No announcement yet.

Universal ECU abd OBD II with sensors for retrofitting?

  • Filter
  • Time
  • Show
Clear All
new posts

  • Universal ECU abd OBD II with sensors for retrofitting?


    so I've been thinking. I'm restomodding a classic Vespa scooter, which is being tricked out with all sorts of gizmos. One thing I really desire however is to toss out all instruments and use a smart phone instead. There are several OBDII bluetooth solutions that can be combined with an IOS or Android app to display all sorts of info.

    problem is that a scooter does not have a single ECU, or OBD, which must be retrofitted. The only sensor is the fuel gauge sender. Is this project feasable, and if so, what would be required? Is there a different solution than the ECU/OBD route? Please keep in mind that a scooter does not provide a lot of jucie to power electronics.

    What I want as a minimum is:
    -Fuel Gauge

    Additional wants is:
    -exhaust temp
    -cylinder head gauge
    -----and a lot more

  • #2
    Hi and welcome to the forum. Looks like you are taking on a unusual project. I think phidgets would be a good choice for this project. good luck and start a worklog to document your journey. SNO


    • #3
      Forget using a conventional ECU/PCM and adapting it.

      You need a multiple analog input data logger.
      link to my CTS-V project:


      • #4
        Thanks for the input - please keep it coming.

        The multiple input data loggers looks like it definantly could do what I need as they seem quite easy to set up even for someone without programming skills. But I wouldn' want to use any of the displays which they are designed for. What I really want is to use the Torque app for Android to display the data on a smartphone. This app gets its signals from an OBD bluetooth adapter, which plugs into the OBD... which I don't have.

        If I've understood correctly, the multiple input data loggers serves kind of like an OBD in this case? Now, please educate me, do these dataloggers share the same programming standard as OBD-II? I can't see that any of them has an OBD port (only ethernet or wireless export or streaming to PC)... Also, can these dataloggers log Fuel Level?

        Click image for larger version

Name:	android-torque-app.jpg
Views:	1
Size:	81.7 KB
ID:	2287896

        For those interested, I've scissored from Fat Charlie on the forum the sensors that Torque will display:
        0-100kph time
        0-60mph time
        1/4 mile time
        1/8 mile time
        Acceleration sensor (total)
        Acceleration sensor (x axis)
        Acceleration sensor (y axis)
        Acceleration sensor (z axis)
        Ambient air temp
        Barometric pressure
        Catalyst temperatire (bank 1, sensor 1)
        Catalyst temperatire (bank 1, sensor 2)
        Catalyst temperatire (bank 2, sensor 1)
        Catalyst temperatire (bank 2, sensor 2)
        EGR commanded
        EGR error
        Engine coolant temperature
        Engine load
        Engine oil temperature
        Engine RPM
        Fuel level
        Fuel pressure
        Fuel rail pressure (diesel)
        Fuel trim bank 1 long term
        Fuel trim bank 1 sensor 1
        Fuel trim bank 1 sensor 2
        Fuel trim bank 1 sensor 3
        Fuel trim bank 1 sensor 4
        Fuel trim bank 1 short term
        Fuel trim bank 2 long term
        Fuel trim bank 2 sensor 1
        Fuel trim bank 2 sensor 2
        Fuel trim bank 2 sensor 3
        Fuel trim bank 2 sensor 4
        Fuel trim bank 2 short term
        GPS accuracy
        GPS altitude
        GPS bearing
        GPS satellites
        GPS vs OBD speed difference
        Intake air temperature Intake manifold pressure
        Kilometers per litre
        Kilometers per litre (average)
        Litres per 100 kilometer
        Litres per 100 kilometer (average)
        Mass air flow rate
        Miles per gallon (average)
        Miles per gallon (instant)
        O2 volts bank 1 sensor 1
        O2 volts bank 1 sensor 2
        O2 volts bank 1 sensor 3
        O2 volts bank 1 sensor 4
        O2 volts bank 2 sensor 1
        O2 volts bank 2 sensor 2
        O2 volts bank 2 sensor 3
        O2 volts bank 2 sensor 4
        Speed (GPS)
        Speed (OBD)
        Throttle position
        Timing advance
        Transmission temperature (method 1)
        Transmission temperature (method 2)
        Trip distance
        Turbo boost & vacuum gauge
        Voltage (control module)
        Voltage (OBD adapter)

        The sensors I would add to my scooter:
        -Engine RPM
        -Fuel Level
        -Wheel Speed Sensor
        -Exhaust Temp
        -Cylinder Head Temp


        • #5
          Look at using one of the above and a USB bluetooth adapter.

          I would suggest you get off the idea of OBD/CANBUS unless you are gonna spend some time getting educated in what they are for and the protocols.
          link to my CTS-V project:


          • #6
            Thanks for input, please keep it coming, I am a long ways away it seems.

            I really want the torque app for android and none of the other off the shelf instruments/displays. If I use one of the above solutions, could it actually be so simple that all I would have to do would be to fit sensors, route these sensors to one of the solutions above, and then plug in a USB bluetooth dongle, let the torque android app handshake with the dongle, and voila (maybe just with some simple tweaks in the complementary software)? If I understand correctly, I need something which will translate the sensor input into OBD-II protocols, with all the proper PIDs. Do any of these solutions do this out of the box or am I looking into some serious programming?

            In my search I stumbled across which might resolve some of my problems, though I am not entirely certain it would really work as a stand alone. Although I can feed sensors into Proenium's PGE-1, and it will generate OBD-II code, and it does come with an OBD-II connector which I could plug a bluetooth dongle to, I'm not so sure it will work as a standalone solution. The examples I've seen all feed into CAN-network stream, and as such will join all the PIDs allready on that network. From what I've read, torque will query all theoretical PIDs, and keep repeating this process until data is found. Without an ECU and OBD2 on the network, 99% of these requests would not return with anything as the sensors don't exist and there is no existing obd or ECU telling it that they don't exist. The consequence would likely be a freeze or reboot of the app, or something similar - rendering it useless. Walk around for Raspberry PI users seems to be to simulate and return the message "no data" for all PIDs/Sensors not available. Then again, all the solutions I've seen so far all starts off with vehicles with at least some ECU and/or OBD on board, with Raspberry etc being used as a code reader or as a means to add custom sensors... they all latch on to the existing network. My project would not be a hybrid, but a "from scratch" system. I have written Proenium, and hopefully they will write back... and with something that makes sense to a newb like me.

            Do for instance data recorders render signals into something which can be read by for instance an OBD-II diagnostic tool, or are they all written for the manufacturers own unique hardware and software? None of the solutions I have seen, besides the one from Proenium actually comes with an OBD-II output port...

            I have failed to source an open source solution which matches my needs. It seems that the raspberry OBD developers don't share, or that they have not been able to get it to work for applications like mine. Question, could raspberry (or other solution) read sensors and convert this into something which Torque could tap into through a USB bluetooth dongle... without all the other ECU, CAN and OBD stuff usually found in a modern car? Simply: Sensor=> Raspberry (data recorder or something else) <=>USB Bluetooth Dongle <=> Torque APP???

            If what I write is confusing, that is because I am utterly confused and don't know what I am talking about at all :P so please shoot me straight. Hopefully your responses are useful to all those with classic vehicles who wants to attempt something similar, even if the conclusion is "don't look any further, drop all further research... it's not feasable"

            Please advice.