Modifying Motronic 2.3.2 ECU hardware and software.

[-ice-]

The schematic is interesting for the analog-inputs. The multiplexer is set by software. So i if you know which Sensor is physically connected to which AIN you know which sensor is actually sampled in the software. This will help to understand the software...

I haven't done this yet. So i you want to trace something, the Analog inputs please

Bye

[prj]

ok, this is going nowhere then as always.
The schematics are under Bosch NDA.

[vwnut8392]

what was your goal for reverse engineering the ECU and figuring out how it works? probably to tune your car like the rest of us wish to do. i just dont believe in being selfish with information, i enjoy to give as much as i can to the community and see how everything evolves. look at all the info that went public for ME7 and how the community has evolved that ECU, alot of useful functions and features have been added which is great! i envy what the DSM and honda community have and i'd like to see open source tuning just like they have.

[prj]

What is your goal? I have factory data on the insides.
No, I am not going to post it, but if you have a specific question then ask.

[vwnut8392]

Been working on this for the past few days. i started tracing out the fuel/timing board, im far from done but its progress. this is bottom side of the PCB over the top of the PCB. I feel this is something that needed to be done to bring this ECU more into the public eye. cant wait till its done to share it with everyone.

[-ice-]

Please think about the logic behind this..

You KNOW the rev limit of 7200. With the equation you can THEREFORE calculate a conversion factor. With this conversion factor you are able to calculate the (real world) rev limit for THE GIVEN hex values.

The conversion factor is the same for auto and manual transmission.

So you have a equation from hex->rev limit an the otherway from rev limit-> hex.


The equation doesn't say devide your result by 8


Bye

[vwnut8392]

That seems right to me. as for them saying that the automatic rev limiter and manual rev limiter bin files have different rev limiters makes no sense to me because im sitting here staring at them in a hex comparison program and the there is no difference at hex offset 0XA37C. The RS2 on the other hand has the values changed.

4A0907551C manual VS 4A0907551D automatic


4A0907551C manual VS 8A0907551C RS2

[andy10v]

Where did 7868800 come from?

[vwnut8392]

so the conversion factor on the motronic.ws web site for the RS2 is wrong because i get 6691.15RPM?

if i do this the conversion factor comes out to be 1058400
7200*147=1058400

This is using the short cut backwards thats posted on the motronic.ws web site.

[andy10v]

Is this correct then?

C/(147x2^3)=7200
C/(147x8)=7200
C/1176=7200
C=1176X7200
C=8467200

8467200/8=1058400

1058400/147=7200

[andy10v]

2 to the power of 3, 2x2x2=8 iirc.

[vwnut8392]

so this is proper?
7868800/(147*2^3) = 7200

what does this mean? (^)

according to the motronic.ws web site if you take the factor and divide it by 8 it creates a short cut to the real world rev limiter value.

7868800/8=983600

than i take the short cut value of 983600 and divide it by 0X9C HEX or 147 DEC. The i get a value of 6691.15. thats pretty far from 7200 RPM.

983600/147=6691.15 RPM.

[-ice-]

I don't know what you're doing there


If you assume 0x9303 is a 7200 rpm limiter you have to solve


convfactor/(X*2^Y) = 7200

with X=0x93 and Y=0x03 to get the conversion factor. You can check the conversion factor by calculating the limiter for the automatic hex values.

Bye

[vwnut8392]

Hi,

I posted an equation in Post 143. Solve the equation to get the conversion factor (by knowing the real world rev limiters of manual and automatic transmission).

Unfortunately i haven't seen the code piece yet, which actually loads the rev limiter word. I guess its loaded via the dptr but i didn't find any hardcoaded pointer (close) to this eprom position.

Byebye

I noticed that a while back also i read about the conversion on http://www.motronic.ws/revlimt2.htm. on that site they refer to the RS2's rev limiter. using the information on that site in reference to the AAN's rev limiter of 7200 RPM the first 2 hex bits should be 0X88. the issue is there is nothing in there with 0X88 in the suggested area where the constants are. im not sure if the rev limiter for the RS2 is the same as the AAN.

basically i did this, 983600/7200=136. 136 is decimal and converting that to hex makes it 0X88. im lost, this is getting very frustrating, ive been working on this for months with no progress.

[-ice-]

Ive been looking at the 4A0 907 551 C binary for the rev limiter and i think i may have found it at 0XA37C. using the info on rev limiters from the motronic.ws web site i think i have successfully calculated the rev limiter.

to figure out the rev limiter you have to locate the binary at 0XA37C and 0XA37D. in the case of the 551C file the hex is 97 and 03. you have to take the 97 and convert it from hex to decimal. 97 converted to decimal is 151. the math goes like this to get the actual real world rev limiter, 983600/151=6513.907284768212. so this makes the rev limiter on that file 6513.90. the value of 983600 came from the motronic.ws web site for the RS2. using the 983600 value is considered a short cut to get the rev limiter. i dont exactly understand the math on there using the whole 16 bit hex code to get the rev limiter. if someone could elaborate on that method of doing it that would be great.

Hi,

I posted an equation in Post 143. Solve the equation to get the conversion factor (by knowing the real world rev limiters of manual and automatic transmission).

Unfortunately i haven't seen the code piece yet, which actually loads the rev limiter word. I guess its loaded via the dptr but i didn't find any hardcoaded pointer (close) to this eprom position.

Byebye