Welcome to SCoA! 
All I know is when there are no bars left on the fuel gauge you need to fill up....
All I know is when there are no bars left on the fuel gauge you need to fill up....
CAN bus, fuel level, odometer
Hi everyone,
I am new in SCoA, and this my very first post here.
I own a 2015 ForTwo 451 and, as electronics engineer, I have not resisted (since some months ago) to the temptation of listening the traffic on the CAN bus (the one broadcasted by the different modules) and also performing some request through PID OBD2 codes. I have had some degree of success with parameters like status of lights, gear selected, doors' status, etc.
But on the other hand, other parameters looks to be hidden very deep, (or I have missed something...:shrug
In this regard I have couple of questions that, hopefully, some of you could answer or just clarify the point a little for me:
1 - Is there any known ID message (proprietary message, of those broadcasted by the modules) that carries the fuel level information, same shown in the instruments cluster by the eight steps fuel bar at the bottom? Or, is there any way I could get that information (perhaps a query code needs to be put on the bus)?
My understanding is that this information reaches the A1 instrument cluster through the CAN bus, so it should be carried by one the messages present on it...
2- A similar question, but in this case for the odometer. I mean, not the total number of kilometers accumulated by the car during its life, but just the trip distance. I have identified the total km on the CAN bus already, but its resolution is 1 Km. The trip accumulated distance increments are 0.1 Km, more accurate...
If some of you have this information and is willing to share it (BTW, this is for personal use, not commercial at all), or just point me in the right direction, it is appreciated in advance very much..!
Thanks,
NorJim
I am new in SCoA, and this my very first post here.
I own a 2015 ForTwo 451 and, as electronics engineer, I have not resisted (since some months ago) to the temptation of listening the traffic on the CAN bus (the one broadcasted by the different modules) and also performing some request through PID OBD2 codes. I have had some degree of success with parameters like status of lights, gear selected, doors' status, etc.
But on the other hand, other parameters looks to be hidden very deep, (or I have missed something...:shrug
In this regard I have couple of questions that, hopefully, some of you could answer or just clarify the point a little for me:
1 - Is there any known ID message (proprietary message, of those broadcasted by the modules) that carries the fuel level information, same shown in the instruments cluster by the eight steps fuel bar at the bottom? Or, is there any way I could get that information (perhaps a query code needs to be put on the bus)?
My understanding is that this information reaches the A1 instrument cluster through the CAN bus, so it should be carried by one the messages present on it...
2- A similar question, but in this case for the odometer. I mean, not the total number of kilometers accumulated by the car during its life, but just the trip distance. I have identified the total km on the CAN bus already, but its resolution is 1 Km. The trip accumulated distance increments are 0.1 Km, more accurate...
If some of you have this information and is willing to share it (BTW, this is for personal use, not commercial at all), or just point me in the right direction, it is appreciated in advance very much..!
Thanks,
NorJim
1 - 10 of 10 Posts
Fuel level input is a voltage signal from the fuel level sensor to the DME (engine control module). That is interpreted and sent to the instrument cluster over the CAN bus to display a fuel level, as far as I can tell. The DME needs to know the fuel level so it can determine whether to run a on-board evaporative emission system test. Both the Snap-On and the OE SDS diagnostic tools show the same thing.
Attachments
Rustedwrench,
Just by chance, have you noticed if the voltage/fuel-level relation is inversely proportional? If so, I guess (probably, hopefully) I got something...
The picture attached shows the change observed on a two-bytes field taken out from one of the many CAN sentences periodically broadcasted by the modules through the CAN bus. To have a good data sample containing the desired information following a predictable pattern easy to identify later, I recorded everything was present on the CAN bus during an entire fill up process. Based on that, I began looking for a two-bytes fields showing the kind of progression that I would expect in the value: a continous change (increasing or decreasing) from one steady value to another steady value.
I was looking for a two-bytes field exhibiting this behaviour based on the assumption that a 10-bits, or even a 12-bits ADC converter was used, which is quite common in many present day microcontrollers.
Originally, the fuel level was in bar number 2. After filling up, it was at 7. From your message, I notice there are discrepancies in the fuel bar / voltage read by the scanner (those you got are real facts, a real reference) and the "voltage" resulting from the mathematical process of the two bytes field I have been playing with. Honestly, I was expecting a 5 Volts reference voltage for the ADC. If such is the case, something is wrong in my math because I'm having greater values as you can see in the graph.
It is too early to state this is in fact the M3/3 output voltage, but the shape of the curve was exciting enough to give it a try..
Just by chance, have you noticed if the voltage/fuel-level relation is inversely proportional? If so, I guess (probably, hopefully) I got something...
The picture attached shows the change observed on a two-bytes field taken out from one of the many CAN sentences periodically broadcasted by the modules through the CAN bus. To have a good data sample containing the desired information following a predictable pattern easy to identify later, I recorded everything was present on the CAN bus during an entire fill up process. Based on that, I began looking for a two-bytes fields showing the kind of progression that I would expect in the value: a continous change (increasing or decreasing) from one steady value to another steady value.
I was looking for a two-bytes field exhibiting this behaviour based on the assumption that a 10-bits, or even a 12-bits ADC converter was used, which is quite common in many present day microcontrollers.
Originally, the fuel level was in bar number 2. After filling up, it was at 7. From your message, I notice there are discrepancies in the fuel bar / voltage read by the scanner (those you got are real facts, a real reference) and the "voltage" resulting from the mathematical process of the two bytes field I have been playing with. Honestly, I was expecting a 5 Volts reference voltage for the ADC. If such is the case, something is wrong in my math because I'm having greater values as you can see in the graph.
It is too early to state this is in fact the M3/3 output voltage, but the shape of the curve was exciting enough to give it a try..
Attachments
I suspect that is the case but I never had any reason to look. You could monitor the actual voltage at the DME connector pins to verify it. Pins 68 and 62, PNK and BRN/PNK. The DME is behind the right side interior panel in the cargo area, just ahead of where the access panel for the right rear lights is located. Click on the attachment to magnify it.
Attachments
1 - 10 of 10 Posts
