Re: Correcting for 80% VE - just more questions then

["Lawson, Dave" <dlawson@ball.com>]

Good questions Randall, I will give it a shot,
>
1. Why?  Are you saying that a larger turbo is less efficient at
producing boost than a smaller turbo?  Correct me if I am wrong, but I
thought that larger turbos were more efficient at making large
quantities of boost and smaller turbos spool up faster at the loss of
top end boost.  One might make the assumption that if a smaller turbo is
better then smallest is better yet.  I know that is not true.  SO why
would a smaller RS2 make more dense air than a larger more efficient at
generating high boost turbo?
>
Your assumption is that larger turbos are more efficient at making large
quantities of boost. A somewhat valid one, but this is where the turbo 
compressor maps need to come out of the file folders. I don't have the
specific audi K26 map, but do have the 944turbo K26 map and the 930
K27 map. But any map will do, look at Turbo magazine, they have
published
some maps when upgrading the eclipse project car.

On the map, the x axis is air flow(cfm or m^3/s) and the y axis is
pressure 
ratio, Pr. And the map itself is like a geographic topo map where the
series of 
contour lines depict altitude. On the compressor map these are turbo 
efficiency lines. I think of the compressor map as a 3-d surface and the
goal
is to climb to the highest part of the map and stay there. This gives
the  
greatest turbo efficiency and therefore heats the air the least when it
is being 
compressed.

The problem arises when you are dealing with real life and running an
engine
over the rpm range which changes the airflow which the engine is using.
Obviously
the engine will be using less air at say 2500 rpm than 5500 rpm, so we
are operating
in a different region of turbo efficiency. But we are lucky that the
5k/200 cars are 
using the ECU to control boost pressure or the PR. This is great because
PR has
an input to the total air flow ie. 173 * 2.79 = 482CFM @ 5500rpm. In my
mind
the smart thing to do is to work with engine speed and Pr and try to
achieve
a somewhat constant engine CFM which keeps the turbo operating in the
high
efficiency region. This would be my desired boost profile. This is a
recursive 
process, you start with an engine speed and PR, get the turbo
efficiency, ask
the question is there a better Pr to use at this engine speed and is it
possible(?),
pick better Pr, look up turbo eff, do the math, look at desity ratio,...
This process
is repeated for each engine speed.

You can do this exercise with 2 turbo maps, one for a smaller turbo and
one for 
a larger turbo. For each engine speed, look up the turbo efficiency for
each turbo.
Most likely for the larger turbo, the engine just doesn't displace
enough volume of
air at any pressure ratio to get into the high eff part of the map until
high engine 
speeds. During this whole exercise, you need to watch other
characteristics of the 
turbo, such as turbine speed and not to be in the surge region. But I
will let someone
else talk about those issues.

I hope this explanation helps, once you get the basic understanding of
what is happening,
it's really just turning the crank and getting values for other Pr and
engine speeds.

Now we ask, big turbo flowing lots of air and PR which gives good
efficiency, whats wrong
with that? There are other engine design parameters which get in the
way, namely intake
runner sizing and inlet valve sizes. Think of the garden hose eaxmple,
no matter how much
pressure or water is backed up on the hose inlet, only so much can flow
out the other end.
If you need more water, you need to find a bigger hose.


>
2.  Why compare a 20v motor with the RS2 to a 10v motor with the hybrid
turbo?
173*2.5=432.5CFM.  By my calculations that is an 11% drop in CFM and HP.
>
Assuming you have a turbo which can operate at 2.79 and flow 475 cfm,
you need to ask the questions, can the MC engine physically flow that
much air
(475 cfm)? Can the intercooler flow that much?, throttle body? intake
manifold?
head ports? can that much air flow through 38mm valves? can the exhaust
system
dispose of that much air? 

We have brought out the math to do some of the rough turbo sizing, but
we need
to return to the engine and see if it is sized for the high cfm job. We
can bring
out the math to do that analysis too.

>
3.  What I still do not understand is how you differentiate the turbos
from one another  in your equations.  Lets say I have the Joe Schmoe
brand turbo that can just barely blow 2.50PR does the efficiency of the
turbo hold equal to the RS2 or is the DR lower because of the air being
hotter due to la ess efficient turbo?  I think any assumptions you are
making about the density of the air to the engine is just that an
assumption.  I think the bigger turbo works less to make the boost and
therefore has a higher air density than the RS2.  
>

What you think are assemptions are really phsical characteristics
of the turbo as a unit(cold side housing/cold side wheel/hot side
housing/hot side wheel). Change any 1 of the above and the turbo 
map changes. 

The density of the air in the engine isn't an assumption, you can
can calculate it and the temps associated with the air. There are some
initial conditions which need to be defines before we start, like
intercooler efficiency, so we are making assumptions there, but
what happens to the air is not assumed, it's a physical phenomena
which we have equations that help describe what is happening.

-
Dave Lawson

-