Installed custom VEMS plug and play (55pin Motronic) to Porsche 964 Tiptronic, works fine but Tiptronic is in limp home mode due to the missing fuel injector pulsewidth signal.
If it needs fuel pulsewidth signal, why not connect a "fuel injector output" ? Tried that, still limp mode. Car has bigger injectors and it seems that pw is too short for it. Since it is injector pw, signal is likely switched GND
- remember that any injector channel can drive any number of FETs (upto 8)
- what conversion function do you think it needs ?
- Eg. 0-100% for 0-15 msec fuel pw ? (this is just guessing, correct this if you have some information or better guess)
- 0/12V or 0/5V digital, or filtered 0-5V analog ?
- perhaps inverted ? (more 0V as injpw grows ?)
I connected the P259 pin 4 (ch0, with pullup) to the corresponding pin (pin 32 on Motronic 55pin connector) and configured boost gauge output hoping it will settle for just some kind of square wave signal but Tiptronic is still in limp mode.
- boost alternate can be configured to almost any frequency, and function (RPM and MAP) so it should really be configurable (maybe needs inversion ? Also configurable).
- Or need RC filter ? eg. 4k7 * 10uF = 47msec time constant
- once the table is satisfactorily "tuned", it is possible to implement in a way that boost-alternate is freed (in case both boost and boost-alternate are needed for boost in the same application + tiptronic also).
Any tips how to proceed with this? I think fuel injector load output signal would be useful for many cars, even Audis use this to calculate fuel consumption, many Porsches etc. With display you can just ignore it but with tiptronic it ruins the driving -> cannot be ignored...
Please produce some scope measurements of the stock ecu having both one injector signal on one probe (ch1) and the pin32 mot55 signal on the other probe (ch2), and take a few (clearly marked by filename) scopeshots @ idle rpm, 3000, 5000 rpm.
Without proper measurements its hard to know whats needed at all.
Brainstorming...
- If fuel inj-pw is sent (or added as new var) via AIM to round device, and round outputs the PWM signal, would that be a solution ?
- Since fuel PW is obviously available on injector outputs, just need some conversion (eg. fixed freq, not varying with RPM)
- an atmega328 (or 168) arduino (5V/16MHz pro-mini, really cheap and fits in a small box or even heatshrink tube) might be a good choice (very simple firmware if the desired conversion formula is known)
Perhaps I can ask customer to install back original DME (easily done since PnP) and to scope the output signal from original DME? PeepPäädam