All
excellent questions.
My
reading of the technical literature combined with some experience suggests that
cylinder/head/induction temperature are the big swingers in detonation occurrence,
with bore size, mixture ratio, hot spots and combustion chamber configuration
also contributing. Anything that makes the brew hot as it approached top
dead center could contribute to an "autoignition" event in which the
mix goes bang, with undesirable results.
I
had to do some intensive work in this area when John Forker was ferrying a
Lancair IV from New Zealand to California. It was (is)
equipped with an engine similar to that in Brent's plane, an IO-540 Lycoming
with dual turbochargers and intercoolers, compression ratio unknown to me,
possibly also 8.5 to 1.
John
unknowingly got a bad load of fuel in Samoa, and half way to Hawaii started
seeing temperature problems which we finally diagnosed as the fuel to the
engine slowly changing from Avgas to whatever crap was pumped in at Samoa,
probably 100 LL cut with jet A. He kept reducing power setting and enrichening
to avoid detonation and heating, and quickly was running so rich that Hawaii
and Samoa were both out of reach. So he turned right and went 900 miles
to the island of Kirabati (and was promptly arrested, but that is another
story.)
We
shipped tools and instruments via the weekly fight from Hawaii,
and the engine inspection suggested he caught the problems in time by reducing
power settings and pumping in a lot more fuel. So problem one solved:
airplane appeared flyable.
Problem
two: only unleaded regular autofuel available, 87 octane. (Aside:
autofuel octane and AVGAS octane are derived differently, so be careful in
comparisons. Mixture ratio also changes apparent octane rating, richer
giving better detonation resistance.) It was out of the question to ship
a barrel of AVGAS to the island. Trust me. Ditto for octane
enhancer. Nobody will fly it, and the delivery ship comes about four
times a year. Maybe.
So
the question of the day: can you burn 87 octane unleaded autofuel in this
aircraft engine? And your life will depend on it.
I
burned the phone lines for a week, talked to many engine builders, consulted
the literature, and concluded that rich enough, with low enough
manifold pressure and low enough CHT, it should be OK. So John
drained one tank, put in some autofuel, and we restricted him to 23 inches (maybe
it was 25, memory fades) and 350F CHT. He took off, climbed to cruise
altitude, circled for one hour, could find no problems, and landed. The
other tank was drained, and fuselage tank and both wing tanks filled with
autofuel. Fortunately he had 6000 feet of runway at sea level, and he
used it all to get the airplane off. It was a real dog being heavy and
taking off at reduced throttle.
John
flew on to Hawaii, although once near, to add insult to injury, ATC gave him a
1 hour delaying vector due to military activities a couple of hundred miles off
shore. But he had LOTS of fuel, and ran nice and rich with well retarded
throttle, and everything stayed cool.
Once
landed, the engine got a thorough going over by some A and Ps at Honolulu.
No problems. John gave away a lot of auto gas to the FBO guys (lots of
cars got filled), flushed the tanks, and later flew on to California. The
engine is still doing fine, as far as I know. John normally runs LOP at
reduced power settings, but lately reported that his fuel controller is a bit
unstable at low flow rates, and needs a rebuild. So lately he runs 50F
rich of peak, also at lower power settings.
One
thing learned from the research was that as a rough rule of thumb,
reducing CHT 20F is equivalent to raising octane one point in terms of
detonation margin. That means if you go from 400F for an air cooled
engine to 200F for a liquid cooled engine (operating at 176F coolant
temperature, allowing for temperature drop across the head), you can reduce
octane requirement about 10 points. Low and behold, in liquid cooled
engines you can run fairly high BMEP and compression ratio with 92 octane
(remember octane comparison caution.)
However,
keep in mind, bore size also counts a lot in this mix. Detonation is
worsened by big cylinders.
But
the tech literature is clear. To avoid detonation, reduce manifold
pressure, reduce temperature, and run rich. Doing two out of three is
probably adequate. What the temperature, manifold pressure, mixture,
compression ratio map looks like, I am not sure. Only data will tell the
deed.
Your
comments, corrections, and suggestions are welcome. Just keep in mind
that like Brent Regan, I know nothing. I only simulate competence.
Use your own judgment and experience. Your mileage may vary.
Fred
Moreno
-----Original Message-----
From: Marvin Kaye [mailto:marv@lancaironline.net]
Sent: Tuesday, 25 April 2006 12:43 PM
Subject: Re: Compression, boost, and detonation
Posted
for "richard titsworth" <rtitsworth@mindspring.com>:
George,
I was at your Jan 06 APS course (well worth it - I'd recommend it anyone)!
I know/realize "tuning" is all very inter-related (and often secret),
but a
few questions.
Was it a TSIO-550 w/ 8.5:1 or a normalized IO-550? Any other mods -
ceramic piston domes, etc?
Any "landmarks" (MP, RPM, FF, CHT) that seem to consistently
represent the
"detonation edge"?
Any info you can share/suggest for spark advance, and/or target manifold
temperatures, to minimize/optimize detonation with the TSIO-550 - 8.5:1?
I fully respect your testing. I love airplane engines and I know you hate
to hear about autos..
But, our detonation challenges (at reasonable mid 30"s MP boost levels,
with
reasonable 8.5:1, and with good 100LL fuel) occur where high perf autos
operate successfully.
(I'm not going there with the comparisons - just focusing on detonation
phenomenon)
I understand the effects of timing, RPM, and mixture, but those are
"patches" not "fixes". (Perhaps variable timing is
the fix?)
Seems like I keep coming back to scavenging/eliminating heat as the key to
delaying the onset of detonation, HP, economy, etc.
Any sense/confirmation on this? Any detailed insight on what heat source
is
the prime cause (internal head face, piston crown, walls, value face, etc)?
How does the air (in the cylinder) get over-heated?
Is the air too hot before it enters the cylinder? (that seems controllable
if we're willing to accept additional cooling drag)
Does heat built too much during compression (adiabatically)? (@ 8.5:1
compression is not un-reasonably high)
Is it absorbed from the piston, head face, walls? (The air isn't in there
very long to be heated by radiation/convection - I think I'll try some heat
transfer modeling - any insights on coatings?
Is it heated in the port while waiting for the intake value to open (sitting
there for 3 strokes)
Have you done much testing with value overlap, intake vs exhaust pressure
ratio, etc - perhaps residual air/exhaust from the prior cycle (hot) is
helping to heat the incoming air? Seems higher compression would help
reduce this effect (less residual air)?
Have you ever had the opportunity to test a TCM LTSIO-550 (liquid)? I
know
it's not the same cross-flow arrangement (and too much trouble for me to
consider for my plane), but I'm curious if it offers any detonation/heat
insights?
Have you ever tested the "Cool Jugs"?
http://www.liquidcooledairpower.com/products.shtml I know these are
for
smaller engines, but perhaps any heat/detonation insights?
I don't recall if you can simulate altitude with your dyno (i.e. throttle
the compressor input)?
Any sense of altitude's effect on detonation margins (say above 18,000)?
Presumed effects (ceteris peribus):
Hotter upper deck temps due to compressor pressurization (thus exercising
the intercoolers)
Higher exhaust back pressure due to waste gate being "more closed" to
drive
compressor to create upper deck pressure
Lower exhaust back pressure due to lower turbine exit pressure (ambient) -
negligible?
Rick Titsworth
APS Jan, 06
Lancair ES - TSIO-550E-8.5
rtitsworth@mindspring.com
cell: 313-506-5064