[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: Tweekin Turbos II (very long and nerdy)

In a message dated 97-03-04 10:48:08 EST, you write:

<< 	Scott--I fear the light thread will rise again like the 
 the ol' pheonix.  Why not take a look at the lights?  I don't know 
 whether the damn things are refracted, contracted, protracted, etc., but 
 I do know I can see, and seeing in an Audi *is* a beautiful thing.
>>>>>>>  I've been driving a car with them for a month now, Bruce...  Look at
my posts, nowhere do I claim the aeros are "bad"....  When you have driven
"better", makes ya think....  And, unlike all you lucky Metrix 440USD buyers
the new Metrix ante is now at more like 750USD...  Now watt?  The real
advantage is the high beam for sight, the H4's lows are "better" (?), but
hardly the difference the highs give.  Is that what you refer?  If that is
the case, a dedicated aux hibeam can easily outperform the H3 designed aero
highs...  On to more interesting stuff....
 >	Also, on your turbo talk.  I know on the '89 2knock car that I 
 >have, the turbo is just as you describe--almost no lag.  Perhaps the 
 >design is to make it feel like an Audi V8 [which what I assume you 
 >meant], I wish it was designed to feel like some other V8's I've had.  I 
 >don't pretend to know anything about turbo tech., but I have heard that 
 >kkk has a turbo either out, or coming out, that provides more thump than 
 >the RS2--and eliminates most of the lag that has always nagged the RS2
 >True or not?  Frankly, I have only driven 1 Audi with the RS2 in it, and 
 >the lag was very pronounced.  So, I guess the question of the month is, 
 >how do you eliminate the turbo lag inherent in the RS2?
My understanding of the KKK "special" turbo is not available yet, and if you
think prices are high now, the cold side of that turbo alone is triple the
price of stock RS2 unit....  Find out what drives the cold side, it might
surprise you...  The RS2 in stock trim without eprom mods has lag....  That
is for a couple reasons, 1 addressed temporarily by porsche, indefinitely
from my seat (more on that later)....  The first is (and I'm not giving away
my "store" here, if you want specifics you can e-mail/call me directly, hence
the <stop> you will see, I do make money at this), the cold side of the RS2
is designed conservatively...  You can put a different cold side on the RS2
unit, to give it a more aggessive MAP...  Guess what?  Porsche did that,
called it the RS2 Competition.  Limited supply, Comps no longer made.....  I
got the numbers off those pups before the last one sold, and RS2 Competitions
are still available thru this nerd....  Next:  The hot side design of the RS2
gives it a propensity to whistle, I mean like, constant cop chasing type....
  It has to do with the inlet/outlet and turbine design <stop>....  The hot
side can be modified to spool that turbo much faster, however, the cops are
now under your hood....  Anyone that has been for a ride in PDQ can tell you,
it's always there...... What I'm sure audi did, was take one ride in the
thing and go "no way" not on a 100k$ car....  So some changes (correctable)
were made to the turbine design to eliminate the whistle.....   

There are several ways to address the RS2 lag, one is in the design of the
motor tweeks, the other is in the computer box itself.  Let's look at a
typical RS2 mod, let's say a 87 5ktq, starting stock,  what to do...
 (baseline assumption is I want an RS2 unit)....  Computer, IC and exhaust
are the simplest of mods....  Specifically, the computer for more boost, the
IC for charge air efficiency, the exhaust for turbo efficiency....  

First, the computer should give massive "off boost" ignition advance, in
fact, as much as possible, cuz what you are referring to as "lag" is the time
it takes for the engine to get positive manifold pressure from vacuum....
 What does that mean?  Well, here's what happens, you get two forces apposing
each other doing the above tweeks....  A turbo wants NO backpressure down
low, some up high while spooling, the less it has, the faster it will spool,
until it becomes supersonic or gets slowed by WG and/or cold side pressure
waves...  A N/A motor, wants backpressure down low, none up high....  So,
when you go after the larger exhaust, the lag becomes greater, and the
performance cost down low rpm's is too high....  But you want larger exhaust
for the turbo to spool unbridled in positive manifold pressure (on boost)....
 A lower compression motor exascerbates low off boost performance....  The
main reason audi and others now use much higher static compression motors in
their turbo cars...  

We are into a major discussion on this particular subject, bear with me,
turbos are not rocket science, but aerodynamics, in fact, the science is
less, the magic of the MAP application more....  

To eliminate lag, what can you do?  Well, increase engine displacement, make
the computer more aggressive on getting the turbo to spool, increase static
compression ratio, increase the efficiency of charge air components(more on
this), use two smaller turbos.  Remember, most computer mods are NOT EPROM
mods, I am speaking of EPROM mods here....  What needs to be done is to get a
very agressive ignition timing map in the chip, then, once on boost, start
dialing the ignition timing back.  This requires some serious effort to do...
 I have one chip that has it, one that is close, and one that is just a
"stageII" mod....  The difference with the one that has it to the limit, is
an on/off switch, little lag in the turbo, in fact, little lag on a turbo
much bigger than the stock RS2....  Given the prices of the "old" tech turbos
(please, I use that term with tongue in cheek, smile on face) that exist, the
above mods need to be maximized before I'd explore a "new" tech turbo.  

"New" Turbo Technologies for the 90's:  KKK is coming out with a radical
"new" design that they "claim" will put the rest of the world behind the
8-ball....  I won't hold my breath, and yes, I've seen and fully understand
the concept.  The competition is not standing still, just taking a different
avenue for the same net result, k24 lowend and k2/3x at the highend @ a
charge air temp @ xPR (Pressure Ratio).  The KKK unit is expensive, is
radically designed in the cold side, and makes the RS2, even some of kkk's
competition, a bargain in comparison.  A larger turbo requires more energy to
spin (exhaust velocity)...  A 20v motor has more energy available to spin,
but lag time increases again off boost, cuz the motor <0 vaccuum just don't
cut it for performance...  So what you need, is a turbo design that gives a
high air flow for minimal velocity AND a high air flow at maximum velocity.
 For that you need to address the components....  KKK design is to
essentially give you both traits on the cold side, and the execution is sound
in principle, expensive in design and execution <stop>....  The other way to
do the same thing is to reduce the frictional losses on the rotating assembly
itself, such as a garrett turbo (reducing reciprocating weight), or look at
the design of the bearing components <stop>...  The KKK design is wallet
prohibitive, the garrett et.al. is more expensive than the RS2 based turbos,
but still about <1/2 the price of the KKK unit.  Personally already exploring
both systems, the latter has MUCH more potential at this stage of the game,
and given the production costs of the KKK design, it will be some time before
that reality becomes an option.  Since the number of modified RS2 units sold
from this "store" is small cuz of the $2M price tag, I doubt a lot of takers
are going to entertain double+ that....  I know of, and have designed several
turbos with the Garrett idea (hybrids, found on lots of HP audi turbos),
quite a few with VNT (variable nozzle turbo) and know of only 1 with the et.
al. execution, and no one that has the "new" KKK...   The 1 that has the et.
al. wasn't designed or sold by Ned.....  Hence my questions.....  Remember,
as we enter the 90's, the only "new" design is the KKK one, and really is a
different execution of the VNT, no more....  Could be the answer...  The
others are technologies made possible in the 90's, but none are really "new"
concepts or designs, just better executions of the "old"

Let's back up some...  There are a few components associated with turbo
performance:  static compression, effective compression (BMEP), rpm,
volumetric efficiency @ 0 vaccuum (includes intake, cam, valves, fuel,
timing, EM, exh, etc), turbo hot side housing size and design, turbo cold
side housing size and design, turbine size and design, compressor wheel size
and design....  What turbo manufacturers do is take a motor and a
turbocharger at an altitude (Po) add rpm's which gives a boost pressure (P1)
and that pressure@rpm gives a volume of air output @ a turbo efficiency (65%
being considered the minimum effective target efficiency, 75% is really
good)...  This plotted on a graph shows PR on the x-axis (which is Po +
P1/Po) and volume of air output on the y axis, a plot of the rpm rises thru
these axies, thru rings of turbine efficiency (which look like lopsided rings
of tree growth in that graph)...  Let's look at a k26 stock for example, it
reaces a maximum efficiency of 72% @ ~2.0PR @ 4500rpm, then slowly tapers to
<slightly lower> PR @ 6200rpm.  Peak torque is achieved at ~ 1.9PR @ 65%
efficiency @ ~3300rpm (stock car).... What happens when you "oversize" a
turbo, is that the velocity of the exhaust gasses isn't enough to effectively
drive the turbine, so efficiency is low when rpms are, efficiency is high
when rpms are....  So you get a sudden rush of power high in the rpm band, at
the expense of the lower....  What happens when you "undersize" a turbo is
the velocity of the gasses is great at a low rpm, but as rpm increases, the
turbo tends to go supersonic, which really means that no volumetric ouput
gains are made beyond a certain rpm, efficiency peaks and decreases
dramatically....  There is the basic philosophy behind a k26 and a k24...
  This also carries to the PR, in that above a certain boost pressure, the
air is heated more that pressured, so you lose efficiency becuz you created
boost at the expense of air density (air temps increased)....  If you venture
outside a turbos designed map, without changing components, you decrease your
efficiency, a k26 has a well defined map, and it peaks at below 2.0PR....
 Some "oversizing" is figgred in for altitude, but beyond design, you aren't
going faster, claims aside

What is the difference between  a k26 and and k24 map.... Well the k24
(<stop> not published) shifts the efficiency curves, just as does an RS2....
 So the maximum k24 efficiency occurs at ~2000rpm instead of 4500, so the
trail off as the rpm's rise is more dramatic, efficiency decreases
significantly before peak rpm's....  Hence an early kick and a flat or
dropping feel up high....  The RS2 design, follows the k26, though it gets
more efficient sooner, has a higher overall efficiency, and can handle
pressures easily in excess of 2.0 without losing efficiency as quickly, all
the while increasing the volume of air available....

Back onto the dock....  To reduce the lag of an RS2?  Enable it to freewheel,
and no, the stock bypass valve isn't enough, and increase the efficiency of
the volume of air going into the motor....  Intercooler baby, if you are
running a stock IC, you really don't have a turbo problem....  Boost is cool,
and a measure, but density is the KEY to the equation <stop>.

I can smile at most stage II box executions out there, and the claims of
22psi, becuz beyond 18 or so psi, you are spinning your wheel, and NOT
gaining performance...  You can't beat physics....  IC's deserve serious
consideration in the equation, bolting a turbo on is easy (my record time
from drive to drive on a 5ktq turbo swap is 1h33min).....

This is getting too long, so time to cut it off, if one make claims, have
some facts and knowledge to back them up...  Same old stuff here...  One asks
for facts, gets flamed...  RSR is a pretty vague thing right now, so vague,
one wonders...  I challenge the "Blows RS2 into the weeds" without any facts
or measurements attached, that misleads everyone here....   The technologies
of turbos is hardly rocket science, neither are the tweeks available....  One
doesn't need to give me many details to make me question the "claims", very
few have spent more time spinning medal than this nerd....

No claims, just questions of them

Scott Justusson S.O.B.
'87 5ktqRS2BITW2B