First installment of 4 blogs on turbos.....
Marc Wiese

--------------------------------------------------------

From: George Braly [GAMI/TAT]
Subject: Turbonormalizing vs Turbocharging

[Chief Engineer at GAMI of GAMIjector fame, and also at Tornado Alley
Turbo  ]

Understanding the important differences

TN and TC engines share the following characteristics:

Both involve compressing outside ambient air to make it available as
denser air to the engine induction system so that the engine can make
more power. The degree to which the air is compressed is just a matter
of design choice when one decides to specify the engine and its ability
to make horsepower across the normal range of desired altitudes.

Any time you compress air - you make it hotter. As discussed below -
almost everybody in the early days missed the importance of
intercoolers. Both TN and TC systems use an exhaust driven turbine to
drive a centrifugal compressor - at speeds that may range as high as
120,000 RPM on some units. Both TN and TC systems can use the same turbo
components. The TN system designer typically starts with a normally
aspirated engine and asks the following question:

What is the minimal number of changes one can make to this existing NA
engine and still have that engine make sea level power at altitude
(typically > 18,000 feet)?

The answer includes a laundry list like this:

>New exhaust system to get the exhaust to the turbo housing  >Add turbo
charger or two of them  >Add some way to control the speed of the turbo
charger i.e. a wastegate.  >Manual or automatic control over the wastegate?
>Intercooler? Very desirable, but not absolutely mandatory. >Fuel
injectors that have the air side referenced to the output of the
turbocharger compressor.
>Fuel pump and fuel metering changes that will let the engine work
gracefully with the higher pressure air over a broader range of
operating conditions.
>Need to enhance the cooling of the overall system (both cylinder head
and possibly oil cooling), since it will be flying in thinner air at
altitude that inherently provides less cooling.
>Mechanically support the weight of the turbochargers.

The reality is that the TC system designer actually started with the
same question. Keep in mind that almost all of the fundamental design
decisions for the TCM TSIO series of engines and the Lycoming TIO series
of engines were made back in the 1963 to 1968 time frame.

In 1963, at TCM, the question was something like this: "What can we do
to modify the existing 260 Hp IO-470, or the new 285 Hp IO-520 normally
aspirated engines - to make a turbocharged engine to sell to Cessna?"

History note: The first C-320 aircraft (in 1964) were equipped with
TSIO-470 engines. A couple of years later, they switched to TSIO-520
engines with 285 HP, just like the Cessna 310s were also switched to
IO-520 normally aspirated engines.

Another history note: Remember, Jack Riley was running amok back then
putting Lycoming normally aspirated IO-540s on C-310's with
turbochargers and operating them as "turbo normalized" engines and
calling them Riley Rockets! TCM just HAD to come up with a turbocharged
engine or they ran the risk that Cessna might go to Lycoming for engines
for the C-310. The term "turbonormalized" was not used, until later.

And guess what? TCM ended up developing exactly the same basic list of 9
items (see above) as for the TN engine - but with one additional item.

Some more history: The then universe of general aviation piston engine
engineers mostly had a background in the big supercharged radial engines
- all of which had very low compression ratios (often down around
6.5:1). All of those engineers just assumed that they would have to
reduce the compression ratio on their new turbo-supercharged flat six
cylinder engines in line with their previous big radial engine experience.

So that is exactly what they did. TCM reduced the CR from 8.5:1 down to
7.5:1. Lycoming typically went from around 8.7:1 down to 7.3:1. There
were some variations.

So, now, take the list of nine items above, and add item number 10.

>Change the piston geometry to reduce the compression ratio by about 1
point, from ~ 8.5 down to 7.5:1. When you do that you have identified
the only meaningful "difference" in the hardware between our common
terminology of turbonormalizing and turbocharging.

Yes, it is true that the typical turbonormalizer on a 285 HP IO-520 TCM
engine only uses around 30" of MP and the typical early turbocharged
engine (TSIO-520D in the 1968 Cessna 320F, for example) uses around
32.5" Hg to make the SAME 285 HP at sea level.

However, bear in mind, the extra 2.5" of manifold pressure (from 30" up
to 32.5" Hg) is there to make up for two things: First, some minor
losses in the exhaust system back pressure and second, and more
important, to make up for the fact that the air was HOTTER in the
non-intercooled induction system in the Cessna 320 and was therefore
"LESS DENSE" and it took higher MP to get the same total mass air flow
through the engine - in order to make the same 285 HP - and to make up
for the loss in overall efficiency because of the lower compression ratio.

Important note: None of those early turbocharged or turbonormalized
engines had intercoolers! Not the Cessna T-210. Not the Cessna 320. Not
the Cessna 310T. Not the Bonanza V35TC. Not the Bonanza A36TC nor the
Bonanza B36TC. None of them.

There were, in the early days, a number of RAY-JAY "turbo normalized"
after market STCs that were added to several Mooneys and to some
Bonanzas. None of them worked well. Most were really bad. That caused
some early heart burn with the OEM's who condemned those early
non-intercooled non-OEM 8.5:1 compression ratio installations - and
often for good reason. Had those early aftermarket STC'd installations
with the 8.5:1 compression ratios been equipped with good intercoolers
and good cylinder and oil cooling, the valid engineering reasons to
object to them would have been eliminated, although, undoubtedly,
Lycoming and TCM would have still found an excuse to complain, just for
marketing reasons.

Another history note: For a while, the Piper factory did do
installations of what we now call turbonormalized engines using the
RAY-JAY hardware. There were no intercoolers. This was done on some
Comanche and Twin Comanche aircraft and some Apache aircraft with
Lycoming IO-540 engines. Then Lycoming changed to low compression
pistons and provided an OEM engine to Piper as a TIO-540x series engine.
Again, there were no intercoolers. Forty years later, the earlier,
higher compression configuration engines are, today, much preferred by
knowledgeable pilot-owners, as opposed to the later models with the
typical 7.3:1 compression ratio engine configurations.

If, in the years around the 1963 time frame, TCM (and, later, Lycoming)
engineers had readily available cost effective air-to-air heat
exchangers (intercoolers) and had the airframe OEM companies then been
willing to go to the trouble and expense to integrate the somewhat more
complex design required to employ the intercoolers effectively, it is
likely that some engineer would have realized that it would be "a good
thing" if the early turbocharged engines used in the Cessna T-210 and
Cessna 320 and the Piper Comanche would have kept their original 8.5
(8.7):1 compression ratio engines. Thus, the several problems (mostly
heat and temperature and fuel consumption) associated with 7.5:1
compression ratio engines would have largely been avoided.

But the OEMs were in a hurry for a quick fix - or a "drop in" turbo
engine that would be a "one design" that would fit in a broad cross
section of existing Cessna and Piper air frames without extensive
airframe modifications. That was a lot easier to do by changing the
piston to get a 7.5:1 compression ratio than it was to keep the 8.5:1
compression ratio and figure out a way to stuff a good intercooler into
the multiple different cowlings that existed among dozens of different
airframes.

A current technology "good" intercooler design can reduce the induction
air temperature so that turbocharged or turbonormalized engines see
induction air temperatures that are much more comparable at all
altitudes and power settings to a hot day normally aspirated engine than
they are to a non-intercooled turbocharged or supercharged engine. But
nobody back in the 1963-1970 timeframe was properly focused on the
importance of the intercooler to optimize and make truly successful the
highly desirable change from normally aspirated engines to turbocharged
engines.

The routine absence of intercoolers in the design of the early
turbocharged (and early turbo-normalized) engines in the 1960s resulted
in a series of unsatisfactory engine life operational experiences for
the owners. Those unsatisfactory experiences became widely circulated as
the prevailing "wisdom" of the time. Those old perceived "truths" are
very hard to correct in the often tradition bound aviation community.

[Keep in mind that no one had engine analyzers back then, either, to
really recognize if they had a cooling problem that needed fixing.  They
found out when a piston holed through, but they still couldn't be sure
what to change or fix, besides the piston.  Paul]

Intercoolers are Really Important

Example: On a hot day (100 F) day at Denver, the TN SR 22 has a
calculated full power induction air temperature of ~ 125 to 130d F.

Many people think the presence of the red hot glowing turbocharger
turbine section must cause a large rise in the induction air
temperature. But because the mass air flow through the compressor is so
large (think in terms of something short of a TON of air per hour) the
temperature rise due to the proximity of the turbocharger is relative
small. It gets lost in the calculations it is so small.

As an aside, the actual measurement of the induction air temperature on
the TN SR 22 agrees with the calculations to a very close margin (+- 5
degrees). Note, even a normally aspirated aircraft on a 100 F day at
Denver will have elevated induction air temperatures, around 110 F, just
due to the heat rise in that simple induction system.

But, if you take away the specially designed (not at all like the
standard TCM intercooler design) intercoolers from the calculation you
get something like 155 to 160 F - even using the extraordinarily
efficient compressors that are on the TN SR 22.

If the compressors used were not well matched on the compressor maps to
the specific aircraft engine application, then that number would go much
higher: Something like 170 to 190 F.

[Again, each 10 degrees increase requires one more number octane, and
reduces power output 1%.  Paul]

Those kinds of numbers are difficult to deal with. That is why carefully
selecting aircraft compressors that have compressor maps that put the
operating point near the center of the "island" of maximum efficiency on
the compressor maps and designing and using highly efficient
intercoolers are all essential ingredients in the "SMART TURBO"(R) which
Cirrus decided to bring to market.

Many people only think of detonation in terms of fuel octane or in terms
of spark timing. But people knowledgeable in the industry know that
detonation is strongly influenced by still another factor - - induction
air temperature. There is lots of good data on that subject, going back
to WWII.

A couple of more points. Even the TN SR 22 is not "fully turbo
normalized". By that is meant, that on hot days, the turbos fully make
up for the loss of ambient pressure as one goes to altitude, but without
a controller that can compensate for higher than standard outside
temperatures, then while the turbo normalized engine does very well in
horsepower, on hot days it will not make full rated HP unless something
more were done to fully compensate for BOTH ambient low pressures and
ambient high temperatures. This requires a very expensive
hydro-mechanical system to fully compensate for density and the only
airplane I know of that even attempts to do that is the Piper Navajo
Chieftain.

Last, the pilots and mechanics should really be aware that there is
nothing "magic" about 29.9" Hg manifold pressure verses 32.5" Hg
manifold pressure or even 36 or 38" Hg manifold pressure. One is not
necessarily "harder" on the engine than is the other. The engineering
parameters that "make a difference" are evident in other data that the
pilot never sees. The important constraints on good engine durability
are only evident in the data from the individual combustion events and
that is primarily the magnitude of the maximum instantaneous internal
cylinder pressure and its occurrence with respect to top dead center of
piston travel, along with the magnitude of the operating cylinder head
temperature.

If a design engineer properly manages the instantaneous peak internal
cylinder pressure (which is a VERY different parameter from the more
widely known BMEP) and the cylinder head temperatures, and the induction
air temperature, then the magnitude of the manifold pressure becomes
very much less important in a discussion of engine durability.

Keep in mind, we have learned over the last 15 years that it really is
not how hard you run the engine, but rather, how you run the engine hard.

If you have comments, questions, or criticism about this article, please
contact gwbraly@gami.com


Curtis Sanford wrote:

Looking at your Denver 100 F day example, I was playing with the ideal
gas law. Air density goes from 1 bar at sea level to .85 bar at Denver.
So just compressing the air would increase the (Rankine) temperature by
1/.85. 100F is 560R, so 1/.85*560R=659R=198F. That seems a lot worse
than the 155-160F you cite. Have I got the wrong math?

[Yup, Curtis, you do!  :-)  Paul]

George Braly wrote:

I lived through most of that. I can remember being 17 years old and
reading the magazines about the "Riley Rocket" . A few years later, I
flew a subsequent updated version of the Riley Rocket with Jack Riley
himself, when he was trying to get us to convert my C-320 over to
Lycomings with his turbos. I flew it - - and declined.

In any event, I decided to try to put some history together with the
engineering and pull it all together to understand how we got from
"there" to "here."

Re, your math...

In a word, yes...

Thermo was always a tough class for me, too!

As the captain said in the Hunt for Red October: "I'm a little rusty on
my Morse Code..."

But here goes, anyway:

(T2/T1) = (P2/P1) EXP (g-1/g) (Isentropic Gas Law)

Where g = gamma = ratio of the specific heats of air = ~ 1.4Cp/Cv for
temperatures in the 100 to 200 d F range.

Where Cp = specific heat capacity of a gas (air) at constant pressure
and Cv = specific heat capacity at constant volume.

When you get the ideal increase in temperature - - then divide that
"delta T" by the compressor efficiency and add it back to the original
ambient temperature to get the "real" Compressor Discharge Temperature
(CDT).

The compressor efficiency of the two tiny Malibu / Mirage turbos in the
TN Cirrus is up in the low 0.71 to 0.75 range.

The compressor efficiency of good bolt-on belt/gear driven automotive
turbo being driven at 1.5 X the typical (slow aircraft) crankshaft
speeds, is, I suspect, rather low.

Probably down around .56 to .62.

The intercooler efficiencies are up in the 0.64 to 0.68 range. The rest
of it is just number crunch combined with some experience as to how
things really work out in the "real world."

But you really want the Induction Air Temperature (IAT). Without
intercoolers, the IAT = CDT.

Intercooler efficiency n = (CDT-IAT) / (CDT-OAT)

Regards, George