205 GTI Pug flow meter to MAP conversion - how hard can it be?
  • Help
Page 1 of 2 12 Last
Results 1 to 25 of 28
  1. #1
    1000+ Posts schlitzaugen's Avatar
    Join Date
    Nov 2010
    Location
    loneliness capital of the world
    Posts
    9,414

    Default 205 GTI Pug flow meter to MAP conversion - how hard can it be?

    Or is it possible at all?

    Main restriction is that I want to keep OEM Jetronic ECU.

    Anyone?

    Advertisement
    ACHTUNG ALLES LOOKENPEEPERS

    Das computermachine is nicht fur gefingerpoken und mittengrabben. Ist easy schnappen der springenwerk, blowenfusen und poppencorken mit spitssparken. Ist nicht fur gewerken bei das dummkopfen. Das rubbernecken sightseeren keepen hands in das pockets-relaxen und watch das blinkenlights.

  2. #2
    1000+ Posts Capago's Avatar
    Join Date
    Oct 2010
    Location
    Hornsby, NSW
    Posts
    1,016

    Default

    dont think it can be done, although both load sensing devices, the MAP sensor would output a different signal to that of an air flow meter, not so much in the sense of the waveform/signal/voltage/resistance etc would be different, but what they are sensing as such (although related) are different. Unless theres an off-the-shelf solution i dont know about, only option is aftermarket ECU with a MAP sensor. cleared up alot of room on my 205 mi16, ran like a dream...briefly...
    The wrong oil is better than no oil at all.

  3. #3
    Demannu-facturing! Demannu's Avatar
    Join Date
    Dec 2002
    Location
    Menzies Creek
    Posts
    2,566

    Default

    When I had my first Mi16, I made an 'airflow meter eliminator'. It was an analogue device consisting basically just of amplifiers, tweaking the inverted signal from a MAP sensor and combining it with a 0-5v signal modulated from the tach pulse.

    It produced a signal between 0 and 5 volts as a function of manifold pressure and RPM. It had adjustment pots for biasing the voltage one way or the other.

    It didn't work very well. It couldn't compensate for changes in volumetric efficiency through the rev range. The only way around that would be a mapped system. And then you may as well just fit an aftermarket ECU.
    Scotty

    Melbourne - Dandenong Ranges

    1956 Peugeot 403 - 'Francois' - resto project

    1969 Peugeot 504 - 'Pascal' - daily driver project

    1970 Peugeot 404 Utility - 'Brutus' - resto project

    1978 Peugeot 604 - as yet unnamed - V6 on straight LPG

    1987 Peugeot 505 - as yet unnamed - project car

    1999 Peugeot 406 Coupé - 'Chloe' - 5 speed manual

    2011 Peugeot 3008 XTE HDi - 'Zoe' - hatchback on steroids

    2014 Peugeot RCZ - 'Remy'

    1999 Range Rover 4.6 HSE - 'Grover' - tow car

  4. #4
    Fellow Frogger! Molerpa's Avatar
    Join Date
    Jul 2005
    Location
    South America, Argentina, That country that looks like hanging from the map... just about to fall...
    Posts
    492

    Default

    This is the first question you start making, before deciding for MegaSquirt


    Quote Originally Posted by luthier View Post
    It is simply impractical to expect a petrol motor to get through too much water. That's why God invented diesels.

  5. #5
    1000+ Posts schlitzaugen's Avatar
    Join Date
    Nov 2010
    Location
    loneliness capital of the world
    Posts
    9,414

    Default

    Demannu, that's the idea. Perhaps your system needs a bit more refining.

    No, I don't want to get rid of the OEM ECU. Don't have the time to mess around with all the mods, just looking into possibilities of small improvements.

    The K-Jetronic on my 17G has a MAP. The LE-Jetronic on the pug can't be that much different. I am sure there is a way to convert the signal and bring it to what the current ECU needs, but my electronics knowledge doesn't go that far.
    ACHTUNG ALLES LOOKENPEEPERS

    Das computermachine is nicht fur gefingerpoken und mittengrabben. Ist easy schnappen der springenwerk, blowenfusen und poppencorken mit spitssparken. Ist nicht fur gewerken bei das dummkopfen. Das rubbernecken sightseeren keepen hands in das pockets-relaxen und watch das blinkenlights.

  6. #6
    VIP Sponsor
    Join Date
    Aug 2000
    Posts
    9,547

    Default

    17G has D Jetronic', K is mechanical, don't know why they call it "tronic.
    Far easier just to fit MS, no mucking around. L Jetronic isn't digital, just an analogue calculator.
    Graham

  7. #7
    1000+ Posts schlitzaugen's Avatar
    Join Date
    Nov 2010
    Location
    loneliness capital of the world
    Posts
    9,414

    Default

    Quote Originally Posted by GRAHAM WALLIS View Post
    17G has D Jetronic', K is mechanical, don't know why they call it "tronic.
    Far easier just to fit MS, no mucking around. L Jetronic isn't digital, just an analogue calculator.
    Graham

    Actually, I think it is L-Jetronic. I'll check again at home. Bloody letters.
    ACHTUNG ALLES LOOKENPEEPERS

    Das computermachine is nicht fur gefingerpoken und mittengrabben. Ist easy schnappen der springenwerk, blowenfusen und poppencorken mit spitssparken. Ist nicht fur gewerken bei das dummkopfen. Das rubbernecken sightseeren keepen hands in das pockets-relaxen und watch das blinkenlights.

  8. #8
    1000+ Posts PeterT's Avatar
    Join Date
    Jul 2000
    Location
    Castle Hill, Sydney
    Posts
    7,567

    Default

    There's more to be gained than just removing the AFM. The ignition curve sucks in those dizzies. Read this thread: http://forum.205gtidrivers.com/index...owtopic=140951
    Do a full conversion and hide the new gear in an old box if you must.

    '92 205 Mi16
    '90 Mi16x4

  9. #9
    1000+ Posts schlitzaugen's Avatar
    Join Date
    Nov 2010
    Location
    loneliness capital of the world
    Posts
    9,414

    Default

    I am trying to get rid of the AFM, not gain more power.

    I see your point, and am sure more power can be unleashed from the beast, but that is beyond my scope. Right now, if I could build a simple electronic circuit which could use a MAP sensor to replace the AFM, I would be quite happy with that.
    ACHTUNG ALLES LOOKENPEEPERS

    Das computermachine is nicht fur gefingerpoken und mittengrabben. Ist easy schnappen der springenwerk, blowenfusen und poppencorken mit spitssparken. Ist nicht fur gewerken bei das dummkopfen. Das rubbernecken sightseeren keepen hands in das pockets-relaxen und watch das blinkenlights.

  10. #10
    Fellow Frogger!
    Join Date
    Nov 2010
    Location
    Blue Mountains, NSW
    Posts
    450

    Default

    Quote Originally Posted by schlitzaugen View Post
    I am trying to get rid of the AFM, not gain more power.

    I see your point, and am sure more power can be unleashed from the beast, but that is beyond my scope. Right now, if I could build a simple electronic circuit which could use a MAP sensor to replace the AFM, I would be quite happy with that.
    It'd be tricky -- they read different things. The AFM reads proportional to the mass of air passing the flap, the MAP sensor just reads the instantaneous air pressure. In mathematical terms, AFM would be related to the integral over time of MAP. So I'd guess your converter would need to average multiple MAP samples and publish this as an analog voltage for the pseudo AFM sensor. So far, so good. You also have to accumulate those samples over varying periods of time depending on engine speed, and you have to publish the "right" signal as expected by the L-Jetronic controller (log scale or whatever).

    I expect you'd end up with a microprocessor based solution rather than analog electronics. By the time you've done all the research and written the software you could have hooked up an off the shelf ECU. But then maybe what you're after is the fun of the challenge.

    Have fun,

    Rob.

  11. #11
    1000+ Posts schlitzaugen's Avatar
    Join Date
    Nov 2010
    Location
    loneliness capital of the world
    Posts
    9,414

    Default

    I was thinking more along the lines of replicating the MAP signal reading part of the ECU on the 17G (with the appropiate sensor) and using that as a signal source in the Pug ECU, bypassing the AFM relevant circuitry. I know this is a step backwards somewhat, but I am not really bothered.

    Your post does make light though of what is involved, thank you. I didn't realise this subtle difference.

    Not sure I understand why you'd need to write software for this?
    ACHTUNG ALLES LOOKENPEEPERS

    Das computermachine is nicht fur gefingerpoken und mittengrabben. Ist easy schnappen der springenwerk, blowenfusen und poppencorken mit spitssparken. Ist nicht fur gewerken bei das dummkopfen. Das rubbernecken sightseeren keepen hands in das pockets-relaxen und watch das blinkenlights.

  12. #12
    Fellow Frogger!
    Join Date
    Nov 2010
    Location
    Blue Mountains, NSW
    Posts
    450

    Default

    You'd need software if you opted for a microprocessor solution.

    I am no electronics expert. I know that such a person would have no bother making something that averaged the MAP readings over time (I suppose this would amount to not much more than a capacitor). And the other operations could no doubt be achieved with analog electronics too. But the trend has been to go digital and use microprocessors for ever more "trivial" problems. I guess this is because tuning digital circuits is generally done in software; you have to change physicial components to tune analog and that loses in the cost/benefit balance.

    Just as an aside, the microprocessor thing is a bit OTT these days. The HCS12 processor used in the Megasquirt II, comfortably able to keep up with engine management duties (squirts and sparks), is suggested for such wondrous applications as "door lock controller" at Freescale's website. Poor thing will be bored out of its skull!

    Have fun,

    Rob.

  13. #13
    1000+ Posts robmac's Avatar
    Join Date
    Nov 2004
    Location
    Melbourne / Caulfield
    Posts
    19,465

    Default

    It's not a matter of being of OTT as much as cost.

    I recently had a project to reproduce the pulse train of a Toyota central locking receiver. The OEM remote was in $500 range and I refused to pay that.

    I looked at a clocked shift register but the time building it on a piece of vero board and the hassle of doing so.

    The solution was an Arduino development board (Freetronics actually-but an equivalent) for less than $40 at Jaycar. This included the board processor, boot loader and usb lead. The software is a free download.

    Writing the code took 30 mins, most of which was reading the blog and support forums to gain familiarity.

    In less that a day I had a working unit installed in the car at around $75 in parts total. If I had gone the discrete processor approach and made a board the cost of the micro controller and parts would have been less $10.00.

    The processors are very powerful and have analogue and digital IO as well as a host of functions.

    For getting rid of an AFM I would suggest it's the only way to go these days.

    Although the programming language can be quickly learnt unless you have a background it take a while to become fully proficient - I'm a long way off that!

    It would looking here if you are interested: http://arduino.cc/en/Guide/HomePage

    The blog is really useful in seeing how experienced programmer tackle the programming.

  14. #14
    1000+ Posts schlitzaugen's Avatar
    Join Date
    Nov 2010
    Location
    loneliness capital of the world
    Posts
    9,414
    ACHTUNG ALLES LOOKENPEEPERS

    Das computermachine is nicht fur gefingerpoken und mittengrabben. Ist easy schnappen der springenwerk, blowenfusen und poppencorken mit spitssparken. Ist nicht fur gewerken bei das dummkopfen. Das rubbernecken sightseeren keepen hands in das pockets-relaxen und watch das blinkenlights.

  15. #15
    VIP Sponsor
    Join Date
    Aug 2000
    Posts
    9,547

    Default

    Quote Originally Posted by schlitzaugen View Post
    If it worked it would need mapping somehow.
    You could have ordered an MS1 ($140) by now.
    Why exactly do you want to get rid of the AFM?
    Graham

  16. #16
    1000+ Posts schlitzaugen's Avatar
    Join Date
    Nov 2010
    Location
    loneliness capital of the world
    Posts
    9,414

    Default

    Clunky and takes up too much space in a crammed engine bay.
    ACHTUNG ALLES LOOKENPEEPERS

    Das computermachine is nicht fur gefingerpoken und mittengrabben. Ist easy schnappen der springenwerk, blowenfusen und poppencorken mit spitssparken. Ist nicht fur gewerken bei das dummkopfen. Das rubbernecken sightseeren keepen hands in das pockets-relaxen und watch das blinkenlights.

  17. #17
    1000+ Posts gezza's Avatar
    Join Date
    Nov 2010
    Location
    Bunbury
    Posts
    2,195

    Default

    Take the windscreen water reservior out that takes just as much space.....

  18. #18
    1000+ Posts schlitzaugen's Avatar
    Join Date
    Nov 2010
    Location
    loneliness capital of the world
    Posts
    9,414

    Default

    The moment I can replace water with electronics consider it done.
    ACHTUNG ALLES LOOKENPEEPERS

    Das computermachine is nicht fur gefingerpoken und mittengrabben. Ist easy schnappen der springenwerk, blowenfusen und poppencorken mit spitssparken. Ist nicht fur gewerken bei das dummkopfen. Das rubbernecken sightseeren keepen hands in das pockets-relaxen und watch das blinkenlights.

  19. #19
    1000+ Posts gezza's Avatar
    Join Date
    Nov 2010
    Location
    Bunbury
    Posts
    2,195

    Default

    And the moment you can replicate a fully working afm that is smaller, that will be done too

  20. #20
    1000+ Posts PeterT's Avatar
    Join Date
    Jul 2000
    Location
    Castle Hill, Sydney
    Posts
    7,567

    Default

    Quote Originally Posted by schlitzaugen View Post
    Clunky and takes up too much space in a crammed engine bay.
    You've folded. Now you've conceded in moving away from originality, this is what I'd do:

    1. gut another AFM, retaining the air temp sensor.
    2. lock up distributor
    3. fit aftermarket ECU in original case and position

    The half decent ECU will have a programmable input for the air temp sensor, so all original wiring can be retained, maintaining the OEM look of the engine bay. The dizzy needs to be locked as it will not be determining advance, but will be used to trigger the new ECU and distribute high tension.

    If you're happy with less cluttered engine bay, replace the AFM with a length of Ø65 PVC pipe painted black and screw an air temp in the side.

    '92 205 Mi16
    '90 Mi16x4

  21. #21
    1000+ Posts schlitzaugen's Avatar
    Join Date
    Nov 2010
    Location
    loneliness capital of the world
    Posts
    9,414

    Default

    Tell you the truth, I was looking at replacing the AFM with a newer filament type model to eliminate the rough idle the flap has, but all this is just me thinking about stuff, not necessarily something that might materialise or not in the near future anyway.

    Performance wise, the dizzy is right now the bit really letting the side down, but it's not an easy modification to do as I found out. Changing the dizzy to something decent say using a megajolt seems to be the easiest way to go, but even then I need mapping, etc. Which in turn means the engine should be top notch otherwise it's not really worth it. By the time that is done, I might as well go full blown megasquirt and that brings another load of work.

    To what end? My engine is just an old DFZ with unknown history. Goes really well as it is, and to tell you the truth, the oil leak past the dizzy internal seal (spiral) is more annoying to me than it's performance shortfalls. Other small niggles include the difficult to access everything and the crammed engine bay that makes it difficult to see anything deeper than the rocker cover. For instance the timing marks on the flywheel. Maybe I'll look at cleaning up the engine bay as much as I can with a redesign and reroute of all that crap (hoses/wires etc) and see how that goes.

    But it's interesting to think about all these things and who knows? Maybe I'll find something that satisfies me.
    ACHTUNG ALLES LOOKENPEEPERS

    Das computermachine is nicht fur gefingerpoken und mittengrabben. Ist easy schnappen der springenwerk, blowenfusen und poppencorken mit spitssparken. Ist nicht fur gewerken bei das dummkopfen. Das rubbernecken sightseeren keepen hands in das pockets-relaxen und watch das blinkenlights.

  22. #22
    1000+ Posts PeterT's Avatar
    Join Date
    Jul 2000
    Location
    Castle Hill, Sydney
    Posts
    7,567

    Default

    The "easiest" way, not necessarily the best way, to fix the dizzy seal issue, is fit a thermostat housing and coil pack from an Si. ie wasted spark.

    '92 205 Mi16
    '90 Mi16x4

  23. #23
    1000+ Posts schlitzaugen's Avatar
    Join Date
    Nov 2010
    Location
    loneliness capital of the world
    Posts
    9,414

    Default

    Quote Originally Posted by PeterT View Post
    The "easiest" way, not necessarily the best way, to fix the dizzy seal issue, is fit a thermostat housing and coil pack from an Si. ie wasted spark.

    Please tell more.
    ACHTUNG ALLES LOOKENPEEPERS

    Das computermachine is nicht fur gefingerpoken und mittengrabben. Ist easy schnappen der springenwerk, blowenfusen und poppencorken mit spitssparken. Ist nicht fur gewerken bei das dummkopfen. Das rubbernecken sightseeren keepen hands in das pockets-relaxen und watch das blinkenlights.

  24. #24
    Fellow Frogger! stew's Avatar
    Join Date
    Apr 2011
    Location
    brisbane
    Posts
    507

    Default

    Gawd, stop trying to change electronics made during the Pyramids. Go and drive an N5 306 Rallye and then you'll give the 205 away.

  25. #25
    Fellow Frogger!
    Join Date
    Nov 2010
    Location
    Blue Mountains, NSW
    Posts
    450

    Default

    Quote Originally Posted by schlitzaugen View Post
    The linked page doesn't make much of a case. Two graphs with different X axes, one curving downwards, the other a straight ramp upwards. Doesn't prove much does it? Just go through a four point, rough thought experiment:

    Closed throttle, idle RPM -- flap closed; MAP 40kPa
    Closed throttle, high RPM -- flap closed; MAP 15kPa
    WOT, idle RPM -- flap slightly open; MAP 100kPa
    WOT, high RPM -- flap wide open; MAP 90-100kPa

    The relationship just isn't as simple as that link suggests (it isn't as simple as the above suggests either).

    And I agree robmac, it makes cost/benefit sense to use microprocessors, but it's still OTT. I guess I'm a bit like the early electronics researchers who thought ICs were an interesting but impractical idea "'cause you'd never be able to repair them". They didn't think they'd just become a throw away item. I have much the same reaction to something about 50x more powerful than my first computer being used to check whether it's time to unlock the door. I know I'm not right about it, but it still hits me as wasteful.

    Have fun,

    Rob.

Page 1 of 2 12 Last

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •