Re: [E36M3] RE: Jim C 2.8 dyno results (from Z3 list?)

[Land Shark <lndshrk@xxxxxxxxxxxx>]

At 05:51 PM 1/30/99 -0500, Tom Peluso wrote:
>
>->  Sure.. lots of them.. BEYOND the fact that I have repeatedly said
>>  that owners cannot reliably dyno an OBD-II upgrade because of the
>>  adaption concerns (and no one seems to listen to me.. they just 
>>  keep wasting money.. and the dynojet operators keep making money)
>>
>How does one properly dyno an OBD-II upgrade?

  Tom, simply.. as an owner.. you CAN'T..

  Let me explain why (again ;)..

  It's called adaption.. simply put, the Siemens boxes in the 
  BMW 6 cylinder cars adapt under a much wider range than the
  previous OBD-I.. and worst off YOU can't clear the adaption..

  See, previous to OBD-II, adaptive information was stored in
  battery backed SRAM.  Unplug the DME to swap chips and woosh
  all adaptive factors set to "nominal" (1 for mult. and 0 for add.)
  Then came Kalifornia, and KARB, and OBD-II.. and EEPROM for
  storage.. this means you can't just "erase" it anymore because
  the last thing the DME does during shutdown is write the SRAM
  locations to EEPROM.. and guess what the first thing it does is.
 
  Yup... zero SRAM, run a galpat memory check, and reload the
  needed ram locations from EEPROM.

  The only way to clear the adaptive locations is via diagnostic
  commands down the serial port (even desoldering the main
  FLASH memory program store doesn't do it, because there is a
  separate EEPROM elsewhere)

  So if you can't clear it.. you can't do an apples to apples
  comparison.

  What's worse is the "quasi-tuners" out there who have no idea
  how it all works, spouting nonsense.. 

  Either you trust the person your buying it from, or don't
  buy it.  It really is like "Natural Flavors" in soda...

  Either you trust the manufacturer, or you put the can back
  on the shelf, because it's not easy to quantify the contents.

  I'll take my "Natural Flavors" in Surge or Mountain Dew flavors ;)

  (Unless you carry Jones Soda, and then I want rasberry!... YUM!)

  Now.. back to adaption.. since you can't clear it
  you can't get a valid comparison.. want to try something..

  Take your car.. OBD-II.. chipped or stock.. drive it one week
  like a MADMAN and dyno it.. then drive it for another week
  very meekly.. and dyno it.. compare the results.. and they
  WILL NOT be anywhere NEAR the same.. and we didn't change the
  car at ALL.. 

  Why?? (Prof. Shark asks the class)

  Exactly.. the car ADAPTED to different areas of the maps
  and either enrichened or enleaned the overall mixture.

  Now.. in whatever state the car is in, you can be rather
  assured that ON AVERAGE the chipped car is making significantly
  more power than the stock car.

  Adaption works like this.. at lower loads and rpms it watches
  the AVERAGE correction (realize we swing rich and lean of 
  stoichiometric continuously, except for WOT)... and it adds or
  subtracts a bit of fuel until the low speed fuel mixture is 
  swinging evenly around Lambda = 1.0 / AFR 14.7:1..

  Now as the rev range increases.. we switch to a multiplicative
  correction which adjusts (if you will) the SLOPE of the fuel
  curve by multiplying the STORED values by a factor (based on 1.00)

  Depending upon how good the manufacturer tuned the low end and lower
  midrange, the environmental conditions, the fuel you are using,
  and the flavor of soda I have in my fridge (*grin*) .. well really
  what AREAS of that lower speed area you are in most of the time
  during the drive cycles.. that determines WHAT those two magic
  numbers will be.

  The kicker.. and bummer.. is that even though lambda control isn't
  active, and adaption isn't HAPPENING at high throttle angles, those
  two numbers ARE STILL USED IN THE CALCULATION regardless as an
  "overall correction" to fueling.

  So.. let's say that at 20% throttle and 3000 rpms the manufacturer
  is a bit rich.. the system adapts lean.. if you drove at very light
  load/throttle for a while at 3000.. you might see your peak HP
  drop 10HP on the dyno.. once the car adapts

  Conversely, let's say the manufacturer is a bit lean at 2000 and 50%
  and you spend your week in LA traffic jams always gassing it in 4th
  and not downshifting.. car goes rich.. now you dyno your car and
  BOOM magic horsepower have appeared (well.. not really)

  Simply put.. while any chip manufacturer can tell you that 90+% of the
  time your chipped car will have more HP than the stock one, unless you
  have cleared the adaption BOTH times when you run the cars you can't
  quantify it at all.

  Jim

  PS: Know that this isn't something I'm pulling out of my butt, it's
      as frustrating to me as it is to you, because to tune these cars
      you disconnect all 4 sensors and then put the car into intentional
      FAULTS to stop adaption.. which you then clear to 0 and 1 respectively
      then tune.. then test.. and on it goes.

  

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