matthew Whiting
Only if the engineers who design and implement the control system didn't
think of these scenarios (certainly this is possible, especially in a
brand new product). A well-designed system will be able to handle any
situation that you can and do it much more quickly, accurately and
consistently. Do you ever feel the need to adjust the mixture or the
spark advance on your car? Even in extremely hot or cold conditions?
An airplane need be no different with appropriate sensors and control
logic.
Matt
Mike O'Malley
>
> lawyers
cost
>
> Frank Henderson wrote in message <37c741de...@news.mindspring.com>...
> >Fellow aviation conversationalists:
> >
> >So what's the problem?
> >
> >Frank Henderson
> >E-Mail: Frankie66 at mindspring dot com
> >Web site: http://frankie66.home.mindspring.com
--
Michael A. O'Malley
\--== PP-ASEL ==--/
ICQ# 32677796
moma...@uiuc.edu
University of Illinois
at Urbana-Champaign
Institute of Aviation
William W. Plummer
scheduled airline. Now, why is it that you prefer to run an airplane
yourself? --Bill
Newps wrote in message <37C759B5...@mcn.net>...
>
>
>Frank, Henderson wrote:
>>
>> Fellow aviation conversationalists:
>>
>> Since the benefits of electronic ignition are well known, and since
>> the engineering effort required would be fundamental, I would like to
>> know why this presents a special problem for us in the aviation
>> community.
>>
>> I know, I know.....with electronic ignition, what would you do when
>> the electrical system went out? This is the excuse I hear for sticking
>> with magneto ignition. But wait a minute. There is an obvious solution
>> to this problem.
>>
>> Simply keep the current two-plug, dual-magneto ignition system that's
>> currently on all existing aircraft engines and incorporate a fail-safe
>> switch, which would house an electronic ignition system tapped into
>> the existing wiring. That way, electronic ignition would operate under
>> all circumstances except upon electrical failure, when the ol'
>> magnetos would kick-in.
>>
>> So what's the problem?
>
> None as far as I'm concerned. Get rid of the stupid mags altogether.
>And while you're in there tear out my mixture control. Why for
>Chrissakes I got to dick with the mixture in the 1990's is beyond me.
>I'm probably pissing away 1-2 gph for just this reason.
>
>
>>
James Bieker
introduced more parts, more elements that could fail, fewer options in a
failure scenario. In theory it sounds easy, but reality is different. There
is alot of "If it ain't broke, don't fix it" to overcome.
Gregory Travis
> None as far as I'm concerned. Get rid of the stupid mags altogether.
>And while you're in there tear out my mixture control. Why for
>Chrissakes I got to dick with the mixture in the 1990's is beyond me.
Automatic mixture controls for aircraft have been available for nearly
sixty years now. If you're up to writing a check, today, to Piper, Mooney or
Raytheon for a new Continental-powered aircraft, one of the options that
you'll be given is whether you want automatic mixture control.
Until very recently, AMC was the default configuration on new Barons and
Bonanzas.
Likewise, you can order a new Cessna, Piper, Maule, etc. today and
specify electronic ignition if you like.
In the history bin, AMC was an option on a number of Lycoming-powered
aircraft throughout the 1950s and 1960s.
And, of course, ten years ago you could order a new Mooney that didn't
have a single mixture or prop control lever. All that, as was ignition,
was automatic.
None of these optiopns, present or past, has ever lived up to what should
have been a huge marketing advantage. In fact, despite repeated attempts
to field them by the engine and airframe makers, no one seems to
want to buy them.
The problem is the market. Not laywers. Not FAA certification
procedures. The market. Pilots have consistently rejected the automation
of certain engine management functions; eschewing the extra cost, weight,
and possible loss of reliability that comes with automated systems
and choosing to live with a bit of extra workload instead.
I believe that two things conspire to make this so. First, unlike
in an automobile, engine operating conditions in aircraft simply don't
change that often. So there simply isn't the return from automated
systems in aircraft that you get in automobiles.
Second, in general a pilot's overall workload is lower flying a plane than
it is driving a car. In other words, there's more free time in
the cockpit that one can devote to system's management. This really came
home for me yesterday. I had to do about 100 miles of freeway driving
and then about 300 miles of lightplane flying. During the freeway driving
I was CONSTANTLY having to worry about my position relative to the
road, to other cars, to my exits, etc. Looking at a map while driving
is something of a stunt, to say the least.
I had none of those concerns while flying. It was a hazy VFR day and
I was on an IFR clearance. My workload consisted of changing headings
over VORs every forty minutes or so and systems monitoring. I had lots
of time to screw with the mixture if I wanted to.
>I'm probably pissing away 1-2 gph for just this reason.
Both Continental and Lycoming built full EFI/FADEC engine management systems
in the early 1980s and flew them on conventional planes. The results
of both of their studies showed a fuel efficiency improvement of "up to
5%" using the system vs. average pilot technique.
My aircraft burns 9.5 GPH block to block. A 5% improvement would drop that
to just over 9 GPH - a savings of under 0.5 GPH. During a typical three hour
trip (like yesterday's), I'll save 1.5 gallons of gas.
That's not too bad, it's like getting a free beer at the end of every
flight. But will the weight and added maintenance of the extra stuff
conspire to rob me of my free beer? I, and the rest of the market,
think so.
Frankly, I like to go for the low hanging fruit. If I wanted to seriously
improve my fuel consumption, I'd spend my $800 on a set of GAMInjectors
and run lean of peak. No moving parts. No damn electronics. No
additional maintenance. And a real 1-2GPH fuel savings.
greg
--
gregory travis |"If you want to build this rich functionality and you're
gr...@littlebear.com|using a proprietary system, you have to start from scratch.
-------------------|But since Windows NT Embedded already has the rich Windows
NT feature set, developers can provide more functionally rich solutions..."
frank_henderson
wrote:
>> lawyers
>cost
I think "lawyers" is closer to the truth. Product liability, while
recently restrained, is still a monster.
"Cost", however, can't be so.
The cost to run the engineering numbers and program that information
into a microchip are not that great. This is just a day's work for
some electrical engineering outfits.
Then, the real work of engineering the whole system would begin:
wiring, switching, and controls. I would add an "Auto" position above
the "L", "R", "Both" ignition switch, allowing total pilot control
over ignition mode.
frank_henderson
wrote:
>Automatic mixture controls for aircraft have been available for nearly
>sixty years now. ...
>
>The problem is the market. Not laywers. Not FAA certification
>procedures. The market.....
That's a great post.........about auto mixture controls. I tend to
feel the same way about mixture: it doesn't change much and you need
to allow total pilot authority.
But when it comes to ignition, this is something that's always just
"on" and would lend itself well to automation. Variable timing is key.
I think the reason for market apathy towards this matter is because no
one has really done it right. The Mooney-Porsche PFM of 1988 is not
what I have in mind.
Auto ignition with magneto backup and full pilot control, using
existing wiring and systems is the answer. For example, I propose an
"Auto" position above the "L", "R", "Both" ignition switch, thereby
allowing full pilot control.
Beautiful, no?
>... If I wanted to seriously
>improve my fuel consumption, I'd spend my $800 on a set of GAMInjectors
>and run lean of peak. No moving parts. No damn electronics.
Electronic fuel injection is another matter, but something that could
still be done allowing for a conventional mixture control.
Bob Noel
at mindspring dot com wrote:
> >> lawyers
>
> >cost
>
> I think "lawyers" is closer to the truth. Product liability, while
> recently restrained, is still a monster.
>
> "Cost", however, can't be so.
>
> The cost to run the engineering numbers and program that information
> into a microchip are not that great. This is just a day's work for
> some electrical engineering outfits.
>
> Then, the real work of engineering the whole system would begin:
> wiring, switching, and controls. I would add an "Auto" position above
> the "L", "R", "Both" ignition switch, allowing total pilot control
> over ignition mode.
unfortunately it can be so. One word: Certification
When was the last time you got an TC amendment or STC approved?
I think it'll take some time before the new AC 23.1309-1A provides
any help.
Gregory Travis
<SPAM BLOCK: Frankie66 at mindspring dot com> wrote:
>>... If I wanted to seriously
>>improve my fuel consumption, I'd spend my $800 on a set of GAMInjectors
>>and run lean of peak. No moving parts. No damn electronics.
>
>Electronic fuel injection is another matter, but something that could
>still be done allowing for a conventional mixture control.
The fuel consumption advantage of electronic fuel injection over
constant mechanical injection would be absolutely minimal in an
aircraft. I daresay one would be hard pressed to detect ANY fuel
consumption change whatsoever. That, of course, begs the question of
why someone would want to exchange a dirt-simple injection system
with a highly complicated system dependent on hundreds of wires,
solenoids, transistors, etc. Not to mention the dangerous high-pressure
fuel pumps and high-pressure fuel lines. There is simply no reason whatsoever
to go with electronic fuel injection in an aircraft engine. It
complicates maintenance, decreases reliability, increases cost, and
increases weight. All for no gain whatsoever save, perhaps, a little
better idle and fewer emissions at idle on the ground.
Gerard Caron
Bob Noel <ihates...@hw1nospam.net.invalid> wrote in message
news:ihatessppaamm-2...@192.168.1.150...
> In article <37c97288...@news.mindspring.com>, SPAM BLOCK:
Frankie66
> at mindspring dot com wrote:
>
> >
> > >cost
> >
> > I think "lawyers" is closer to the truth. Product liability, while
> > recently restrained, is still a monster.
> >
> > "Cost", however, can't be so.
> >
Amen, brother. Not only is it expensive to get the design approved,
there is the added costs due to the environmental requirements and the
FAA approved QA program, configuration control, and the 'hard to
estimate' product support costs--warranty, etc. And then you have
little problems like redesign and recertification due to an obsolete
part. And all this NRE has to come out of the gross margin, in addition
to those other trivial business expenses; overhead, G&A, liability...
Makes you wonder why anybody gets into this business. Remember the old
joke:
Q: Do you know how to make a small fortune in aviation?
A: Start with a large one.
> When was the last time you got an TC amendment or STC approved?
I stick to TSO's. Although I spend a lot of time answering questions
from the TC/STC applicant.
> I think it'll take some time before the new AC 23.1309-1A provides
> any help.
It probably will help the little companies the most. The problem is
that by using a lower cert level, you limit your market (to say, single
engine recips under 6K.) A small guy might do that, but someone like
AlliedSignal (or is it Honeywell, or Honey-Allied?) won't accept that
limitation. They'd rather spend a little more, and amortize it over a
larger market.
BTW, the question of FADECs and electronic ignitions came up during the
AC rewrite (I was there along with about 2 dozen others from around the
industry and FAA.) The sore point is that the FAA considers an engine
controller failure to be potentially catastrophic, meaning it needs the
highest level of cert. On a single engine a/c it isn't really true, but
the argument was that it could be installed on any Part 23 a/c, and on a
ME it is critical. So here the new requirements can help by allowing
the part to be limited to SE a/c.
Gerry
frank_henderson
wrote:
>The fuel consumption advantage of electronic fuel injection over
>constant mechanical injection would be absolutely minimal in an
>aircraft.
You're probably right, good point.
>.....a highly complicated system dependent on hundreds of wires,
>solenoids, transistors......dangerous high-pressure
>fuel pumps and high-pressure fuel lines.
This made me think.
Does it REALLY have to be that complicated? Isn't the fuel already
delivered under pressure to the injectors? The only thing you would
hope to accomplish with electronic control would be timing and
duration, right? So the only change would be those two parameters of
the existing valve, so I'm not sure why existing pumps, hoses, and
pressures wouldn't work.
> All for no gain whatsoever save, perhaps, a little
>better idle and fewer emissions at idle on the ground.
If that's all there is to gain with EFI, you're definitely right. But
it seems to me electronic ignition would yield much more.
Frank Henderson
E-Mail: Frankie66 at mindspring dot com
Web site: http://frankie66.home.mindspring.com
Ron Natalie
>
g away 1-2 gph for just this reason.
>
> would be useful, e.g., a hot cylinder(s) and maybe even high altitude
> operations?
>
for itself just fine. The reason is purely a combintation
of tradition and KISS, rather than technological.
George R. Patterson III
>
> Simply keep the current two-plug, dual-magneto ignition system that's
> currently on all existing aircraft engines and incorporate a fail-safe
> switch, which would house an electronic ignition system tapped into
> the existing wiring. That way, electronic ignition would operate under
> all circumstances except upon electrical failure, when the ol'
> magnetos would kick-in.
>
> So what's the problem?
It costs nearly three times as much as a set of magnetos. Slick makes
it. It replaces your mags, but it works as you describe - electronic
ignition primary with magneto backups.
George Patterson, N3162Q.
Gregory Travis
<SPAM BLOCK: Frankie66 at mindspring dot com> wrote:
>Does it REALLY have to be that complicated? Isn't the fuel already
>delivered under pressure to the injectors? The only thing you would
>hope to accomplish with electronic control would be timing and
>duration, right? So the only change would be those two parameters of
>the existing valve, so I'm not sure why existing pumps, hoses, and
>pressures wouldn't work.
The fuel pressure used on aircraft injection systems is, as I
recall (don't have references here) in the 15-25PSI range. The
EFI system on my Toyota has pressures in the 80PSI area. The potential
for catastrophy following a leak in the Toy system is pretty
high. I wouldn't want it to happen to me at any altitude above
ground. :-)
The injectors used in aircraft engines are constant mechanical
injectors. Fuel is ALWAYS flowing through them when the engine is
running; even when the intake valve is closed. The only variable
is the rate of fuel flow through the injector. There is no injection
timing as there is with electronic (solenoid) injectors or with
diesel injection.
Yes, you can automate the injection flow rate. That is, essentially,
what the automatic mixture control does. It's also what your mixture
control knob does.
>If that's all there is to gain with EFI, you're definitely right. But
>it seems to me electronic ignition would yield much more.
Eectronic ignition gives you about a 5% improvement in block-block fuel
consumption given average pilot technique. This is real-world data that
Unison, Lycoming, and Continental have all generated during independent
tests of electronic ignition systems.
It's actually variable ignition timing that gives you the 5%, not the
electronic part. It's just so much easier to implement variable ignition
with electronics.
Ron Natalie
>
>
> It's actually variable ignition timing that gives you the 5%, not the
> electronic part. It's just so much easier to implement variable ignition
> with electronics.
Also gives you some better performance/reliability with regard
to plug fouling. Magnetos are pretty weak sparks.
Gregory Travis
Magneto output is a function of magneto RPM which is a function of
engine RPM. Magnetos do, indeed, produce a weak spark at low engine RPM
and starting.
However, magneto output at cruise RPM is significant. Several kilovolts
above that of a regular distributor/coil arrangement. Not as hot as
full electronic ignition to be sure but still no "weak spark."
Interestingly, one of the bugs that surfaced during LASAR testing was
that the hotter spark generated by LASAR was destroying aviation
ignition harnesses and plugs. Champion was working on a new set of
harnesses and plugs last that I heard.
Charles K. Scott
Frank Henderson writes:
> If that's all there is to gain with EFI, you're definitely right. But
> it seems to me electronic ignition would yield much more.
But would it? You really can't mix what happens to an automobile
engine in it's normal city driving cycle (which is where electronic
ignition tends to be most helpful with it's ability to instantly and
constantly change it's advance) with how an aircraft engine runs.
Aircraft engines really don't need instantaneous hyper ajustable
variable timing because once you get to cruise, you set the rpms and
that's it. All that is necessary is that the timing be correct for
that setting and perhaps for starting so as to make it easier to start.
The way aircraft engines operate really precludes the need for variable
adjustable timing.
Corky Scott
Jim Sokoloff
> In article <37C7F1B5...@mcn.net>, Newps <ne...@mcn.net> wrote:
> > > (fwiw) I agree that it would be really nice to have a single
> > > engine control rather than 2 or 3 per engine. But having a
> > > human in the loop can reduce the cost of control systems because
> > > the human can be assigned all or most of the control logic.
> >
> > A computer can do it so much better than you or I.
>
> If and only if it is programmed correctly.
That is only presuming that the sensors work correctly. I wouldn't fly
my engine solely by reference to the 6 cyl EGT/CHT. (I set the mixture
from it, but I consciously cross-check that with how it feels, what the
TAS is, etc. If I come up 10 kts short on TAS, and it feels a tiny bit
rough, I'm probably too lean, though I can't read it on the GEM.)
> I know it's
> a tired old cliche, but the reality is that computers
> can and do screwup in truly spectacular ways.
Indeed.
---Jim, computer jockey by day (and night lately)
Jim Sokoloff
> A well-designed system will be able to handle any
> situation that you can and do it much more quickly, accurately and
> consistently.
Would you fly behind an engine that you couldn't control manually? I
write computer code all day for financial systems and I wouldn't fly
behind the code I write. (It's good code and I'll completely entrust my
money to it, but not my life and the lives of my friends...)
> Do you ever feel the need to adjust the mixture or the
> spark advance on your car?
Do you manually adjust the spark advance on your airplane? (Maybe you
do, but not if you fly anything modern...)
In only 12 years of driving, I've been stranded by a failed spark
advance unit on a distributor, failed engine-driven fuel pump (no
standby electric boost pump), carb ice in an old VW, and a handful of
transient anomolies. A complete engine failure every 4 years (and
perhaps 70,000 miles) is more risk than I am willing to take.
Certainly, aircraft level redundant systems would have helped two of the
failures, and a control system could be programmed to have handled each
of those cases. But, I don't trust that covering each individual failure
means you can cover failures in combination, nor that you can even
figure out programmatically what's wrong. Certainly it's challenging to
program a computer to solve problems "out of the box" as well as a human
whose very life is tied to the outcome.
Taking control from the human pilot removes yet another bit of
information that the pilot can use to manage an emergency to an outcome
that minimizes injury and loss of life, IMO. (And yes, I'd hate flying
an Airbus...)
---Jim
Jim Sokoloff
> The sore point is that the FAA considers an engine
> controller failure to be potentially catastrophic, meaning it needs the
> highest level of cert. On a single engine a/c it isn't really true, but
> the argument was that it could be installed on any Part 23 a/c, and on a
> ME it is critical.
Um, how is a FADEC failure in a single NOT "potentially catastrophic"? I
can admit that I'd rather have one fail on takeoff in a single than in a
twin, but other than that, I'd rather have a single engine failure in
something with as many engines as I can get.
---Jim
Matthew Whiting
>
> Matthew Whiting wrote:
>
> > A well-designed system will be able to handle any
> > situation that you can and do it much more quickly, accurately and
> > consistently.
>
> Would you fly behind an engine that you couldn't control manually? I
> write computer code all day for financial systems and I wouldn't fly
> behind the code I write. (It's good code and I'll completely entrust my
> money to it, but not my life and the lives of my friends...)
Yes, and probably have several times while flying on USAirways
airplanes! I used to write software for real-time process control
systems and I wouldn't fly behind code written by you either! Software
development for hard realtime systems is nothing like writing software
for most commercial applications. Correctness is a function not only of
the numerical correctness of the computations, but ALSO a function of
meeting the time deadlines. A "correct" calculation that is 10
milliseconds too late, maybe no better than a numerically incorrect
calculation.
I no longer write software (I manage a group that does), but would fly
behind an engine controlled by software produced by my engineers ...
well, most of them anyway!! :-)
> > Do you ever feel the need to adjust the mixture or the
> > spark advance on your car?
>
> Do you manually adjust the spark advance on your airplane? (Maybe you
> do, but not if you fly anything modern...)
No, which is my point. You don't mind having automatically controlled
spark timing so why not mixture control?
> In only 12 years of driving, I've been stranded by a failed spark
> advance unit on a distributor, failed engine-driven fuel pump (no
> standby electric boost pump), carb ice in an old VW, and a handful of
> transient anomolies. A complete engine failure every 4 years (and
> perhaps 70,000 miles) is more risk than I am willing to take.
Well, in 24 years of driving, I've been stranded once ... also by carb
ice in a VW. But this wasn't the fault of the car. The car had been
converted from FI to carburetion and whoever did the job neglected to
install the heater hose from the heat excanger to the air cleaner
snorkel.
In 21 years of flying, I've been stranded by a vacuum failure (hose from
a Precise Flight standby system came loose) and an electrical failure.
> Certainly, aircraft level redundant systems would have helped two of the
> failures, and a control system could be programmed to have handled each
> of those cases. But, I don't trust that covering each individual failure
> means you can cover failures in combination, nor that you can even
> figure out programmatically what's wrong. Certainly it's challenging to
> program a computer to solve problems "out of the box" as well as a human
> whose very life is tied to the outcome.
>
> Taking control from the human pilot removes yet another bit of
> information that the pilot can use to manage an emergency to an outcome
> that minimizes injury and loss of life, IMO. (And yes, I'd hate flying
> an Airbus...)
The information can be provided by the computer if necessary.
Matt
frank_henderson
wrote:
>Electronic ignition gives you about a 5% improvement in block-block fuel
>consumption given average pilot technique.....
One more thing........
Remember, the goal of automating ignition (and perhaps fuel injection)
is not necessarily fuel efficiency.
The primary objective is reliability. In aircraft engines, this is the
supreme goal. Reliability can be increased by having an engine run
more smoothly and with consistent operating temperatures.
In opinion, many of the problems on light aircraft can be traced to
vibration and fatigue. A Malibu pilot commented once that a
significant portion of his annual bill was due to vibration (caused
partially by the two-bladed prop. He now has a three-blader and is
hoping for smaller repair bills.). A smooth-running engine would pay
big dividends toward longer parts life, reduced metal and joint
fatigue, and electronics reliability - not to mention increased pilot
& passenger comfort.
This is why turbine power can be such a boon.
And, all sorts of engine metal fatigue - cracked crankcases, for
instance - can be reduced with lower, consistent operating temps.
So really, lower fuel consumption is just icing on the cake!
Frank Henderson
E-Mail: Frankie66 at mindspring dot com
Web site: http://frankie66.home.mindspring.com
Roy Smith
> A smooth-running engine would pay
> big dividends toward longer parts life, reduced metal and joint
> fatigue, and electronics reliability - not to mention increased pilot
> & passenger comfort.
>
> This is why turbine power can be such a boon.
I remember reading somewhere that when turbine power came along, they
started having problems with failing instruments. Seems the vibration
from a piston engine keeps the bearings in old steam gauge instruments
from sticking. Without the constant vibration, the needles would tend to
get stuck. The solution was to add a small mechanical vibrator to the
instrument panel!
Might be urban legend, but it sounds plausable.
--
Roy Smith <r...@popmail.med.nyu.edu>
New York University School of Medicine
Gregory Travis
<SPAM BLOCK: Frankie66 at mindspring dot com> wrote:
>wrote:
>
>>Electronic ignition gives you about a 5% improvement in block-block fuel
>>consumption given average pilot technique.....
>
>One more thing........
>
>Remember, the goal of automating ignition (and perhaps fuel injection)
>is not necessarily fuel efficiency.
>
>The primary objective is reliability. In aircraft engines, this is the
>supreme goal. Reliability can be increased by having an engine run
>more smoothly and with consistent operating temperatures.
>
>In opinion, many of the problems on light aircraft can be traced to
>vibration and fatigue. A Malibu pilot commented once that a
>significant portion of his annual bill was due to vibration (caused
>partially by the two-bladed prop. He now has a three-blader and is
>big dividends toward longer parts life, reduced metal and joint
>fatigue, and electronics reliability - not to mention increased pilot
>& passenger comfort.
How would electronic ignition make my engine run "smoother?" About the
only thing it can do is advance the timing to compensate for lower manifold
pressures and thus recover some power. That will tend to make the engine
vibrate more, not less.
Piston engine "smoothness" is first a function of design (inline and (especially)
opposed engines are best in this regard), number of cylinders (pick a number that
eliminates any primary or secondary forces or moments - 6 is a good one). Secondarily
it's a function of power balance. Make sure each cylinder is getting an equal
amount of fuel, air, and that the ignition timing is the same for each
cylinder. Finally, it's a function of manufacturing. Namely, get the balance
and volumes consistent.
>This is why turbine power can be such a boon.
Depends on the turbine and the piston engine in comparison. Take a ride in
a Garrett-powered Turbine Commander and then take a ride in a Continental-powered
Cessna 414. I think you might well revise your statement.
>And, all sorts of engine metal fatigue - cracked crankcases, for
>instance - can be reduced with lower, consistent operating temps.
>>So really, lower fuel consumption is just icing on the cake!
For any given specific fuel consumption, operating temperature is a function
of power developed. The only way to lower it (operating temperature) is to
either reduce power or increase cooling. Or you can lower specific fuel
consumption (increase engine efficiency). So you see, it's really much more
than just icing on the cake.
frank_henderson
wrote:
>How would electronic ignition make my engine run "smoother?"
Remember the old saying: "Timing is everything".
> About the
>only thing it can do is advance the timing to compensate for lower manifold
>pressures and thus recover some power. That will tend to make the engine
>vibrate more, not less.
I think what you meant to say was that, at lower rpm, retarding the
timing (not advancing it) would recover some power. I would agree.
But it also would smooth things out.
The point on the crankshaft at which ignition occurs is critical.
Older cars (without electronic ignition) required the ol' fashioned
"tune-up" to keep them running smoothly. Smoothing out the power band
is, I believe, the whole reason for spark advance in the first place.
Not only that, but a retarded timing should also yield quicker starts
and less starter wear.
>Piston engine "smoothness" is ...a
>function of design.....of power balance.......of manufacturing.
All true too.
>.....Take a ride in
>a Garrett-powered Turbine Commander and then take a ride in a Continental-powered
>Cessna 414. I think you might well revise your [smoothness] statement.
Well, I have to stick to my guns on this one....
A turbine engine is inherently smoother than a piston. There isn't the
changes in direction and momentum required of a piston engine in order
to produce power. A turbine continuously turns in one direction.
If a turbine isn't running smoothly, then I would say the prop is the
culprit, not the engine. The turbine-powered Cessna P210 makes a great
comparison, and the consensus is unanimous.
>....The only way to lower it (operating temperature) is to
>either reduce power or increase cooling.
That's one way to do it.
Although I'm not sure what role ignition/injection plays in operating
temperature, my intuition tells me there is something there. We all
know that enriching the mixture cools EGT and leaning does the
opposite, so I have to conclude that perhaps greater control over
other ignition factors my influence operating temps as well.
The engineering parameters governing this topic I'm sure are well
known and I would welcome comments from any knowledgeable source,
just for kicks.
Frank Henderson
E-Mail: Frankie66 at mindspring dot com
Web site: http://frankie66.home.mindspring.com
MJones
is free) for a great discussion of this topic. Wade through the mixture
stuff to get to discussion of ignition timing. (Actually, the mixture
stuff is interesting, too.)
M.J.