RE: Overboost defined - long

[QSHIPQ@aol.com]

Scott M writes:
>...
>I have experimented a bit with increasing this allowed "Overboost" time
>"slightly" during one test, where my car was hitting the overboost fuel
>pump cutout, after installing a bypass valve, which allowed the boost to
>shoot up quicker past the 1.8 bar set point during a 3rd/4th gear shift.
>For safety purposes, you do not want to increase this time too much. A
>better solution would be to use a 2.5 bar pressure sensor for an increased
>control range.
>A slight "Overboost" can be seen on these MAC11/14 vehicles running stock
>boost levels, (1.4 bar) and will also show the same under damped response,
>with the ECU controlled boost, and the over damped response if the upper WG
>hose is disconnected. 
>This "Overboost" is a nice feature as it gives the car a shot of additional
>torque to help move it down the road.
>...
>"The dynamic pattern of boost pressure build-up can be influenced by the
>boost pressure control system. For rapid acceleration, "excess boost" can
>be obtained for a limited period."


I'm intrigued by the claims of audi, Scott, since a second or two of 
"overboost" readings really isn't affecting a measureable time.  In fact, to 
test this, I tested a 1.8bar HKS mod (1.8bar peak) to a 1.8bar ecu mod 
(2.0bar 'overboost' seen).  The results would indicate that audi makes a more 
comprehensive boost controller (more inputs), but that the WGFV control 
circuit is somewhat "primitive".  The HKS car (second to last gen controller) 
hit the programmed rpm range/boost profiles on the money (no overboost), and 
was faster to spool up the turbo, and was consistently faster in 1/4mile 
times.  Advantage to the boost controller.

However, this is a hardware issue, IMO.  The boost profile and it's feedback 
loop isn't as quick in the audi PT>ECU>WGFV>PT as the HKS.  A better WGFV 
would probably negate the mention of the 10vt "overboost" as a feature, 
desired or not.  When I see a relatively simple boost controller doing a 
better job at controlling just boost, it becomes obvious that some 
compromises are in the audi systems.  Is it enough to justify a boost 
controller?  Not IMO, but I've installed a few of them, and I haven't seen 
any failures.  I know Dave Jones @ Revsport has put dozens in, and he swears 
by them

The stiffer the WG spring, the more critical it is to really address the WGFV 
values, IMO.  The ECU is looking for a specific boot profile, when you 
already have a high baseline spring pressure, the WGFV addition is helped 
(less to add) and hindered (any addition is usually too much).  To test this, 
try a 1.8 or 2.0bar mod with a stock WG spring.  I never got a measured 
"overboost" profile with this test, and it took forever to get to the proper 
boost.  Hardware sucks.  HKS unit didn't care what spring you put in> no 
"overboost".

It seems apparent, that audi is liberally using the "overboost" readings as a 
good thing ("...can be obtained" doesn't at all mean it 'is' or is a 
measurable "feature").  Not sure I agree or disagree, the side benefit might 
be a slightly higher instant reading, but translating that into some sort of 
"performance" really is stretching the argument, SAE paper or not.  A peak 
reading that immediately falls, is a tough argument to measure, more 
specifically, it doesn't really make a difference.  10 or 15 seconds of that 
overboost profile does.  However, even that has some slop in it.  I've 
measured boost spikes in true "overboost" mode in the 20vt cars as well.  
Neither the 10vt or the 20vt use exceptional WGFV IMO.  The 20vt ones fail 
pretty regularly with ecu mods too, the 10vt ones don't.

WRT what you posted on the 944turbo cars, I'd be careful here Scott.  Several 
in the dodge technical department found out that this modification to the 
2.2l turbo cars (boost bleed mod) gave some lean fuel results, and warnings 
about this modification were widespread.

Overboost cutoff and overboost mode have accepted defintions in the turbo 
arena.  Using a slow feedback loop WGFV control system in the context of 
"overboost" is a great technical spin of terms, but really isn't a correct 
presentation, IMO.  The proof is in the simple and cheap HKS pudding.

Thanks for the post Scott, most interesting.

Scott Justusson