Joost

The speedometer sensor attached to the transmission is just a reed switch which grounds the sensor/signal wire from the tachometer. It will give 8 pulses per right rear wheel revolution.

Your question about extracting information from the temperature sensor is a question of impedance (or, if you will, resistance). Long ago simple meters to measure voltages and the like drew a lot of current, relatively speaking, from the circuits they measured, which distorted the signal, so to speak. Along came the vacuum tube volt meter - it drew a tiny current from the measured circuit, and used the vacuum tube as a switch to multiply the current needed by the meter. Modern digital volt/ohm meters have input values in the mega ohms, and don't load the circuits being tested.

I don't know all that much about Arduino inputs, but my guess is that they are similar, don't load the tested circuits, and you can easily look up the specifications. The same likely applies to reading speedometer pulses - the Arduino connection to the signal input on the speedometer shouldn't confuse the speedometer into thinking the input was always low or always high (open), despite pull up or down resistors in the Arduino circuitry or the speedometer circuitry.

The input impedance of the arduino analog input is 500 Mega Ohm, so it will draw a very little current.
Other dependency is the maximum voltage of 5V on the input.
For measuring frequency of a pulse signal, it works using a voltage divider (2 resistors), followed by a zenerdiode for protecting the arduino. This wont influence the pulse signal.
But the temperature is not a pulse signal, but a current through a temperature dependent resistor. In this case choosing the right values for the voltage divider will be crucial. We will have to do some tests without breaking things.

Just like reading the lambda sonde.
The original 1 wire Bosche Lambda Sonde, Reading the signal through a resistor with zenerdiode, pulldown resistor of 10 Mega Ohm, connected to the analog in of the Arduino.
After a couple of weeks programming and testing (in my spare time) i can read / use the signal.
Even the Arduino can decide itself when to switch to closed loop regulation.

I still have one (or maybe several) question in my head.

At which point do other DIY lambda controllers startup the system.
Because starting the Ardiuno takes several seconds. (3 - 5).

I power the Arduino with the parallel power relay under the seat.
So i switch on contact pos I (nothing happens)
Switch on contact pos II (generator light in dash on, Arduino starts)
As soon as i hear the frequency valve, or see on display dutycycle 50,
Turn the contact key to start the engine.

This is my practice, how have other DIY lambda controller inventors done this ?

The 2-3 second startup is arduinos default.

On switch Pos 2 you should not have current on the fuel pump! If yes then it's its not the factory default.
So as the ECU, TV, WUR and AAR are getting their current over the same wire from the fuel pump relay, I guess you did pull the plug behind the Air Sensor plate housing? This is a very bad idea, because its a safety construct where on accidents i.E., where the engine stops and key is still "on", no more fuel floats to the outside.

This is quite impressive work. Are there any plans to implement a MAP sensor, this would be helpful for turbo cars I would think.
David

The airflow sensor is still a hurdle for me to take.
I see in my own diagram the arduino should boot when the sensorplate starts the fuelpump. :D

From the beginning, my car is a 1976.
The engine is a 1983 930/16 and i was glad the wiring of the engine was complete.

In the diagram i see the air flow sensor with 1 brown wire to earth and 1 bruwn/black wire to the T2a connector.
But where is this connector? It should go to the fusebox in the front where the fuelpump relay is plugged.
You guess, the fusebox is from 1976.

How is the brown/black wire going in your car ?
I dont see the T14 or T12 connector in the wiring diagram for this wire.
While these are the big connectors with the engine wiring.

http://forums.pelicanparts.com/uploa...1602863929.jpg

http://forums.pelicanparts.com/uploa...1602863929.jpg

I will have to complete this connection myself, but i want to complete this as close to original as possible.

The sensorplate switch is wired.

http://forums.pelicanparts.com/uploa...1602972267.jpg

But how is it reaching the fusebox in the front ?
Or better: how should it reach the fusebox in the front ?

The Air Flow Sensor Switch is working correctly.
This is wired to the T2a connector below the fusebox in the engine compartment.
But due to the year of the car (1976) the brown/black wire is in the T1a connector (also below the fusebox in the engine compartment).
Now i created a brown/black wire of 10 cm from the T1a to the T2a connector.
This is correctly wired to the T2e connector below the fusebox in the front.
Which is wired to the fuel pump relay.

So, now my fuel pump starts running as soon as I turn the key from pos II to Start and the engine is cranking.
At that point the Arduino starts booting (3 sec), after which the signal to the frequency valve goes to 50%.

Is this the same way as others made this ?
Because the original OXS Control Unit has output to the frequency valve right at the power on.

Does the engine start without problems, then I wouldn't worry.
It should work as the first Second(s) the cold start valve will inject enough fuel into the manifolds air flow to make the engine start properly.