Bug description: Switching on the auxiliary (electric) fuel pump and advancing the throttle does not generate fuel flow; it only increases fuel pressure.
On some piston aircraft, before starting the engine from cold‑and‑dark, the checklist procedure to verify the auxiliary fuel pump requires switching on the pump, advancing the throttle, and observing a fuel‑flow peak.
This works with the old [FUEL] system but not with the modern [FUEL_SYSTEM].
Repro steps: Load any piston aircraft using [FUEL_SYSTEM] that is equipped with an auxiliary fuel pump (e.g., C172) on the apron, and follow these steps to check the fuel pump:
Please note that most piston aircraft (analog gauges or glass) use the simvar A:ENG FUEL FLOW GPH:n to display fuel flow. With the legacy [FUEL] system this simvar reports a value both when the engine is running and when the engine is off but the fuel pump is on and the throttle is advanced. With the modern [FUEL_SYSTEM], this simvar reports a value only when the engine is running, because there is no fuel flow when only the fuel pump is on.
This doesn’t sound like the same issue.
I haven’t dived in the code yet, but what’s reported here sounds like a design regression on the new fuel system that causes the fuel flow to not be evaluated when the engine isn’t running.
What you describe in your 2020 bug report seems more related to the way fuel flow is dispatched in different lines and how much fuel flow can a line take.
I fail to see the similarities, can you enlighten me?
Absolutely! In my test case in FS2020, there is a fuel line connecting to the engine, which was tested and at full power the fuel flow of the feed line and the RECIP ENG FUEL FLOW variables matched.
My problem was located in two cases:
Case 1 - low fuel pressure: when in engine was idling and the engine-driven pump was shut down, the electrical pump can only provide fuel at 50% the pressure that’s required for idle. In this scenario, fuel pressure and flow was correctly calculated at the feeder line but the RECIP ENG still burnt at idle.
Case 2 - before engine start-up: electrical pump was on, mixture was 100% open and fuel was flowing through the feeder line at the correct value. Still, the RECIP ENG variables showed zero flow.
So the common point between my report and what @Arantis_Sim is showing here is that the flow/pressure values in the final engine feed line don’t match the RECIP ENG values, they are completely ignored and the engine simulation runs at its ideal fuel flows.
Case 1 is more about engine combustion simulation when FF is insufficient then, no?
I assume you expect RPM to drop and engine to stall eventually?
After investigating the case of the C172 in 2020 vs 2024, I see what you mean for case 2. @Arantis_Sim So yes, the legacy fuel system shows an increase of FF in the ENG_FUEL_FLOW_GPH simvar that is used in the gauge, when switching aux pump on and increasing throttle.
This no longer works in the 2024 version of the C172.
However, I’m not yet sure if the issue is the modern fuel system of the 2024 C172 has been too simplified to show this or if this cannot be simulated at all with the modern fuel system.
The first thing I’m not sure to understand is if this FF increase is supposed to be temporary or not. On the C172 fuel system diagrams I could find, the FF sensor is placed in the fuel distribution valve that then dispatches fuel to injectors.
But there does not appear to be fuel return lines from that valve. (fuel return lines are upstream of this valve)
So is the FF gauge supposed to show an increase just for the time it takes for all lines to injectors to be filled and then drop back to 0?
On the 2024 version of the C172, if you look at FUELSYSTEM_LINE_FUEL_FLOW:8, which is the line that goes to the engine, the fuel flow increases when switching the auxiliary pump on.
But it’s almost instantaneous, as all lines are filled at 90% by default and FF drops back to 0 when all lines are filled.
And of course, the gauge would need to be modified to use something else than the engine fuel flow simvar that will not measure that increase.
Yes. I would expect the RPM to drop in accordance with the fuel flow and eventually stall the engine.
On the 2024 version of the C172, if you look at FUELSYSTEM_LINE_FUEL_FLOW:8, which is the line that goes to the engine, the fuel flow increases when switching the auxiliary pump on.
The line flow is always right. The question pertaining to case 2 and what I assume @Arantis_Sim is asking is, will the aircraft burn what is shown in FUELSYSTEM LINE FUEL FLOW:8, or will it burn what the RECIP ENG shows? Because if it burns what the legacy RECIP ENG A:var shows, then there will be a mis-match in fuel burn between modern and legacy fuel system. This is important in the event of fuel leaks or failures, as well as during start-up.
I still think it’s worth investigating separately then.
Please move your 2020 ticket to 2024 or create a dedicated one and provide us with your package so we can trace what’s happening in your specific case.
Our core systems team suggested that some of these engine simvars return what FF the engine is asking, not necessarily what the fuel system is able to provide. So the gauges relying on such simvars will not really be representative of what’s really happening.
Then the impact on the engine has to be investigated separately. I’m not sure what the effect on RPM will be as the FF is decreasing. I expect the engine to at least stall when fuel pressure drops below min_fuel_press_for_combustion_psf.