Received-SPF: none
 receiver=logan.com; client-ip=69.147.92.97; envelope-from=keltro@att.net
DomainKey-Signature:a=rsa-sha1; q=dns; c=nofws;
  s=s1024; d=att.net;
  h=Message-ID:X-YMail-OSG:Received:X-Mailer:References:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type;
  b=r37ComdcNJeOm8Rj3y4bnrXRD4yGvjWt4KcfsZKbeF+6F5Ld2sUhpw6KyL+wd1G2tiGfLtF5fO2xpZ3gqgBOTEluLmvsU5mvjgvqYycVXgkH/t6Csy0o+UHMlsR0TT99qkjv+BVB76dSoEu9LZqD4nbyTUN/4YXzsKWkodQQsk0=;
Message-ID: <298241.32302.qm@web83901.mail.sp1.yahoo.com>
References: <list-4509376@logan.com>
Date: Sun, 17 Oct 2010 11:27:19 -0700 (PDT)
From: Kelly Troyer <keltro@att.net>
Subject: Fw: The Case For Turbocharging
To: Rotary motors in aircraft <flyrotary@lancaironline.net>
In-Reply-To: <list-4509376@logan.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="0-1081704649-1287340039=:32302"

--0-1081704649-1287340039=:32302
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

=0A=0A=0A=0ASorry guys..........Did not think about my "Open Office" files =
not opening for =0Aeveryone !!=0AHope this is not too big for the forum !!.=
..................=0A=0ATurbocharging Automotive Engines for Aircraft Appli=
cations=0AApril 27/98 =0AWith automotive engines finding increasing use pow=
ering homebuilt aircraft, we =0Afind many people entertaining the idea of t=
urbocharging to boost the power to =0Aweight ratios and altitude performanc=
e of these engines. We will examine the =0Avarious facts and myths regardin=
g turbocharging as applied to light aircraft =0Ause.=0AHistory=0ATurbos in =
aircraft were first successfully applied in mass numbers during WWII. =0ATh=
e P-38 and P-47 were the best known turbocharged fighter types and both had=
 =0Aexcellent high altitude performance.=0AIn general aviation use as appli=
ed to air cooled opposed engines, turbocharging =0Ahas a less revered reput=
ation. Much of this has to do with the air cooled engine =0Aitself and the =
fact that most of these engines were never designed to be boosted =0Ato beg=
in with. They may have been modified a bit for the turbo but with power to =
=0Aweight ratios being a major concern, many are structurally and thermally=
 not up =0Ato the task. Other concerns such as turbo cooling and lubricatio=
n by inferior =0Aaviation type oils also led to premature turbo failures.=
=0AWhat is a Turbocharger?=0AA turbo consists of a centrifugal compressor c=
onnected to a turbine wheel which =0Ais spun at high rpm by the energy of t=
he engine exhaust gasses. It is a very =0Asimple device, but the high tempe=
ratures and stresses acting on it require =0Aexotic materials in the turbin=
e section which are somewhat expensive. A turbo =0Awill cost in the neighbo=
rhood of $800 to $1300 with the wastegate assembly.=0A=0ATurbine wheels com=
e in many shapes and sizes. Constructed of high temperature =0Aalloys such =
as Inconel or Hastaloy. This is the hot end. =0A=0A=0ATurbine housings are =
cast from a high nickel/iron alloy. =0AThe compressor pressurizes the intak=
e manifold to achieve higher hp or maintains =0Asea level pressure as the a=
ircraft climbs. This permits higher climb rates at =0Aaltitude and increase=
d cruise speeds.=0A=0ACompressor wheels are cast from aluminum alloy and co=
me in many sizes or trims. =0AThis is the cold end. =0A=0A=0ACompressor hou=
sing is cast from aluminum alloy. Shape is designed to slow the =0Aair, thu=
s compressing it. =0A=0A=0ACast iron center section or bearing housing cont=
ains floating sleeve bearings or =0Amore recently ceramic ball bearings. Hi=
gh pressure oil from engine feeds the =0Abearing and is drained out the bot=
tom of the housing back into the engine. Some =0Ahousings are water cooled =
to prevent coking of oil deposits leading to bearing =0Afailure. =0A=0AEleg=
ant Simplicity=0ATurbocharging is the most efficient method of boosting hp =
with the least weight =0Apenalty known. A typical turbo installation on a l=
ight aircraft including =0Aintercooler and exhaust is around 35-50 pounds. =
Such an installation is capable =0Aof tripling hp at sea level or adding 50=
% to the naturally aspirated output up =0Ato 15,000 feet. Sea level manifol=
d pressures can be maintained to 25,000 feet in =0Asome cases. An aircraft =
capable of 200 knots at sea level would true out at =0Anearly 300 knots wit=
h a good turbo system at high altitude.=0ASpinning the compressor with an e=
xhaust driven turbine is a much better way than =0Aby the mechanical means =
as in a supercharger. It is lighter, more reliable and =0Amore efficient an=
d has the added advantage of having more drive energy available =0Aas the a=
ircraft climbs due to lower backpresure, which is exactly what is =0Aneeded=
. =0A=0AFuel Flow vs. Power=0AThere is a common misconception that turbocha=
rging increases fuel flows because =0Aof the backpressure of the turbine. T=
his may be true in air cooled engines using =0Aa poorly matched turbo witho=
ut intercooling because the air cooled engine cannot =0Amaintain temperatur=
e stability without adding fuel for cooling purposes. Reduced =0Afuel flows=
 at altitude in naturally aspirated and turbine engines is not due to =0Aso=
me magic process, it is due to the fact that as altitude is increased, less=
 =0Apower is produced thus less fuel is required. Many people also don't se=
em to =0Arealize that fuel flow is related to hp developed and that a turbo=
 engine will =0Adevelop more power at altitude than a naturally aspirated e=
ngine, obviously =0Arequiring more fuel and delivering higher speeds at the=
 same time. =0A=0ATesting done back in the '40s and '50s on turbo engines a=
ctually confirm that =0Afuel flows were LOWER at the same BMEP with turboch=
arging than in a naturally =0Aaspirated engine developing the same power. T=
his suggests that when properly =0Aapplied, the turbo actually recovers mor=
e energy from the exhaust to be applied =0Ato reduce pumping losses during =
the compression stroke than is consumed during =0Athe exhaust stroke workin=
g against higher backpressure. With approximately 50% =0Aof the energy in t=
he fuel going out the exhaust in the form of heat and =0Apressure, turbocha=
rging can recover some of this waste and put it to good use.=0AIt was shown=
 that turbos could achieve fuel flows 3-6% lower and =0Aturbocompounding co=
uld achieve up to 15% lower fuel flows. Cylinder head and =0Apiston cooling=
 were shown to be the primary limitations to achieving lower BMSFC =0Afigur=
es. We can conclude from this that as applied to liquid cooled engines with=
 =0Atheir superior thermal rejection rates, turbocharging really has few =
=0Adisadvantages compared to the many advantages it offers.=0AIt should be =
no surprise that the latest certified liquid cooled engines from =0AToyota =
and Orenda are turbocharged. People also should remember the awesome =0Atur=
bocompound radials of the '50s. These engines offered extremely low BSFC an=
d =0Ahigh power. It seems foolish not to turbocharge most good aircraft eng=
ines used =0Aat any kind of altitude as you are throwing away performance w=
ithout it.=0ATurbocharged Air Cooled Engines=0AAs mentioned earlier, turboc=
harged , conventional opposed air cooled engines =0Ahave a less than stella=
r track record and many lay people blame the turbo. In =0Afact, the fault l=
ies with the engines it is applied to. The average Lycoming or =0AContinent=
al has a lot of things not going for them. Many of these engines suffer =0A=
premature failure of crankcases, cylinders, heads and valves in their =0Atu=
rbocharged iterations. Let's examine why.=0AProbably the leading cause of c=
ase and barrel cracking is the lack of rigidity =0Athese engines have in th=
e cylinder/ case area. For ease of barrel replacement I =0Aassume, the desi=
gners of these engines chose to bolt the cylinders individually =0Ato the c=
rankcase at the bottom of the cylinder using a flange. This arrangement =0A=
creates a very willowy structure and with each firing impulse, combined wit=
h the =0Asheer mass of the huge reciprocating parts used in these engines, =
a definite =0Ahigh amplitude, cyclic stress is put on these parts eventuall=
y leading to =0Acracking. The increased gas pressures associated with turbo=
charging compound =0Athis problem. This design problem is actually the caus=
e of a major portion of =0Athe high vibration levels associated with conven=
tional aircraft engines.=0AThe flange method puts the barrel under a tensio=
n load with every firing impulse =0Atoo which is just plain stupid given th=
e thickness of the cylinders. Modern =0Aautomotive racing practice would us=
e through studs to retain the barrels, thus =0Aputting the barrels under co=
mpression.=0ACylinder head cracking and exhaust valve burning and sticking =
are primarily due =0Ato insufficient cooling in these areas which is exagge=
rated by the extra heat =0Aintroduced by turbocharging. Air cooling is rela=
tively inefficient at sinking =0Aheat off around the exhaust port and valve=
 seat area. This problem is compounded =0Aby the use of very low compressio=
n ratios resulting in higher EGTs thus even =0Ahigher heat flux in the crit=
ical exhaust port area. Typical air cooled aircraft =0Aengines run cylinder=
 head temperatures in excess of 400 degrees on a regular =0Abasis. With alu=
minum alloys losing roughly 50% of their strength at this =0Atemperature it=
 is easy to see why cracking is all too common here. The high =0AEGT's are =
also detrimental to the life of the turbo's turbine section.=0AValve proble=
ms are due to the same cause, temperatures being too high from =0Ainefficie=
nt heat transfer. Many factory turbo aircraft are forced to either open =0A=
cowl flaps or richen the mixture at high altitude to keep temperatures with=
in =0Alimits. With power being maintained at altitude and air density falli=
ng off, =0Athere is often insufficient mass flow available for cooling. Thi=
s is rarely a =0Aproblem on naturally aspirated engines as power and coolin=
g requirements are =0Adropping off with increasing altitude.=0ANot a lot of=
 design went into the induction system on many of these engines =0Ahence fu=
el distribution is usually not the greatest. This contributes to engine =0A=
roughness and problems with leaning. Leaning is limited by the mixture in t=
he =0Ahottest cylinder with the others running richer than desired.=0AThe o=
ils used in aircraft engines are relatively crude by modern synthetic =0Aau=
tomotive standards and consequently not the best for the turbocharger. The =
use =0Aof these oils appears to be a result of the very loose tolerances re=
quired in an =0Aair cooled engines because of the high operating temperatur=
es which also leads =0Ato the high oil consumption characteristic of these =
engines.=0AFinally, many certified installations never used intercooling wh=
ich is simply =0Aamazing. The advantages and necessity of it was clearly un=
derstood in WWII. High =0Acharge temperatures at altitude required even mor=
e fuel for cooling and yet more =0Acowl flap opening to control head temper=
atures.=0ATurbocharged Liquid Cooled Engines=0ALiquid cooled engines are a =
natural for turbocharging because of their higher =0Aheat rejection capabil=
ities. Water cooling also enables the turbo center section =0Ato be water c=
ooled which alleviates a primary cause of turbo failure- coked =0Abearing h=
oles from high heat and fried oil. Water cooled engines can use modern =0As=
ynthetic oils because of their tighter clearances. Synthetic oils with thei=
r =0Amuch higher temperature capabilities are more suitable to lubricate th=
e turbo =0Athan old fashioned, aircraft oils derived from natural base stoc=
ks. The popular =0AMobil 1 synthetic oil, common in the racing world, can w=
ithstand temperatures of =0A350 degrees continuously and even 450 degrees i=
ntermittently without significant =0Adegradation. With proper lubrication a=
nd cooling, turbochargers will easily last =0Athe life of the engine.=0AIf =
we compare the Popular Subaru EJ22 and EJ25 horizontally opposed 4 cylinder=
 =0Aengines to their certified counterparts we find many advantages especia=
lly in =0Aturbocharged trim. The Subaru has 5 main bearings instead of 3 wh=
ich means that =0Athe crank is better supported. The Subaru has a higher co=
mpression ratio for =0Ahigher thermal efficency. Its overhead cam, 4 valve =
per cylinder arrangement =0Aprovides lower friction and higher volumetric e=
fficiency than a 2 valve pushrod =0Asetup. Finally, the integrated block an=
d head design is many times stiffer and =0Astronger than the aircraft engin=
e.=0AAs long as the radiator and ducting are properly designed, cooling pro=
blems are =0Anot a concern on the liquid cooled engine.=0AThe induction sys=
tem on the Subaru has proper runner lengths to boost torque, =0Arelatively =
equal airflow to each cylinder and can easily use a modern digital, =0Aelec=
tronic fuel injection system to precisely meter fuel. All of these things =
=0Aaid in producing more power with less fuel.=0APower vs. Weight vs. Manif=
old Pressure vs. Life vs. Cost=0AThe EJ engines when turbocharged and equip=
ped with a radiator, PSRU and =0Acomposite propeller are slightly heavier t=
han the popular Lycoming and =0AContinental 360 cubic inch engines which ar=
e rated at 180- 200 hp for takeoff =0Aand around 135- 150 hp for maximum cr=
uise. The Subaru can easily attain 200- 250 =0Ahp at 50 inches for takeoff =
and reliably deliver 140 to 170hp at 35-38 inches =0Afor cruise. The 3.3L E=
G33 six cylinder is capable of 300+ hp for takeoff and =0A200-250 hp in cru=
ise making it a viable alternative to the O-540 engines. These =0Apower lev=
els can be maintained during climb and cruise at altitude thanks to =0Aturb=
ocharging and liquid cooling.=0AProjected TBO is in the 1000-1200 hour rang=
e at this time. Most people flying =0Athe EJ engines are confident that no =
interim work will be required during this =0Aperiod such as cylinder barrel=
 replacement and valve grinding operations which =0Aare quite common on air=
craft engines before reaching their stated TBO. In other =0Awords, if you c=
hange the oil and check the water, that is about all that is =0Arequired du=
ring those 1000 hours. There are no magneto or even valve adjustments =0Ato=
 worry about. Expect spark plug replacement every 100- 250 hours or so at $=
3 =0Aeach. Overhaul costs will range from about $500 to $1500 for parts and=
 machine =0Awork depending on condition. Labor would be in the $500 to $120=
0 range. We are =0Aall familiar with the costs to overhaul a certified airc=
raft engine. =0A=0AMatching a Turbocharger=0AAs applied to the popular Suba=
ru EJ22 and EJ25 engines for general purpose use =0Abelow 15,000 feet we ca=
n narrow the choice of turbos down a bit. Garrett T3 =0Aturbos are recommen=
ded as they are relatively cheap, reliable and easy to fix =0Awith a vast a=
rray of combinations available. The compressor side may be either a =0A-50,=
 -60 or super 60 wheel with a stage 3 turbine wheel in a .82 A/R housing, =
=0Adepending on the exact use and hp. These may be ordered with an integral=
 =0Awastegate and a water cooled bearing housing which are both highly reco=
mmended. =0AThese turbos are quite capable of producing enough boost to ext=
ract 250 hp at =0Asea level and still have 175+ hp available at 15,000 feet=
.=0AThe integral wastegates are generally more reliable than an external ty=
pe =0Aespecially on an aircraft application in which it is almost constantl=
y bypassing =0Aexhaust. The wastegate is simply a valve used to bypass exha=
ust gasses around =0Athe turbine to control turbine and compressor speed an=
d thus boost pressure. It =0Ais usually actuated automatically by a diaphra=
gm sensing manifold pressure.=0AMany people have tried to use the OE turbo =
on auto aircraft conversions, often =0Awith limited success. The turbos on =
cars were not matched to operate at high =0Aaltitude and continuous high po=
wer settings so they are mismatched and =0Ainefficient when used on aircraf=
t. High exhaust back pressures and overspeeds =0Awith consequent catastroph=
ic failure are not uncommon. In extreme cases, =0Acompressor or turbine whe=
el bursts can cause engine damage and failure due to =0Apart ingestion or o=
il loss. This is not a good idea. Use a turbo which is =0Aprofessionally ma=
tched for your engine and intended use by someone familiar with =0Ahigh alt=
itude turbocharging. =0A=0AIntercooling=0AIntercoolers are heat exchangers =
placed between the compressor and intake =0Amanifold. Their purpose is to r=
educe the temperature of the compressed air =0Abefore it enters the engine.=
 Whenever air is compressed, its temperature =0Aincreases. In the case of a=
 high boost turbo at high altitude, the air exiting =0Athe compressor may r=
each 300-350 degrees F. This lowers the mass flow of air =0Ainto the engine=
 and reduces power. It also raises EGTs and lowers the detonation =0Alimits=
. All of these things are detrimental to performance and engine longevity. =
=0A=0AAn effective intercooler will lower the charge temperature to within =
a few =0Adegrees of the ambient temperature.=0AExhaust Systems=0AThe exhaus=
t system on turbocharged engines is critically important. High =0Atemperatu=
re materials such as stainless steel or Inconel should be used for =0Atheir=
 construction to ensure longevity. Tubing thickness should be a minimum of =
=0A.060 inch to withstand the constant 1500 degree temperatures. Careful at=
tention =0Ato thermal expansion is necessary so that cracking possibilities=
 are minimized. =0ABellows or slip joints are often required at junctions. =
The turbo itself should =0Anever be supported by the headers. A strong, tri=
angulated structure should =0Asupport the weight of the turbo on the engine=
.=0AIf possible, the primary header tubes should be equal length and mandre=
l bent =0Afor low restriction and maximum pulse energy with equal pulse spa=
cing. The =0Aturbine discharge pipe should be 2.25 to 2.5 inches to minimiz=
e backpressure. A =0Amuffler is often unnecessary to reduce exhaust noise a=
s the turbo usually does a =0Agood job of this. This helps to offset the we=
ight of the turbo installation to =0Asome degree.=0AFuture Developments=0AW=
atch the Aircraft section for pictures and updates on turbo installations a=
s =0Athey become available. We are currently flying our turbocharged EJ22 p=
owered =0ARV6A seen elsewhere on this site.=0AR.F.=0A=0AFuel Octane vs. HP=
=0A03/13/98=0AIn turbocharged engines there is a fine balancing act when it=
 comes to making a =0Alot of power on low octane fuel. In most cases, ignit=
ion timing must be retarded =0Aas the boost pressure rises above a critical=
 point and finally there reaches a =0Afurther point where the engine simply=
 loses power. If the timing was not =0Aretarded with increasing boost, dest=
ructive preignition or detonation would =0Aoccur. Normal combustion is char=
acterized by smooth, even burning of the =0Afuel/air mixture. Detonation is=
 characterized by rapid, uncontrolled temperature =0Aand pressure rises mor=
e closely akin to an explosion. It's effects are similar =0Ato taking a ham=
mer to the top of your pistons. =0A=0AMost engines make maximum power when =
peak cylinder pressures are obtained with =0Athe crankshaft around 15 degre=
es after TDC. Experimentation with increasing =0Aboost and decreasing timin=
g basically alters where and how much force is =0Aproduced on the crankshaf=
t. Severely retarded timing causes high exhaust gas =0Atemperatures which c=
an lead to preignition and exhaust valve and turbo damage.=0AWe have a hypo=
thetical engine. It's a 2.0L, 4 valve per cylinder, 4 cylinder =0Atype with=
 a 9.0 to 1 compression ratio and it's turbocharged. On the dyno, the =0Amo=
tor puts out 200hp at 4psi boost with the timing at the stock setting of 35=
 =0Adegrees on 92 octane pump gas with an air/fuel ratio of 14 to 1. We ret=
ard the =0Atiming to 30 degrees and can now run 7psi and make 225hp before =
detonation =0Aoccurs. Now we richen the mixture to 12 to 1 AFR and find we =
can get 8psi and =0A235 hp before detonation occurs. The last thing we can =
consider is to lower the =0Acompression ratio to 7 to1. Back on the dyno, w=
e can now run 10psi with 33 =0Adegrees of timing with an AFR of 12 to 1 and=
 we get 270 hp on the best pull.=0AWe decide to do a test with our 9 to 1 c=
ompression ratio using some 118 octane =0Aleaded race gas. The best pull is=
 490 hp with 35 degrees of timing at 21 psi. On =0Athe 7 to 1 engine, we ma=
nage 560 hp with 35 degrees of timing at 25psi. To get =0Atotally stupid, w=
e fit some larger injectors and remap the EFI system for126 =0Aoctane metha=
nol. At 30psi we get 700hp with 35 degrees of timing!=0AWhile all of these =
figures are hypothetical, they are very representative of the =0Agains to b=
e had using high octane fuel. Simply by changing fuel we took the 7 to =0A1=
 engine from 270 to 700 hp.=0AFrom all of the changes made, we can deduce t=
he effect certain changes on hp;=0ARetarding the ignition timing allows sli=
ghtly more boost to be run and gain of =0A12.5%.=0ARichening the mixture al=
lows slightly more boost to be run for a small hp gain =0Ahowever, past abo=
ut 11.5 to 1 AFR most engines will start to lose power and even =0Aencounte=
r rich misfire.=0ALowering the compression ratio allows more boost to be ru=
n with less retard for =0Aa substantial hp gain.=0AIncreasing the octane ra=
ting of the fuel has a massive effect on maximum =0Aobtainable hp.=0AWe hav=
e seen that there are limits on what can be done running pump gas on an =0A=
engine with a relatively high compression ratio. High compression engines a=
re =0Atherefore poor candidates for high boost pressures on pump fuel. On h=
igh octane =0Afuels, the compression ratio becomes relatively unimportant. =
Ultimate hp levels =0Aon high octane fuel are mainly determined by the phys=
ical strength of the =0Aengine. This was clearly demonstrated in the turbo =
Formula 1 era of a decade ago =0Awhere 1.5L engines were producing up to 11=
00 hp at 60psi on a witches brew of =0Aaromatics. Most fully prepared stree=
t engines of this displacement would have =0Atrouble producing half of this=
 power for a short time, even with many racing =0Aparts fitted.=0AMost fact=
ory turbocharged engines rely on a mix of relatively low compression =0Arat=
ios, mild boost and a dose of ignition retard under boost to avoid =0Adeton=
ation. Power outputs on these engines are not stellar but these motors can =
=0Ausually be seriously thrashed without damage. Trying to exceed the facto=
ry =0Aoutputs by any appreciable margins without higher octane fuel usually=
 results in =0Asome type of engine failure. Remember, the factory spent man=
y millions =0Aengineering a reasonable compromise in power, emissions, fuel=
 economy and =0Areliability for the readily available pump fuel. Despite wh=
at many people think, =0Athey probably don't know as much about this topic =
as the engineers do.=0AOne last method of increasing power on turbo engines=
 running on low octane fuel =0Ais water injection. This method was evaluate=
d scientifically by H. Ricardo in =0Athe 1930s on a dyno and showed conside=
rable promise. He was able to double power =0Aoutput on the same fuel with =
the aid of water injection. =0A=0AFirst widespread use of water injection w=
as in WW2 on supercharged and =0Aturbocharged aircraft engines for takeoff =
and emergency power increases. The =0Awater was usually mixed with 50% meth=
anol and enough was on hand for 10-20 =0Aminutes use. Water/methanol inject=
ion was widely used on the mighty =0Aturbocompound engines of the '50s and =
'60s before the advent of the jet engine. =0AIn the automotive world, it wa=
s used in the '70s and '80s when turbos suddenly =0Abecame cool again and w=
here EFI and computer controlled ignitions were still a =0Abit crude. Some =
Formula 1 teams experimented with water injection for qualifying =0Awith su=
ccess until banned. =0A=0AMy personal experience with water injection is co=
nsiderable. I had several turbo =0Acars fitted with it. One 2.2 liter Celic=
a with a Rajay turbo, Weber carb and no =0Aintercooler or internal engine m=
ods ran 13.3 at 103 on street rubber on pump gas =0Aback in 1987. This was =
accomplished at 15psi. With the water injection switched =0Aoff, I could on=
ly run about 5 psi before the engine started to ping. I think you =0Amight =
see water injection controlled by microchips, catch on again in the coming =
=0Ayears on aftermarket street turbo installations. It works.=0A=0A=A0=0AKe=
lly Troyer=0A"DYKE DELTA JD2" (Eventually)=0A"13B ROTARY"_ Engine=0A"RWS"_R=
D1C/EC2/EM2=0A"MISTRAL"_Backplate/Oil Manifold=0A"TURBONETICS"_TO4E50 Turbo
--0-1081704649-1287340039=:32302
Content-Type: text/html; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

<html><head><style type=3D"text/css"><!-- DIV {margin:0px;} --></style></he=
ad><body><div style=3D"font-family:arial, helvetica, sans-serif;font-size:1=
4pt"><DIV>&nbsp;</DIV>=0A<DIV style=3D"FONT-FAMILY: arial, helvetica, sans-=
serif; FONT-SIZE: 14pt"><BR>=0A<DIV style=3D"FONT-FAMILY: times new roman, =
new york, times, serif; FONT-SIZE: 12pt">=0A<DIV style=3D"FONT-FAMILY: aria=
l, helvetica, sans-serif; FONT-SIZE: 14pt">=0A<DIV style=3D"FONT-FAMILY: ar=
ial, helvetica, sans-serif; FONT-SIZE: 14pt">=0A<DIV style=3D"FONT-FAMILY: =
arial, helvetica, sans-serif; FONT-SIZE: 14pt">=0A<DIV></DIV>=0A<DIV>Sorry =
guys..........Did not think about my "Open Office" files not opening for ev=
eryone !!</DIV>=0A<DIV>Hope this is not too big for the forum !!...........=
........</DIV>=0A<DIV>&nbsp;</DIV>=0A<DIV>=0A<H2>Turbocharging Automotive E=
ngines for Aircraft Applications</H2>=0A<P></P>=0A<P>April 27/98 =0A<P>With=
 automotive engines finding increasing use powering homebuilt aircraft, we =
find many people entertaining the idea of turbocharging to boost the power =
to weight ratios and altitude performance of these engines. We will examine=
 the various facts and myths regarding turbocharging as applied to light ai=
rcraft use.</P>=0A<P><STRONG>History</STRONG></P>=0A<P>Turbos in aircraft w=
ere first successfully applied in mass numbers during WWII. The P-38 and P-=
47 were the best known turbocharged fighter types and both had excellent hi=
gh altitude performance.</P>=0A<P>In general aviation use as applied to air=
 cooled opposed engines, turbocharging has a less revered reputation. Much =
of this has to do with the air cooled engine itself and the fact that most =
of these engines were never designed to be boosted to begin with. They may =
have been modified a bit for the turbo but with power to weight ratios bein=
g a major concern, many are structurally and thermally not up to the task. =
Other concerns such as turbo cooling and lubrication by inferior aviation t=
ype oils also led to premature turbo failures.</P>=0A<P><STRONG>What is a T=
urbocharger?</STRONG></P>=0A<P>A turbo consists of a centrifugal compressor=
 connected to a turbine wheel which is spun at high rpm by the energy of th=
e engine exhaust gasses. It is a very simple device, but the high temperatu=
res and stresses acting on it require exotic materials in the turbine secti=
on which are somewhat expensive. A turbo will cost in the neighborhood of $=
800 to $1300 with the wastegate assembly.</P>=0A<P><BR>Turbine wheels come =
in many shapes and sizes. Constructed of high temperature alloys such as In=
conel or Hastaloy. This is the hot end. =0A<P><BR>Turbine housings are cast=
 from a high nickel/iron alloy. =0A<P>The compressor pressurizes the intake=
 manifold to achieve higher hp or maintains sea level pressure as the aircr=
aft climbs. This permits higher climb rates at altitude and increased cruis=
e speeds.</P>=0A<P><BR>Compressor wheels are cast from aluminum alloy and c=
ome in many sizes or trims. This is the cold end. =0A<P><BR>Compressor hous=
ing is cast from aluminum alloy. Shape is designed to slow the air, thus co=
mpressing it. =0A<P><BR>Cast iron center section or bearing housing contain=
s floating sleeve bearings or more recently ceramic ball bearings. High pre=
ssure oil from engine feeds the bearing and is drained out the bottom of th=
e housing back into the engine. Some housings are water cooled to prevent c=
oking of oil deposits leading to bearing failure. =0A<P><STRONG>Elegant Sim=
plicity</STRONG></P>=0A<P>Turbocharging is the most efficient method of boo=
sting hp with the least weight penalty known. A typical turbo installation =
on a light aircraft including intercooler and exhaust is around 35-50 pound=
s. Such an installation is capable of tripling hp at sea level or adding 50=
% to the naturally aspirated output up to 15,000 feet. Sea level manifold p=
ressures can be maintained to 25,000 feet in some cases. An aircraft capabl=
e of 200 knots at sea level would true out at nearly 300 knots with a good =
turbo system at high altitude.</P>=0A<P>Spinning the compressor with an exh=
aust driven turbine is a much better way than by the mechanical means as in=
 a supercharger. It is lighter, more reliable and more efficient and has th=
e added advantage of having more drive energy available as the aircraft cli=
mbs due to lower backpresure, which is exactly what is needed. </P>=0A<P><S=
TRONG>Fuel Flow vs. Power</STRONG></P>=0A<P>There is a common misconception=
 that turbocharging increases fuel flows because of the backpressure of the=
 turbine. This may be true in air cooled engines using a poorly matched tur=
bo without intercooling because the air cooled engine cannot maintain tempe=
rature stability without adding fuel for cooling purposes. Reduced fuel flo=
ws at altitude in naturally aspirated and turbine engines is not due to som=
e magic process, it is due to the fact that as altitude is increased, less =
power is produced thus less fuel is required. Many people also don't seem t=
o realize that fuel flow is related to hp developed and that a turbo engine=
 will develop more power at altitude than a naturally aspirated engine, obv=
iously requiring more fuel and delivering higher speeds at the same time. <=
/P>=0A<P>Testing done back in the '40s and '50s on turbo engines actually c=
onfirm that fuel flows were LOWER at the same BMEP with turbocharging than =
in a naturally aspirated engine developing the same power. This suggests th=
at when properly applied, the turbo actually recovers more energy from the =
exhaust to be applied to reduce pumping losses during the compression strok=
e than is consumed during the exhaust stroke working against higher backpre=
ssure. With approximately 50% of the energy in the fuel going out the exhau=
st in the form of heat and pressure, turbocharging can recover some of this=
 waste and put it to good use.</P>=0A<P>It was shown that turbos could achi=
eve fuel flows 3-6% lower and turbocompounding could achieve up to 15% lowe=
r fuel flows. Cylinder head and piston cooling were shown to be the primary=
 limitations to achieving lower BMSFC figures. We can conclude from this th=
at as applied to liquid cooled engines with their superior thermal rejectio=
n rates, turbocharging really has few disadvantages compared to the many ad=
vantages it offers.</P>=0A<P>It should be no surprise that the latest certi=
fied liquid cooled engines from Toyota and Orenda are turbocharged. People =
also should remember the awesome turbocompound radials of the '50s. These e=
ngines offered extremely low BSFC and high power. It seems foolish not to t=
urbocharge most good aircraft engines used at any kind of altitude as you a=
re throwing away performance without it.</P>=0A<P><STRONG>Turbocharged Air =
Cooled Engines</STRONG></P>=0A<P>As mentioned earlier, turbocharged , conve=
ntional opposed air cooled engines have a less than stellar track record an=
d many lay people blame the turbo. In fact, the fault lies with the engines=
 it is applied to. The average Lycoming or Continental has a lot of things =
not going for them. Many of these engines suffer premature failure of crank=
cases, cylinders, heads and valves in their turbocharged iterations. Let's =
examine why.</P>=0A<P>Probably the leading cause of case and barrel crackin=
g is the lack of rigidity these engines have in the cylinder/ case area. Fo=
r ease of barrel replacement I assume, the designers of these engines chose=
 to bolt the cylinders individually to the crankcase at the bottom of the c=
ylinder using a flange. This arrangement creates a very willowy structure a=
nd with each firing impulse, combined with the sheer mass of the huge recip=
rocating parts used in these engines, a definite high amplitude, cyclic str=
ess is put on these parts eventually leading to cracking. The increased gas=
 pressures associated with turbocharging compound this problem. This design=
 problem is actually the cause of a major portion of the high vibration lev=
els associated with conventional aircraft engines.</P>=0A<P>The flange meth=
od puts the barrel under a tension load with every firing impulse too which=
 is just plain stupid given the thickness of the cylinders. Modern automoti=
ve racing practice would use through studs to retain the barrels, thus putt=
ing the barrels under compression.</P>=0A<P>Cylinder head cracking and exha=
ust valve burning and sticking are primarily due to insufficient cooling in=
 these areas which is exaggerated by the extra heat introduced by turbochar=
ging. Air cooling is relatively inefficient at sinking heat off around the =
exhaust port and valve seat area. This problem is compounded by the use of =
very low compression ratios resulting in higher EGTs thus even higher heat =
flux in the critical exhaust port area. Typical air cooled aircraft engines=
 run cylinder head temperatures in excess of 400 degrees on a regular basis=
. With aluminum alloys losing roughly 50% of their strength at this tempera=
ture it is easy to see why cracking is all too common here. The high EGT's =
are also detrimental to the life of the turbo's turbine section.</P>=0A<P>V=
alve problems are due to the same cause, temperatures being too high from i=
nefficient heat transfer. Many factory turbo aircraft are forced to either =
open cowl flaps or richen the mixture at high altitude to keep temperatures=
 within limits. With power being maintained at altitude and air density fal=
ling off, there is often insufficient mass flow available for cooling. This=
 is rarely a problem on naturally aspirated engines as power and cooling re=
quirements are dropping off with increasing altitude.</P>=0A<P>Not a lot of=
 design went into the induction system on many of these engines hence fuel =
distribution is usually not the greatest. This contributes to engine roughn=
ess and problems with leaning. Leaning is limited by the mixture in the hot=
test cylinder with the others running richer than desired.</P>=0A<P>The oil=
s used in aircraft engines are relatively crude by modern synthetic automot=
ive standards and consequently not the best for the turbocharger. The use o=
f these oils appears to be a result of the very loose tolerances required i=
n an air cooled engines because of the high operating temperatures which al=
so leads to the high oil consumption characteristic of these engines.</P>=
=0A<P>Finally, many certified installations never used intercooling which i=
s simply amazing. The advantages and necessity of it was clearly understood=
 in WWII. High charge temperatures at altitude required even more fuel for =
cooling and yet more cowl flap opening to control head temperatures.</P>=0A=
<P><STRONG>Turbocharged Liquid Cooled Engines</STRONG></P>=0A<P>Liquid cool=
ed engines are a natural for turbocharging because of their higher heat rej=
ection capabilities. Water cooling also enables the turbo center section to=
 be water cooled which alleviates a primary cause of turbo failure- coked b=
earing holes from high heat and fried oil. Water cooled engines can use mod=
ern synthetic oils because of their tighter clearances. Synthetic oils with=
 their much higher temperature capabilities are more suitable to lubricate =
the turbo than old fashioned, aircraft oils derived from natural base stock=
s. The popular Mobil 1 synthetic oil, common in the racing world, can withs=
tand temperatures of 350 degrees continuously and even 450 degrees intermit=
tently without significant degradation. With proper lubrication and cooling=
, turbochargers will easily last the life of the engine.</P>=0A<P>If we com=
pare the Popular Subaru EJ22 and EJ25 horizontally opposed 4 cylinder engin=
es to their certified counterparts we find many advantages especially in tu=
rbocharged trim. The Subaru has 5 main bearings instead of 3 which means th=
at the crank is better supported. The Subaru has a higher compression ratio=
 for higher thermal efficency. Its overhead cam, 4 valve per cylinder arran=
gement provides lower friction and higher volumetric efficiency than a 2 va=
lve pushrod setup. Finally, the integrated block and head design is many ti=
mes stiffer and stronger than the aircraft engine.</P>=0A<P>As long as the =
radiator and ducting are properly designed, cooling problems are not a conc=
ern on the liquid cooled engine.</P>=0A<P>The induction system on the Subar=
u has proper runner lengths to boost torque, relatively equal airflow to ea=
ch cylinder and can easily use a modern digital, electronic fuel injection =
system to precisely meter fuel. All of these things aid in producing more p=
ower with less fuel.</P>=0A<P><STRONG>Power vs. Weight vs. Manifold Pressur=
e vs. Life vs. Cost</STRONG></P>=0A<P>The EJ engines when turbocharged and =
equipped with a radiator, PSRU and composite propeller are slightly heavier=
 than the popular Lycoming and Continental 360 cubic inch engines which are=
 rated at 180- 200 hp for takeoff and around 135- 150 hp for maximum cruise=
. The Subaru can easily attain 200- 250 hp at 50 inches for takeoff and rel=
iably deliver 140 to 170hp at 35-38 inches for cruise. The 3.3L EG33 six cy=
linder is capable of 300+ hp for takeoff and 200-250 hp in cruise making it=
 a viable alternative to the O-540 engines. These power levels can be maint=
ained during climb and cruise at altitude thanks to turbocharging and liqui=
d cooling.</P>=0A<P>Projected TBO is in the 1000-1200 hour range at this ti=
me. Most people flying the EJ engines are confident that no interim work wi=
ll be required during this period such as cylinder barrel replacement and v=
alve grinding operations which are quite common on aircraft engines before =
reaching their stated TBO. In other words, if you change the oil and check =
the water, that is about all that is required during those 1000 hours. Ther=
e are no magneto or even valve adjustments to worry about. Expect spark plu=
g replacement every 100- 250 hours or so at $3 each. Overhaul costs will ra=
nge from about $500 to $1500 for parts and machine work depending on condit=
ion. Labor would be in the $500 to $1200 range. We are all familiar with th=
e costs to overhaul a certified aircraft engine. =0A<CENTER></CENTER>=0A<P>=
</P>=0A<P><STRONG>Matching a Turbocharger</STRONG></P>=0A<P>As applied to t=
he popular Subaru EJ22 and EJ25 engines for general purpose use below 15,00=
0 feet we can narrow the choice of turbos down a bit. Garrett T3 turbos are=
 recommended as they are relatively cheap, reliable and easy to fix with a =
vast array of combinations available. The compressor side may be either a -=
50, -60 or super 60 wheel with a stage 3 turbine wheel in a .82 A/R housing=
, depending on the exact use and hp. These may be ordered with an integral =
wastegate and a water cooled bearing housing which are both highly recommen=
ded. These turbos are quite capable of producing enough boost to extract 25=
0 hp at sea level and still have 175+ hp available at 15,000 feet.</P>=0A<P=
>The integral wastegates are generally more reliable than an external type =
especially on an aircraft application in which it is almost constantly bypa=
ssing exhaust. The wastegate is simply a valve used to bypass exhaust gasse=
s around the turbine to control turbine and compressor speed and thus boost=
 pressure. It is usually actuated automatically by a diaphragm sensing mani=
fold pressure.</P>=0A<P>Many people have tried to use the OE turbo on auto =
aircraft conversions, often with limited success. The turbos on cars were n=
ot matched to operate at high altitude and continuous high power settings s=
o they are mismatched and inefficient when used on aircraft. High exhaust b=
ack pressures and overspeeds with consequent catastrophic failure are not u=
ncommon. In extreme cases, compressor or turbine wheel bursts can cause eng=
ine damage and failure due to part ingestion or oil loss. This is not a goo=
d idea. Use a turbo which is professionally matched for your engine and int=
ended use by someone familiar with high altitude turbocharging. =0A<P><STRO=
NG>Intercooling</STRONG></P>=0A<P>Intercoolers are heat exchangers placed b=
etween the compressor and intake manifold. Their purpose is to reduce the t=
emperature of the compressed air before it enters the engine. Whenever air =
is compressed, its temperature increases. In the case of a high boost turbo=
 at high altitude, the air exiting the compressor may reach 300-350 degrees=
 F. This lowers the mass flow of air into the engine and reduces power. It =
also raises EGTs and lowers the detonation limits. All of these things are =
detrimental to performance and engine longevity. </P>=0A<P>An effective int=
ercooler will lower the charge temperature to within a few degrees of the a=
mbient temperature.</P>=0A<P><STRONG>Exhaust Systems</STRONG></P>=0A<P>The =
exhaust system on turbocharged engines is critically important. High temper=
ature materials such as stainless steel or Inconel should be used for their=
 construction to ensure longevity. Tubing thickness should be a minimum of =
.060 inch to withstand the constant 1500 degree temperatures. Careful atten=
tion to thermal expansion is necessary so that cracking possibilities are m=
inimized. Bellows or slip joints are often required at junctions. The turbo=
 itself should never be supported by the headers. A strong, triangulated st=
ructure should support the weight of the turbo on the engine.</P>=0A<P>If p=
ossible, the primary header tubes should be equal length and mandrel bent f=
or low restriction and maximum pulse energy with equal pulse spacing. The t=
urbine discharge pipe should be 2.25 to 2.5 inches to minimize backpressure=
. A muffler is often unnecessary to reduce exhaust noise as the turbo usual=
ly does a good job of this. This helps to offset the weight of the turbo in=
stallation to some degree.</P>=0A<P><STRONG>Future Developments</STRONG></P=
>=0A<P>Watch the Aircraft section for pictures and updates on turbo install=
ations as they become available. We are currently flying our turbocharged E=
J22 powered RV6A seen elsewhere on this site.</P>=0A<P>R.F.</P></DIV>=0A<DI=
V>&nbsp;</DIV>=0A<DIV>=0A<H2>Fuel Octane vs. HP</H2>=0A<P></P>=0A<P>03/13/9=
8</P>=0A<P>In turbocharged engines there is a fine balancing act when it co=
mes to making a lot of power on low octane fuel. In most cases, ignition ti=
ming must be retarded as the boost pressure rises above a critical point an=
d finally there reaches a further point where the engine simply loses power=
. If the timing was not retarded with increasing boost, destructive preigni=
tion or detonation would occur. Normal combustion is characterized by smoot=
h, even burning of the fuel/air mixture. Detonation is characterized by rap=
id, uncontrolled temperature and pressure rises more closely akin to an exp=
losion. It's effects are similar to taking a hammer to the top of your pist=
ons. </P>=0A<P>Most engines make maximum power when peak cylinder pressures=
 are obtained with the crankshaft around 15 degrees after TDC. Experimentat=
ion with increasing boost and decreasing timing basically alters where and =
how much force is produced on the crankshaft. Severely retarded timing caus=
es high exhaust gas temperatures which can lead to preignition and exhaust =
valve and turbo damage.</P>=0A<P>We have a hypothetical engine. It's a 2.0L=
, 4 valve per cylinder, 4 cylinder type with a 9.0 to 1 compression ratio a=
nd it's turbocharged. On the dyno, the motor puts out 200hp at 4psi boost w=
ith the timing at the stock setting of 35 degrees on 92 octane pump gas wit=
h an air/fuel ratio of 14 to 1. We retard the timing to 30 degrees and can =
now run 7psi and make 225hp before detonation occurs. Now we richen the mix=
ture to 12 to 1 AFR and find we can get 8psi and 235 hp before detonation o=
ccurs. The last thing we can consider is to lower the compression ratio to =
7 to1. Back on the dyno, we can now run 10psi with 33 degrees of timing wit=
h an AFR of 12 to 1 and we get 270 hp on the best pull.</P>=0A<P>We decide =
to do a test with our 9 to 1 compression ratio using some 118 octane leaded=
 race gas. The best pull is 490 hp with 35 degrees of timing at 21 psi. On =
the 7 to 1 engine, we manage 560 hp with 35 degrees of timing at 25psi. To =
get totally stupid, we fit some larger injectors and remap the EFI system f=
or126 octane methanol. At 30psi we get 700hp with 35 degrees of timing!</P>=
=0A<P>While all of these figures are hypothetical, they are very representa=
tive of the gains to be had using high octane fuel. Simply by changing fuel=
 we took the 7 to 1 engine from 270 to 700 hp.</P>=0A<P>From all of the cha=
nges made, we can deduce the effect certain changes on hp;</P>=0A<P>Retardi=
ng the ignition timing allows slightly more boost to be run and gain of 12.=
5%.</P>=0A<P>Richening the mixture allows slightly more boost to be run for=
 a small hp gain however, past about 11.5 to 1 AFR most engines will start =
to lose power and even encounter rich misfire.</P>=0A<P>Lowering the compre=
ssion ratio allows more boost to be run with less retard for a substantial =
hp gain.</P>=0A<P>Increasing the octane rating of the fuel has a massive ef=
fect on maximum obtainable hp.</P>=0A<P>We have seen that there are limits =
on what can be done running pump gas on an engine with a relatively high co=
mpression ratio. High compression engines are therefore poor candidates for=
 high boost pressures on pump fuel. On high octane fuels, the compression r=
atio becomes relatively unimportant. Ultimate hp levels on high octane fuel=
 are mainly determined by the physical strength of the engine. This was cle=
arly demonstrated in the turbo Formula 1 era of a decade ago where 1.5L eng=
ines were producing up to 1100 hp at 60psi on a witches brew of aromatics. =
Most fully prepared street engines of this displacement would have trouble =
producing half of this power for a short time, even with many racing parts =
fitted.</P>=0A<P>Most factory turbocharged engines rely on a mix of relativ=
ely low compression ratios, mild boost and a dose of ignition retard under =
boost to avoid detonation. Power outputs on these engines are not stellar b=
ut these motors can usually be seriously thrashed without damage. Trying to=
 exceed the factory outputs by any appreciable margins without higher octan=
e fuel usually results in some type of engine failure. Remember, the factor=
y spent many millions engineering a reasonable compromise in power, emissio=
ns, fuel economy and reliability for the readily available pump fuel. Despi=
te what many people think, they probably don't know as much about this topi=
c as the engineers do.</P>=0A<P>One last method of increasing power on turb=
o engines running on low octane fuel is water injection. This method was ev=
aluated scientifically by H. Ricardo in the 1930s on a dyno and showed cons=
iderable promise. He was able to double power output on the same fuel with =
the aid of water injection. </P>=0A<P>First widespread use of water injecti=
on was in WW2 on supercharged and turbocharged aircraft engines for takeoff=
 and emergency power increases. The water was usually mixed with 50% methan=
ol and enough was on hand for 10-20 minutes use. Water/methanol injection w=
as widely used on the mighty turbocompound engines of the '50s and '60s bef=
ore the advent of the jet engine. In the automotive world, it was used in t=
he '70s and '80s when turbos suddenly became cool again and where EFI and c=
omputer controlled ignitions were still a bit crude. Some Formula 1 teams e=
xperimented with water injection for qualifying with success until banned. =
</P>=0A<P>My personal experience with water injection is considerable. I ha=
d several turbo cars fitted with it. One 2.2 liter Celica with a Rajay turb=
o, Weber carb and no intercooler or internal engine mods ran 13.3 at 103 on=
 street rubber on pump gas back in 1987. This was accomplished at 15psi. Wi=
th the water injection switched off, I could only run about 5 psi before th=
e engine started to ping. I think you might see water injection controlled =
by microchips, catch on again in the coming years on aftermarket street tur=
bo installations. It works.</P></DIV>=0A<DIV><BR>&nbsp;</DIV>=0A<P>Kelly Tr=
oyer<BR><STRONG><FONT size=3D4>"DYKE DELTA JD2"</FONT> <FONT size=3D1>(Even=
tually)</FONT></STRONG></P>=0A<P>"13B ROTARY"_ Engine<BR>"RWS"_RD1C/EC2/EM2=
<BR>"MISTRAL"_Backplate/Oil Manifold</P>=0A<P>"TURBONETICS"_TO4E50 Turbo</P=
>=0A<DIV><BR></DIV>=0A<DIV style=3D"FONT-FAMILY: arial, helvetica, sans-ser=
if; FONT-SIZE: 14pt"><BR>&nbsp;</DIV></DIV></DIV></DIV></DIV></DIV></div></=
body></html>
--0-1081704649-1287340039=:32302--