Delay in fuel flow indication with increase in RPM or percentage power

[Jamal haider]

We are facing a problem of delay in fuel flow indication with increase in RPM or percentage power. We havel Lycoming engine IO-540 with Skyview suite.
The problem is that during operation, whenever the throttle is advanced or retarded the fuel flow increases, but it takes approximately 10-12 seconds to reach the desired value.

I am attaching a picture for understanding. In the picture you can observe that RPM rise from 2000 to 2600 in 5 secs whereas the FF takes almost 12-14 secs to stabilize. The same issue is encountered once the throttle is retarded. The fuel flow sensor I use is FT-60 Red Cube and it is installed as recommended by the OEM.
Any help in this regard would be highly appreciated.

 

[airguy]

When you go into the setup menus for the sensor setup, you can select how much filtering is done for the income data to smooth it - this an averaging technique that prevents very jittery data and gives a smoother output. Sometimes you want that (like a fuel level sensor) and sometimes you don't (like oil pressure or fuel pressure). If I remember correctly there are a couple different levels of smoothing for this data and you will need to set it at the lowest level. In my airplane the fuel flow will lag throttle and mixture movements perhaps 2-3 seconds and stabilize, that's realistically the best you are going to get.

 

[Jamal haider]

When you go into the setup menus for the sensor setup, you can select how much filtering is done for the income data to smooth it - this an averaging technique that prevents very jittery data and gives a smoother output. Sometimes you want that (like a fuel level sensor) and sometimes you don't (like oil pressure or fuel pressure). If I remember correctly there are a couple different levels of smoothing for this data and you will need to set it at the lowest level. In my airplane the fuel flow will lag throttle and mixture movements perhaps 2-3 seconds and stabilize, that's realistically the best you are going to get.

Are you talking about the K-factor of the fuel flow sensor, because I could only find this parameter which can be changed for the sensor in setup?
For my case, the sensor has been provided with a range of k factor from 62000-68000 which is changeable and it is working on 68000.

 

[airguy]

No, not the K-factor - this is a value that you can tweak in the sensor definition file (.SFG file extension). This tells Skyview how any individual sensor works, and how to interpret the data from that sensor and display it. There is a value there called "Shift of X" where X is a number, used as a power of 2, representing 1/16 of a second, and the computed value affects how long the smoothing/averaging takes place. A smaller number for X will give faster changes, but may also create jittery data - there is a happy medium somewhere there that you will need to experiment with.

A sample sensor file is attached here to look at. You'll need to download your sensor file on a USB stick to modify it on your computer and then upload it again to Skyview.

 

[Jamal haider]

Hey man, thanks for your help in this regards. I tried changing the value, the default value was 4, i changed it to 3 and then 2, loaded for each case but I observed same response.

 

[airguy]

Hey man, thanks for your help in this regards. I tried changing the value, the default value was 4, i changed it to 3 and then 2, loaded for each case but I observed same response.

If you have any voids (air or fuel vapor bubbles) in the fuel system downstream of the pumps, this may in fact be legitimate behavior as far as the flow transmitter can tell. Those voids would act like a pressure reservoir, and it would take a couple seconds to fill or empty that void area as the engine demand changes, and the fuel flow changes with it. Is it possible you have any places that an air bubble might be trapped, like a line rising up to deadhead on a pressure transducer or something like that?

 

[baldcaldwell]

If you have any voids (air or fuel vapor bubbles) in the fuel system downstream of the pumps, this may in fact be legitimate behavior as far as the flow transmitter can tell. Those voids would act like a pressure reservoir, and it would take a couple seconds to fill or empty that void area as the engine demand changes, and the fuel flow changes with it. Is it possible you have any places that an air bubble might be trapped, like a line rising up to deadhead on a pressure transducer or something like that?

Regarding air vapors in the lines to the transducers... I'm curious about this as well. My fuel and oil pressure sensors are located on and at the top of the firewall. The engine has not yet been started, but we are close. So no oil or fuel is in the lines. But how do the lines charge (after running)? I wondered about this as I was installing the lines to the transducers. Do the lines need to be cracked with the engine running to push the air out until the fluids hit the sensors? What ensures air leaves these lines? Thanks!

 

[Jamal haider]

If you have any voids (air or fuel vapor bubbles) in the fuel system downstream of the pumps, this may in fact be legitimate behavior as far as the flow transmitter can tell. Those voids would act like a pressure reservoir, and it would take a couple seconds to fill or empty that void area as the engine demand changes, and the fuel flow changes with it. Is it possible you have any places that an air bubble might be trapped, like a line rising up to deadhead on a pressure transducer or something like that?

We have an aircraft with Skyview and another with a similar suite of Garmin. Both aircraft have the same fuel sensor Ft-60 but their engine monitoring system and displays are different. The one with Garmin is not giving this delay.
We have checked the fuel system performed its purging but issue persists.
What you told me is that we can change Shift of x in sfg file. I tried changing the value but it had no impact on the system. Then we observed that the sfg file has CAN_FUEL FLOW without any pin definition which we assumed is not our concerned value. Then we added another sensor definition for FUEL FLOW with pin definition and tried varying the shift but no response. Then we tweaked age and window also but still no change in the response. It seems like that the sensor definition is not actually changing even if we change and load it.
Also we observed that the voltage value in sensor debug window is also coming at the same rate.