> >Paul,
 > >
 > >I believe this turbo, the Garrett GT-2860R with the 76 trim and optional
 > >0.86 A/R housing might be just what we are looking for in our aviation
 > >uses. I`m considering it as the turbine for the turbocompound test
 > >engine use but perhaps it might be a good choice for Dave`s plane as
 > >well.
 > >The bigger A/R will slow the wheel down but provide more torque at
higher
 > >rpms.
 > >
 > >The high content Ni body and Inconel turbine wheel will handle the heat
 > >better than the standard IHI turbochargers from Mazda which are, lets
 > >face it, 20 year old technology. The Garrett in addition has ball
 > >bearings, is compact, has a built in waste gate and has both water and
 > >oil cooled jackets. Whats not to love? Beside the price we will have
 > >to drill the threads out of the holes. I don`t know if the lack of a
 > >split turbine mouth will an issue I imagine it depends on the exhaust
 > >manifold design.
 > >
 > >No info on weight unfortunately.
 > >
 >
 >http://www.turbobygarrett.com/turbobygarrett/catelog/Turbochargers/GT28/GT2860R_707160_5.htm
 > >
 > >Doug in Japan
 >
 > I'll study it Doug. And get back to everybody on it. .86 A/R is not
all that big.
 > Could be bigger. I have seen them with over one. If it is a
automotive turbo
 > there is excess HP built in to accelerate. We are going to need an
unusual
 > combination. Might be OK for turbo compound as the engine will slow
the turbine.
 >
-----------------------------------------------------------------------------
 >
 > A quick thought.
 >
 > Lets separate turbo compounding from turbo charger for the moment.
 >
 > For turbo charger the diverted valve manifold and turbo tach is the
way to go as
 > you can use any turbo charger. You set a turbo RPM redline and the
diverter
 > takes care of holding that. Manual or automatic. Boost is whatever
you get
 > at that turbo RPM. If you are feeling rich you can up the turbo RPM
red line
 > and risk a turbine failure.
 >
 > Apparently what we have gathered so far is there is so much kinetic
energy in the
 > rotary exhaust it tends to ignore more or less built in waste gates.
 >
 > Got to run. Be back later today.
 > --
 > Paul Lamar ...No rotor no motor.
 >
 >
 >
 > Paul,
 >
 > Yes, in the GT 3482R series they come without a waste gate, and two
 > options for the exhaust pipe mounting. Unfortunately they don`t mention
 > a Inconel turbine as an option. This model is recommended for engine
 > displacements from 2000 to 6000 cc. One option has an turbine A/R of
 > 1.06. Turbine wheels are all 68.0mm dia. and 84 trim. I definitely
 > agree with you that we need larger wastegates so that when flying we can
 > isolate the turbine completely. Whether that mean enlarging the waste
 > gate hole on the built-in models, making a complete prior by-pass system
 > as you have shown or a combination of the two.
 >
 > BTW, I made a Rhino drawing of the spin muffler turbo mount system,
 > saved it as JPEG file and now I can`t find it in the computer. Hmmmm?
 >
 > Doug in Japan



 > Not a waste gate. A motor controlled diverter valve.
 > Once you have that device you can experiment with turbo's until
 > we find one that gives the most boost and still lives.
 >
 > Turbine 101.
 > Peak turbine torque occurs at zero-RPM.
 > Zero torque occurs at max no-load-RPM.
 > Max HP occurs at .5 X max no-load-RPM.
 > All RPM's are the turbine RPM. Not the engine RPM.
 > Boost will be proportional to turbine HP.
 >
 > 1) Given a constant R or turbine diameter and a given
 > exhaust gas flow volume.
 >
 > Larger A:
 >
 > Lower exhaust gas velocity.
 >
 > Lower peak torque.
 >
 > Lower max no-load-RPM.
 >
 > Lower max turbine HP.
 >
 > Lower back pressure on engine.
 >
 >
 > 2) Given a constant A and variable turbine diameter and a given
 > exhaust gas flow volume.
 >
 >
 > Larger R:
 >
 > Same exhaust gas velocity.
 >
 > Higher peak torque due to longer moment arm.
 >
 > Lower max no-load-RPM due to longer moment arm and
 > circumferential velocity being the same.
 >
 > The same max turbine HP ????. Higher torque lower RPM.
 > (Better for our purposes?)
 >
 > Same back pressure on engine.
 >
 > The key to figuring out what is going on is turbine diameter
 > >>>> AND <<<< A/R. These are the critical parameters as far as
 > turbine HP is concerned.
 >
 > Next question? HP consumed (required) and boost as a function of
 > compressor diameter and A/R.
 >
 > Paul Lamar ...No rotor no motor.

Doug in Japan wrote:  "we need larger wastegates"

Paul Lamar wrote:  ".86 A/R is not all that big.   Could be bigger. I
have seen them with over one. If it is a automotive turbo
 there is excess HP built in to accelerate" and "The key to figuring out
what is going on is turbine diameter
 > >>>> AND <<<< A/R. These are the critical parameters as far as
turbine HP is concerned."

It sounds like both of you want to be able to vary the sizes of all the
components of the turboCHARGING (since this is the turbocharging
thread) system, to come up with the optimum combination for
turbocharging the engine.

We know that turbochargers are already built with variable _vanes_, that
pass variable amounts of air to the turbine blades.  Let's create what
Mr. Lamar suggests up in this thread: " . . . _variable turbine
diameter_ . . . " (Suggestion #2).

What do I mean?

I mean, why not make a turbocharger with varies _everything_:
 -  the amount of exhaust that flows into the turbocharger
 -  the diameter of the turbine
 -  the size of the turbocharger _housing_(if that is what you meant,
Paul, by "variable diameter")
 -  the size of the wastegate
so that all the variables can be tested together?

Now, if building this kind of turbocharger is too expensive, is there a
way we can do it digitally?  Build in the computer a digital model of
the turbocharger, the turbines, the wastegate, and the airflows through
all of them, so that we can vary each component, or all of the
components together, until we arrive at an optimum solution?

Les Nordman :-)

The variable vane turbo chargers are ideal no doubt about it.
They are just not readable available. I could be wrong. I may be looking
in the wrong places.

If you are talking about a computer model I think that can be done with
a spreadsheet. Spread sheets have been done on turbo chargers but not integrated
with variable A/R and all the other parameters in the engine turbo charger system.
I just don't have time to work on it yet.


My immediate focus is talking Dave Leonard into a turbo tach so we have
some hard and fast data to work with.

Also into building a diverter valve/manifold so we don't lose another
turbocharger. Probably cheaper than an HKS manifold if it can be found
at all. I have a couple of turbo charger lying around I could loan
Dave if he had such a manifold and tach. One of which is a TO4 I bought
from Dave. I just don't want to destroy them for no purpose.

BTW I forgot to add a statement to Turbine 101.
Relative gas velocity to blade speed is zero at max no-load-
turbine-RPM. In other words the blade tangential speed and
the exhaust gas velocity are the same.

Paul Lamar

Paul

Also into building a diverter valve/manifold so we don't lose another
turbocharger. Probably cheaper than an HKS manifold if it can be found
at all. I have a couple of turbo charger lying around I could loan
Dave if he had such a manifold and tach. One of which is a TO4 I bought
from Dave. I just don't want to destroy them for no purpose.

BTW I forgot to add a statement to Turbine 101.
Relative gas velocity to blade speed is zero at max no-load-
turbine-RPM. In other words the blade tangential speed and
the exhaust gas velocity are the same.

Paul Lamar

Paul,

Yes I meant exhaust divertor not wastegate exclusively.  Do you have a
picture of the HKS manifold?  Maybe I can find one over here.

 I have my doubts every turbine puts out maximum hp at "0.5 X the no
load rpm."  I wish it were that easy:)   I think we need a turbine dyno
for any of this stuff in any case and  since the turbine is fixed we
have to let the load, (the compressor housing) most likely be free to
rotate a bit to gather any torque data. With an over sized compressor we
can throttle the output and increase the load on the turbine causing it
to run slower stop and then what, start to bake:)?

Les, Yep the variable vanes turbo like BORG WAGNER makes for the Porche
are sweet but not without troubles. The unit is too small for our use.
This technology,  was originally `borrowed` from Garrett Corp. but
later, through loss of a patent infringement claim, it cost B.W.  a tidy
sum.  The vanes, which don`t really change the size of the turbine,  but
the angle at which the hot gases contact the blades, tend to get carbon
deposits on them and stick.  Perhaps variable vanes could be said to
change to be like a variable  A/R ratio of the turbine chamber. Garretts
new sliding plate version, which is in their labs but apparently not for
sale yet, appears to be simpler to make and perhaps more robust. It
sounds like it has one circular moving plate instead of all those vanes,
connecting shafts, linkages and circular drive plate.

Just so as to not to confuse the issue, my interest is in the turbine
only for turbo compound experimental use.  The compressor data is
interesting only it in tells me what type of `load` the compressor is
drawing. As Paul has often said this tells us what the turbine is
supplying but not it full capabilities.

Doug in Japan


It is just a rule of thumb. All turbines have high torque when they
are stalled. That is when the air velocity relative to the blade
is a maximum. True, the blade might be stalled but we are not operating
there anyway. You should have a bunch of curves on your wind generator.

It is also  handy to make that simplification when sizing turbines.
That is what happens when you assume the torque is a straight
line function of RPM. HP peaks around half max turbine RPM.
It is not exactly half but it is close enough to understand
how turbine work. See the attached chart. Good enough to get us in
the ball park. Right at the moment we are a couple of miles away :)