aem infinity standalone on e36 obd 1

[bry195]

grabbed the right rear and front left wheel speed signal at the abs unit and ran it to the infinity.. I'll check functionality soon.

[bry195]

so here is what the idle invert circuit looks like. Green with white monitors a PWM out of standalone to the close side of the 3 wire valve. yellow with white is the inverted signal to the close side of the valve. 1k resistor. nnvn1 and rectifier. thing gets hot! I found a 25 mm bldc 2 wire fan that keeps it chilly. Now if I just had a steady hand. surface mount would be great if it was about 3 times bigger.

[e21jps]

I am also considering a 8H, and having it tuned by Jordan. The instructions they give for wiring an s50 is pretty damn easy, they literally tell you exactly which pins to connect.

I've been looking for a downside to this stand alone, and cannot find a single one.

Are you kidding me!!!!!!..... look at all the crap he has needed to do to make this work!!

Im about to do the exact same thing on an m50 over the next few days but using a diferent brand of ECU and it will be a direct plug in! all i will need to do is make up a jumper harness, nothing else needed not even igniters! and i guarantee it is a far cheaper ecu too! 3 wire idle valve no problem, full 3d vanos control (even duel vanos m/s54 etc) no problem, 6 coils driven directly no problem, all the oem sensors incl vehicle speed and flex fuel read directly no problem, features traction control antilag, launch control all handled internally plus almost anything else you can throw at it.... that AEM Ecu actually looks like a bit of a nightmare!!

[bry195]

Are you kidding me!!!!!!..... look at all the crap he has needed to do to make this work!! Im about to do the exact same thing on an m50 over the next few days but using a diferent brand of ECU and it will be a direct plug in! all i will need to do is make up a jumper harness, nothing else needed not even igniters! and i guarantee it is a far cheaper ecu too! 3 wire idle valve no problem, full 3d vanos control (even duel vanos m/s54 etc) no problem, 6 coils driven directly no problem, all the oem sensors incl vehicle speed and flex fuel read directly no problem, features traction control antilag, launch control all handled internally plus almost anything else you can throw at it.... that AEM Ecu actually looks like a bit of a nightmare!!

The infinity does all that. The idle circuit just save you from using 2 pwm outputs. you could go to a 2 wire valve and do the same. I took too many things to the Nth degree for people to get a feel for if the AEM is easy. the drawing for the M3 shows you every pin on the dme and where it lands on the AEM. its really not bad.

[bry195]

Are you kidding me!!!!!!..... look at all the crap he has needed to do to make this work!! Im about to do the exact same thing on an m50 over the next few days but using a diferent brand of ECU and it will be a direct plug in! all i will need to do is make up a jumper harness, nothing else needed not even igniters! and i guarantee it is a far cheaper ecu too! 3 wire idle valve no problem, full 3d vanos control (even duel vanos m/s54 etc) no problem, 6 coils driven directly no problem, all the oem sensors incl vehicle speed and flex fuel read directly no problem, features traction control antilag, launch control all handled internally plus almost anything else you can throw at it.... that AEM Ecu actually looks like a bit of a nightmare!!

E46 is plug and play. E36 is just a jumper harness that you have to make. all of the calibrations and connections are published and included. Almost every step I have taken has had about 3 revisions so that it felt right to me including the idle circuit (about 5).

the only faults I have found are in the firmware revision. if I disabled injectors I had to take a few steps to get them back on.
the idle circuit to condense 2 pwm channels to 1 is a little generic.

a jumper harness could be made in a day or 2 at the worst.

Im doing things aem doesnt suggest nor support so I have to be very careful. the ground and shielding work I have done is way too much. Im also not using a wideband or MAP sensor the way AEM intended.

it looks like allot of work. Its really nod different than any other standalone install. the ignitors are separate though. But i prefer to replace an ignitor rather than send the whole ecu off for repair.

[bry195]

I prefer to really detail things and revise them. This thing has the fastest processor on the market (2 processors) and the way it works allows you to do just about anything. including things not written into the firmware. I am really happy I selected the AEM. it is as open source as you want to make it or you can use it out of the box. the support is immediate. there are shops in just about every town as well. Im excited to have a chance to be decent on this platform. its really powerful but simple at the same time.

[e21jps]

Ok i admit i didnt study this whole thread really closely some good information for others doing similar installs etc. Sounds like its got plenty of computing power, not worth considering an ecu unless it has at least 32bit processors these days but unfortunately many if not most dont!... and they often wont disclose the size either!!! So im surprised it needs so much extra stuff. I was just looking at all the little gadgets extra wiring, extra soldering, little auxiliary circuits etc that is needed, where many other units will do all that internally with a tick of a box etc. I think the safest place for igniters is tucked away inside the ecu just like OEM I have never heard of internal ones dying. We will use an internal wideband for logging and tuning, prob need to put the o2 sensor on its own flying lead but that will be it for additional wiring

[bry195]

Ok i admit i didnt study this whole thread really closely some good information for others doing similar installs etc. Sounds like its got plenty of computing power, not worth considering an ecu unless it has at least 32bit processors these days but unfortunately many if not most dont!... and they often wont disclose the size either!!! So im surprised it needs so much extra stuff. I was just looking at all the little gadgets extra wiring, extra soldering, little auxiliary circuits etc that is needed, where many other units will do all that internally with a tick of a box etc. I think the safest place for igniters is tucked away inside the ecu just like OEM I have never heard of internal ones dying. We will use an internal wideband for logging and tuning, prob need to put the o2 sensor on its own flying lead but that will be it for additional wiring

I think most people looking at this thread would get the same impression. you can tick a box. there is a wizard that allows you to tell it this is a bmw e36 with so and so sensors and it will setup all of the rest of the data (Base session). then you can run the car or make changes.

its had a couple issues but the pain in the arse it appears to be is really about (for the most part) making it a pain in the arse. I pull things apart over an over again.

[bry195]

car is running again. Sorry, took my time to test everything before lighting it off.

so I enabled the oil temp and pressure failsafes
got the idle and idle decel working pretty well but I still need a couple tweaks


has anyone looked at their fuel pressure when the car is heat soaked? I have perfect pressure until coolant temps are 105 ish. 29psi at idle oem regulator.

DEF-when you look at the fuel pressure scaled. and its referenced in the software to manifold vacuum....wouldnt pressure always hover around 50 psi? I wonder if the scaler is working in reverse or the fpress is not scaled to the manifold vacuum at all. as vacuum increases fuel pressure drops??

[bry195]

I think I found the reason for the (almost) hydrolock. the electronics in the coil ignitor 1st channel appear to be damaged. YAY

[Butters Stoch]

I think I found the reason for the (almost) hydrolock. the electronics in the coil ignitor 1st channel appear to be damaged. YAY

This is a brand new unit ?

[bry195]

This is a brand new unit ?

yes, brand new but to be fair.

1-huco makes these for aem and they make pretty good stuff

2-I had some sort of issue with one of my coils and in the process plugged it into the second pack. in both cases it was that channel. I cant say it fried the transistor for sure because I dont really know and its probably going to be replaced under warranty (fingers crossed). dont mess this up for me!

I found lots of ems mfg's selling these for 100 to 125 each (2). If I knew which Huco it was they are about 35 bucks street price. Anyone know?

[Def]

car is running again. Sorry, took my time to test everything before lighting it off.

so I enabled the oil temp and pressure failsafes
got the idle and idle decel working pretty well but I still need a couple tweaks


has anyone looked at their fuel pressure when the car is heat soaked? I have perfect pressure until coolant temps are 105 ish. 29psi at idle oem regulator.

DEF-when you look at the fuel pressure scaled. and its referenced in the software to manifold vacuum....wouldnt pressure always hover around 50 psi? I wonder if the scaler is working in reverse or the fpress is not scaled to the manifold vacuum at all. as vacuum increases fuel pressure drops??

I'm pretty sure fuel pressure logs as psig. There's a separate channel called "Fuel pressure differential" or something like that. It references back to manifold pressure and should be fairly constant. Fuel pressure differential is what is used for fueling calcs.

I have no fuel pressure deviation as my car heats up with a ~11-12 yr old Walbro 255 that's been through hell and back.

[bry195]

Thanks DEF that worries me but I may hit you up for a fuel pressure sensor and/or map sensor.

Got the tach working. 50% duty and 3 pulses per rev (low side 12 volt) on the TD signal from the DME. now the MPG sensor appears to have homed leading me it is ready for some of the same.

got the barometric pressure sensor wired and spot on.

got the flex fuel reading correctly I believe. alcohol content looks right and with a few changes in the virtual channel I got the derived fuel temperature logging as well. where would I find some math on how fuel temperature modifies AFR? I think I can create a table that modifies injector efficiencies or lambda target.

got the wheel speed sensors wired in and programmed. just need to test the response but it looks right.

as an FYI I bought some cheap 3 bar knockoff DELPHi GM map sensors. read 4.5 volts at sea level so that aint 3 bar. oh well 20 bucks each. now one of them is my barometric pressure. they do respond the right way to vacuum and pressure just limited to 1 bar.

[bry195]

this is tapped into my knock sensors. Maybe I can hear detonation with it. the point is to watch the frequencies in the aem without them being degraded by listening at the same time. Still waiting for coil drivers to come back. One was bad. Looks like its covered by warranty. I very well could have damaged it. Oh well, thanks AEM.

[bry195]

new coil drivers made it happen.

waiting for sealed logging cable to seal everything up.

[bry195]

drove the car today. runs like crap but I forgot to set the coil dwell back up where it belongs. maybe that is the problem.

3 fuel pressure sensors and fuel pressure is always low so next stop is a fuel pump. probably should do that before a bunch of tuning.

[Butters Stoch]

drove the car today. runs like crap but I forgot to set the coil dwell back up where it belongs. maybe that is the problem.

3 fuel pressure sensors and fuel pressure is always low so next stop is a fuel pump. probably should do that before a bunch of tuning.

I have a 255 and a brand new 340 sitting in my box.

[bry195]

I have a 255 and a brand new 340 sitting in my box.

aww, man. I pulled the trigger on the walbro 485 kit from RallyRoad. Looks like Eric dropped it in the mail about the same time I sent the money.
Thanks though!

I looked in my tank with a flashlight. looks like a bunch of flaked of nickel plating. I fished most of it out with a magnet. I threw a new filter sock on the 255 and I had a solid 45psi while idling. now to get it hot and see if it stays that way. thats more pressure than I have seen so far. Thats about right for idle isnt it?

[bry195]

2 discoveries while im sorting the cylinder 1 issues.

I now control the econometer. I'l post pulse frequency and the relationsip between speed and the gauge once I look up the numbers I used.

the 240ohm reistor on the ground for the coils has a purpose! it sends a pulse back to the DME everytime a coil pulses. something like 50ma and I'll measure everything later and post.. Im hoping width amplitude and frequency are all representative of each cylinder.

something like a knock window should allow for monitoring misfire. possibly even closed loop coil control.

[bry195]

I hate to repeat myself to the guys who saw some of my other threads but I use this one to keep a log of the install so just for the record:

1-tacho working
2-drven and non driven wheels scaled and working
3-econometer ranged and scaling soon
4-cylinder 1 not running was bad injector connection
5-cylinder 4 not running bad coil driver
6-all sensors working except for voltage differential on wideband failsafe
7-ac working but no freon
8-need to install real map sensor
9-need to install wideband
10 idle invert board working great except the inversion circuit creates voltage even with aem off cuz cont power not switched power
11-need to seal honda connector and fit external junction connector.
12-485 walbro from rallyroad installed and solved pressure issues (still some metal in tank)
13-golan 10 micron filter installed
14-new fpr installed
15 flex fuel on wrong voltage still but reading correctly. change to 12 volt soon
16 misfire detection resistor
17 moved throttle plate closed from .8 to .6 to get icv off 100 percent duty at 800 rpm.
18 load comp for acc done
19 lambda and icv pid was good but off now from bringing number 1 in.
20 wall wetting and dynamic lambda PID needs to be redone

[bry195]

knock wired but now it needs calibration
I stole Jons timing
stole Jons 2.65ms dwell on coils (thanks Jon)
URO parts made some donations need to install and review
be real nice to boris coming soon

[bry195]

here is another lesson.

I used the aem failsafe gauge as my map sensor. it did ok at idle but once loaded it didn't. it felt like the motor was missing. the afr part of the gauge seems to be kinda ok but since I was buying stuff I got a wideband for the infinity too (thanks DEF).

Maybe I'll pull the whole aem failsafe wideband logger and fit a nice EGT gauge in their. trades welcome.

[bry195]

so now that the car is running I have had a little time to spend on understanding how things are working together. I also mentioned previously that something was still a little off about it.

so I see crank duration looking parabolic at higher rpms. so after looking at crank and cam pulses some are getting lost. a 10 or 20k resistor between sensor legs is posted all over the web but before I go through that 2 things.

I calculated the filter levels for the chipset. I'll put the right values in.
my wiring shouldnt be an issue but I'll check the shielding and resistance.
then resistors.

[bry195]

nice little sealed USB port installed in dash, it also has a headphone connection for listening to the knock sensors. PWM drive for fuel pump. no the wire nuts are not staying. New rims look pretty good. harting connector on the ECU compartment for add on sensors. still have to do some house keeping to make everything clean and safe but the car runs and drives great at low rpm and load.