|
|
<<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>>
<< Lancair Builders' Mail List >>
<<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>>
>>
[John Schroeder asked about FADEC systems:]
I have been paying attention to the public comments and published material
by the OEMs about the leaning schedules. Some of those comments and claims
have changed from time to time.
Here is my best take on it all:
I do not think any of the presently proposed FADEC units will allow you to
operate the engine at WOT and LOP when the MP is above some arbitrarily set
(lower) MP.
That means that if you are 1500' MSL and wanted to operate at 27" MP x 2500
RPM, and well lean of peak, it would not be possible. That is my present
understanding. That is a major limitation that is unwarranted by good and
efficient operating practices. In my view, that is a show stopper. MUCH
better flexibility with three levers.
The OEMs don't understand or believe in operations at normal high power
settings but lean of peak. Thus, they have designed their FADECs to either
operate rich of peak at high power or to operate at peak or maybe lean of
peak at lower manifold pressure settings. The problem is, at lower MP
settings, it is harder and harder to get the engines to operate smoother
while lean of peak!!! They are programming themselves into a corner... in
my view.
Electronic ignition provides the opportunity to make things better in
several directions at once. The problem is, all of the existing systems
rely upon some version of a programmed schedule or a memory map control
algorithm. What is needed is a full blown unrestricted closed loop control
algorithm. It has been well understood in the engineering literature since
about 1975 that the only "right" way to do this is to use a sensor that
monitors the internal combustion pressure events in relation to crank angle.
None of the OEM proposals are pursuing that in any way.
Advancing the timing while lean of peak at lower power settings will help
the engines operate more smoothly.
But right now, at 75% power, a 520/550 c.i. class engine needs the timing
set at about 22degrees while 50 to 70F lean of peak. Guess where it is set
on an IO-550? Yes... 22d. That is one reason that the turbonormalized
engine in the AOPA give-away Bonanza runs so darn well... it sits there all
day long, at WOT and 2500 RPM and 75-100F LOP with 22d spark timing and that
combination ends up being almost ideal for that engine and configuration.
>>3. Would a combination of some kind of electronic/computer directed fuel
injection with the electronic ignition improve the performance of this
IO-550?
The folks at the small engine research program at U of Wisconsin have been
studying this and are hopeful that the engine manufacturers will see some
wisdom
of it
Well... I really don't think so. There is a HUGE amount of misunderstanding
about this subject.
The way the pulsed injection systems currently work is that they seek to
"coordinate" and time the pulse stream of fuel to coincide with valve
openings. That is nice, as far as it goes... but... BUT.... BIG but,
here.... by the time the engine is up to normal cruise RPM in the 2300 to
2500 RPM range, these systems no longer attempt to sequence the fuel pulses
with valve openings, but rather, they merely pulse the nozzle to control
total fuel flow.... and the result is... these systems perform very similar
(essentially identical) to a well tweaked continuous flow port injection
system... which is what you already have in your engine, now!!!
The airplane is a very different operating environment than the automotive.
An aircraft spends 98% of its time at essentially one power setting... and
it is at an engine RPM that is sufficiently high enough so that it cannot
make good use of sequential pulsed injection to gain any advantage I am
aware of, compared to existing much simpler systems. About the only
significant advantage the sequential pulsed injection systems have is during
ground taxi and idle... they should run smoother under those conditions.
I have seen no data that suggests that any of the proposed pulsed fuel
injection systems will actually achieve more uniform cylinder to cylinder
fuel/air ratios than what the dramatically simpler existing systems are able
to achieve if one just goes to the trouble to get the fuel injectors
"right". Simple is good. Especially in aircraft engines.
BTW... those were all good questions. ****************************
As a thought experiment, ask yourself how many SENSORS there are on one of
the new FADECs.
Two? Four? Eight? 12? 16? 20? 24? More ?
Six CHTs. Six EGTs... two fuel pressure. two induction air temp...
two manifold... two... two... two..
Then ask yourself how many of those sensors have to be functional before the
aircraft will be legal to takeoff?
THEN... ask yourself how many trips might have to be canceled because a
sensor would not allow the aircraft to be dispatched.
SIMPLE is better...
Regards, George
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
LML website: http://www.olsusa.com/Users/Mkaye/maillist.html
LML Builders' Bookstore: http://www.buildersbooks.com/lancair
Please send your photos and drawings to marvkaye@olsusa.com.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
|
|