Erratic TPS Indication

The throttle position sensor information is typically transmitted from the Rotax ECU to the EFIS over the Rotax CAN bus, which is a particular type of data bus.  With Garmin and Dynon, there is an engine interface module that is separate from the EFIS where the CAN bus terminates.  From there, the engine interface module transmits the information on to the EFIS in one of several date formats. I looked at the GRT website and documentation for the Horizon and they don’t describe a data connection from their product to the Rotax CAN bus, but I assume they have an option for it and it’s just not posted on their site.  If your EFIS is getting other Rotax CAN bus data like oil temp, oil pressure, EGT, MAP, then we can probably assume the CAN bus connection is OK, and the problem is on the Rotax side. However, it might be a good idea to speak with GRT and see if they recognize the problem.  

Looking at the Rotax side of things, the TPS sensor is a three wire circuit that operates on a 0-5v signal.  These types of sensors are quite sensitive to poor connections.  You can check the connector at the TPS, which is mounted on the side of the throttle body.  From there the wires go directly to the ECU A2 connector, which is the middle connector on the ECU, so make sure that connector is properly latched.  You only get 20 unlpug/replug cycles on the ECU connectors so only disconnect it if you have exhausted other troubleshooting options.  When you remove any ECU connector you need to log it, and it’s a good idea to put a sticker on the connector with the number of cycles it has had. if the wiring looks good, and GRT does not have the answer, you could try a new TPS.  

Excellent advise thank you!  (I had no idea about the limited number of ECU connection cycles)

If the issue is on the Rotax side, either with the sensor itself or a loose connection, would the engine run differently or show any other signs of a problem?

All other CAN bus data is displaying properly on the EFIS so I don't *think* the issue is on the EFIS side but I'll double check with GRT.  A friend suggested pulling the EFIS data log and having them review it.

Thanks again,

Chris

Yes, pulling the EFIS data log is a good idea, as is connecting the Rotax BUDS software and looking at both the fault log and watching the TPS real-time from the BUDS screen on the computer.  

This is a good point you bring up about a TPS problem causing the engine to run differently, and also showing a fault via the lane indicator lights.  I believe the ECU needs accurate TPS data to execute proper fuel metering. So it seems logical that without TPS data the engine would revert to a failsafe fuel mapping and you would get a fault indication on Lane A. This line of thinking points me back in the direction of a display problem.  I don’t know much about the GRT architecture, but if there is an engine interface module that interprets the CAN bus data, then that part would be suspect.  

Be careful here...the TPS is not a stand alone part, it comes with the complete unit as it is calibrated by the supplier.  Before you start all this please connect to a BUDS diagnostic and check to see if this is something in your instrument wiring or in the actual ECU itself.  if this fault is shown in the ECU logs then proceed to get the logs looked at before you start trying to work with the wire harness.  

Cheers

Just a note on ECU connection times permitted.  This information is found in the installation manual.  The problem is the gold plating can wear out and then errors start to appear.  It might not give a problem until number 25 or 35 but they state you are good up to 20 only. It will all depend. 

Cheers