Instrumentation project for MBZ diesel
Instrumentation project for MBZ diesel
08-10-2013, 02:26 PM
#1
I am new in the Forum
My Project is to install instrumentation for real time monitoring of fuel consumption (and more variables in the future) using existing sensors in my 1993 W124 OM603 turbo Mercedes.
Any one has done this before? If so, show me the thread. I didn´t find one.
The car has VSS signal available. The difficult part is to measure actual fuel consumption per time unit.
My approach seeks getting a lt/hour variable to combine with VSS and getting km/lt or Miles/gallon figures, in real time.
This info is very useful to evaluate mods, and the majority of improvements of power also determine an improvement in efficiency.
Looking forward for comments,
Oldbeaver
Hello guys,
I am new in the Forum
My Project is to install instrumentation for real time monitoring of fuel consumption (and more variables in the future) using existing sensors in my 1993 W124 OM603 turbo Mercedes.
Any one has done this before? If so, show me the thread. I didn´t find one.
The car has VSS signal available. The difficult part is to measure actual fuel consumption per time unit.
My approach seeks getting a lt/hour variable to combine with VSS and getting km/lt or Miles/gallon figures, in real time.
This info is very useful to evaluate mods, and the majority of improvements of power also determine an improvement in efficiency.
Looking forward for comments,
Oldbeaver
08-11-2013, 06:26 PM
#3
(08-11-2013, 10:18 AM)aaa I suspect fuel consumption can be directly derived from the pump's internal rack position and engine speed. There is an existing rack position sensor on the pump.
The alternative would be installing fuel flow meters on both the supply and return lines.
Dear Rolf,
I already tried to use the two flow meters method. They got stuck very son due to the debris floating in the fuel, despite an extra filter they had. The problem may have been that the measuring mecanism was very tiny. The thing worked only for a month or so, I couldn´t have time refine the equations in Arduino.
The other method you mention makes sense to me.
The car has a VSS signal in km/h. Else, I need fuel consumption in lt/h, right? Dividing the first by the second I am done.
For calculating consumption in fuel I have the internal governor rack position, which will tell me how open it is at any time. In one revolution, the injection will last around 1 to 40 milliseconds (depending of rpm), lap of time every injector will inject a tiny amount of fuel.
There are 3 injections per revolution in a 6L engine.
So I need to know the rack governor signal (opening), multiply it by the time each injector is open (depends on rpm), multiply the result by the maximum possible fuel the Bosch injector is capable, multiply the result by 3, to know the fuel injected each revolution, "fr".
Then multiply fr x rpm will give me the fuel injected each minute.
So the equation will be:
fr = L7 * Id * qmax * 3
where:
fr = fuel injected in each rev in lt
L7 = governor rack position
Id = injection duration
gmax = máximum injection quantity of a Bosch injector
3 = number of strokes per revolution of the engine
And total fuel injection volumen will be:
Q = fr * rpm
where:
Q = total injected fuel volumen per minute
fr = fuel injected each revolution
rpm = revs per minute of the engine
Injection duration is calculated as a function of rpm.
For example, if the engine is rotating at 2000 rpm, the time available for each injector is 1/3 * 2000 minutes = 1/6000 min.
As each minute has 60 sec, the time in seconds will be 1/6000/60 =
1/100 sec, 10 miliseconds if I am not wrong.
Opening of the governor rack will depends on many factors, all of them managed by the ECU which gives an output signal that controls the governor, air mass input at the turbo, air temp of this air, coolant temp, rpm, governor position, throttle position.
This means that the amount of fuel injected in each stroke will differ.
So, I need to measure L7 signal in each revolution. This is my only problem: have a measure of L7 in each rev.
For that, I need a tailored electronic circuit to take samples of L7 each revolution, calculate the equations above to determine fuel injected per revolution and per minute .
I don´t know if there is a device able to sample L7 each rev. I don´t know how to construct such a circuit. Yet.
Any comments, ideas?
Oldbeaver
This post was last modified: 08-11-2013, 06:37 PM by oldbeaver.
(08-11-2013, 10:18 AM)aaa I suspect fuel consumption can be directly derived from the pump's internal rack position and engine speed. There is an existing rack position sensor on the pump.
The alternative would be installing fuel flow meters on both the supply and return lines.
Dear Rolf,
I already tried to use the two flow meters method. They got stuck very son due to the debris floating in the fuel, despite an extra filter they had. The problem may have been that the measuring mecanism was very tiny. The thing worked only for a month or so, I couldn´t have time refine the equations in Arduino.
The other method you mention makes sense to me.
The car has a VSS signal in km/h. Else, I need fuel consumption in lt/h, right? Dividing the first by the second I am done.
For calculating consumption in fuel I have the internal governor rack position, which will tell me how open it is at any time. In one revolution, the injection will last around 1 to 40 milliseconds (depending of rpm), lap of time every injector will inject a tiny amount of fuel.
There are 3 injections per revolution in a 6L engine.
So I need to know the rack governor signal (opening), multiply it by the time each injector is open (depends on rpm), multiply the result by the maximum possible fuel the Bosch injector is capable, multiply the result by 3, to know the fuel injected each revolution, "fr".
Then multiply fr x rpm will give me the fuel injected each minute.
So the equation will be:
fr = L7 * Id * qmax * 3
where:
fr = fuel injected in each rev in lt
L7 = governor rack position
Id = injection duration
gmax = máximum injection quantity of a Bosch injector
3 = number of strokes per revolution of the engine
And total fuel injection volumen will be:
Q = fr * rpm
where:
Q = total injected fuel volumen per minute
fr = fuel injected each revolution
rpm = revs per minute of the engine
Injection duration is calculated as a function of rpm.
For example, if the engine is rotating at 2000 rpm, the time available for each injector is 1/3 * 2000 minutes = 1/6000 min.
As each minute has 60 sec, the time in seconds will be 1/6000/60 =
1/100 sec, 10 miliseconds if I am not wrong.
Opening of the governor rack will depends on many factors, all of them managed by the ECU which gives an output signal that controls the governor, air mass input at the turbo, air temp of this air, coolant temp, rpm, governor position, throttle position.
This means that the amount of fuel injected in each stroke will differ.
So, I need to measure L7 signal in each revolution. This is my only problem: have a measure of L7 in each rev.
For that, I need a tailored electronic circuit to take samples of L7 each revolution, calculate the equations above to determine fuel injected per revolution and per minute .
I don´t know if there is a device able to sample L7 each rev. I don´t know how to construct such a circuit. Yet.
Any comments, ideas?
Oldbeaver
10-30-2013, 09:01 AM
#4
I am using the Christian Zeitnitz AudioScope for studying the signals of the OM603 ECU.
For that I use a homemade set of probes I made putting together old staff.
The sample graph attached is of the rpm signal.
Salute,
OldBeaver
Attached Files
Image(s)Hello all,
I am using the Christian Zeitnitz AudioScope for studying the signals of the OM603 ECU.
For that I use a homemade set of probes I made putting together old staff.
The sample graph attached is of the rpm signal.
Salute,
OldBeaver
Attached Files
Image(s)
Users browsing this thread:
4 Guest(s)
4 Guest(s)
Users browsing this thread:
4 Guest(s)
4 Guest(s)