Moderated Discussion Areas
ContinuousWave: Small Boat Electrical
Modern NMEA-2000 Engine Fuel Data; Fuel Tank Levels
|Author||Topic: Modern NMEA-2000 Engine Fuel Data; Fuel Tank Levels|
posted 06-28-2015 09:21 AM ET (US)
Modern outboard engines typically provide data about their rate of fuel consumption as part of their NMEA-2000 engine data. I have never seen any documentation of the particular method used by a particular modern engine to obtain this flow rate. I presume the method involves computation of a rate of fuel flow based on the engine speed, the throttle position, the number of cylinders, the engine temperature, the barometric pressure, the exhaust back pressure, and the engine's management module algorithm for fuel-air ratio under those conditions, which would be based on the fuel amounts believed to be provided by the fuel injectors based on their operating cycle and their individual parameters. As far as I know, there is no actual measurement device or instrumentation, such as a spin turbine sensor in a fuel line, in a typical modern engine that measures the actual flow of fuel.
The data reported by a modern engine is most likely based on computation from the parameters I mentioned above. The data reported is the current rate of fuel flow. An NMEA-2000 engine reports the fuel flow rate in units of 0.0001 cubic-meters/hour to the network in the parameter group number (PGN) 127489, Engine Parameters, Dynamic. It does not report the volume in gallons-per-hour. The frequency of the reporting is around 1-Hz, that is, once per second.
A NMEA-2000 display device receives the flow rate from the network. The display device converts the data into the unit preferred by the operator. If the operator prefer to see the flow rate in gallons per hour, the display converts to that unit of measurement. Other units may be available, such as liters per hour.
Fuel tank levels can be computed using this flow rate data, but typically the modern engine does not perform any sort of calculation or manipulation of data about a fuel tank level. The general method of finding a fuel tank level from fuel flow rate data is an indirect method of measuring fuel tank level.
Fuel tank levels are typically tracked by some sort of dedicated function, often called a FUEL MANAGER, that is part of a NMEA-2000 display device. The FUEL MANAGER receives the fuel flow rate data from the modern engine via the network. The FUEL MANAGER integrates the flow rate over time to compute a FLOW VOLUME. The FLOW VOLUME is deducted from the previously calculated FUEL REMAINING (in tank) value, computing a new FUEL REMAINING value. The value of FUEL REMAINING is compared to TANK VOLUME, and a TANK LEVEL is computed. Again, all of this happens away from the modern engine and its engine management module.
The accuracy of a TANK LEVEL is dependent on the integrity of the data entered by the operator when any fuel is added to the tank. If the operator is diligent and properly informs the FUEL MANAGER when fuel has been added to the tank, the computed FUEL REMAINING and TANK LEVEL should be accurate.
There are many possibilities for error to enter this system:
--failure of operator to properly configure FUEL MANAGER
--failure of operator to diligently enter fuel volume added to tank
--failure of the device containing the FUEL MANAGER function to get data from the modern engine due to various problems
--failure of the FUEL MANAGER to properly integrate FLOW RATE over time to get FLOW VOLUME
I have simultaneously used two systems which employ engine flow rate data from an NMEA-2000 network to compute fuel used volume. Their computed fuel used volumes involving perhaps 50-gallons of fuel tend to differ by perhaps 0.25-gallons at most, when they are working properly. When not working properly, the computed volume of fuel from an indirect measurement system like this can be way off, that is, way off as in tens of gallons off.
Having said that, and having used several systems like this, I still retain a mechanical, direct-indication fuel tank level float gauge to tell me the fuel tank level as a second source of this data. That is the source I trust the most.
When you have just filled the tank to its capacity and you reset the FUEL MANAGER to the tank capacity, and you have configured the FUEL MANAGER with the actual tank capacity, the accuracy will be quite good, as long as:
--you properly entered the fuel tank capacity into the FUEL MANAGER
--you actually added fuel to actually fill the tank to capacity.
It often happens that adding fuel to a tank fails to actually fill the tank to full tank capacity due to various problems in the arrangement of the tank, the location of the filler hose, the trim on the boat, and the location of the tank vent.
For more about the engine data provided to a NMEA-2000 network by a modern engine, see:
NMEA-2000 Parameter Groups
For more about the measurement of a fuel tank level using the indirect method, see
Fuel Tank Level Measurement
posted 06-28-2015 09:33 AM ET (US)
An additional source of error in the reporting of rate of fuel flow by a modern engine may be a defective fuel injector. A fuel injector that malfunctions may not be delivering the amount of fuel intended based on its operating cycle and original performance. For example, if debris blocks the fuel injector orifice and prevents complete closure, additional fuel could be flowing into the engine. Or, if debris blocks delivery of fuel to the injector, less fuel might be flowing than intended. Discrepancy between the actual rate of fuel flow from the intended rate would not be reported by a modern engine which was providing fuel flow rate data based on calculations rather than actual measurement.
posted 06-28-2015 09:36 AM ET (US)
For an example of a source of error in computation of flow rate data into flow volume, see the long discussion about errors in the Lowrance system implemented with their EP-85R Data Storage Device:
Lowrance Fuel Management Problems
posted 07-06-2015 05:18 PM ET (US)
Can a Garmin GMI 20 use the [E-TEC EMM] as the source for fuel usage? Garmin sell a fuel flow metering unit. I am assuming it can also estimate the tank level without a tank level sender.
posted 07-06-2015 09:07 PM ET (US)
The GARMIN GMI-20 would not get data from the E-TEC EMM, but rather from a NMEA-2000 network. The E-TEC sends NMEA-2000 PGN 127489, Engine Parameters, Dynamic, to the network, and that PGN contains the fuel flow rate data.
I checked the GARMIN literature for the GMI-20 at
On page 4 it notes that PGN 127489 is supported.
I checked the Garmin owner's manual for the GMI-20. It was rather sparse and did not specifically say that the fuel flow rate data can be configured to come from the engine NMEA-2000 data. I suspect that it can, but you'd better contact Garmin technical support to get clarification.
posted 07-07-2015 10:29 AM ET (US)
As for the GMI-20 computing fuel tank level with the indirect method, again, the operating instructions are really very limited and don't specify this explicitly. Better check with Garmin about that, too.
posted 07-07-2015 10:40 AM ET (US)
Garmin publishes a document about the technical details of their NMEA devices. See
Even this document seems ambiguous about describing the source for fuel flow rate data to be used in fuel economy computations. In a large table under the heading of
General NMEA 2000 Data Type Requirements
they list a data type of FUEL FLOW RATE, indicate that it would come from PGN 127489, but describe the typical sensor as a FUEL FLOW SENSOR. All of the other data contained in PGN 127489 are listed and the typical sender is shown as NMEA-2000 COMPATIBLE ENGINE. That seems very curious.
Powered by: Ultimate Bulletin Board, Freeware Version 2000
Purchase our Licensed Version- which adds many more features!
© Infopop Corporation (formerly Madrona Park, Inc.), 1998 - 2000.
Powered by: Ultimate Bulletin Board, Freeware Version 2000