F-16 Fuel Flow Drift

[Glide]

I am aware that fuel flow will change with altitude and no throttle input, but is it supposed to be stable during straight and level flight?  My fuel flow always seems to drift slowly up or down even with AP on.  I have a deadzone of 3 in my throttle axis, and I am wondering if my throttle is the issue. 

[Glide]

After keeping a sharper eye on the behaviour, I see the fuel flow drifts after a throttle input until the airspeed levels off.  For example, if I am at 300 kts  with ff 3000 and I increase throttle until ff is 4000, the jet will accelerate and the ff will drift upwards until the airspeed levels off. 

[Terzi]

That's correct behavior because as the speed changes the blades will hit faster air, will spin easier and breathe more oxygen.

[Glide]

30 minutes ago, Terzi said:

That's correct behavior because as the speed changes the blades will hit faster air, will spin easier and breathe more oxygen.

True, but faster air means hotter intake and doesn't the engine limit the intake temps? 

[Glide]

Cool reference: https://www.grc.nasa.gov/www/k-12/airplane/turbfan.html

[Terzi]

9 hours ago, glide said:

True, but faster air means hotter intake and doesn't the engine limit the intake temps? 

I don't know about intake temperatures, but going faster means more volumetric air intake per given time, thus slight increase in PPH.

[Glide]

7 hours ago, Terzi said:

thus slight increase in PPH

Slight?  Seems huge.  Fly straight and level at 300kts with 3000pph.  Move the throttle to just before the AB detent.  Then watch the fuel flow creap up by over 1000pph.  This does not seem correct to me. 

Edited March 22, 2021 by glide

[Terzi]

Well maybe more than slight In low altitudes difference would be more and high altitudes less.

[Glide]

well, another week or so, and it could all change with 2.7.   Cheers!

[Glide]

I tested this on the Hornet as well.  The behaviour is there but not as pronounced.  Pull a turn and watch the fuel flow drop in the Hornet by 400pph in 360 degrees.  Try the same thing in the Viper.  Fuel flow drops like a rock.  I don't think either model is correct, but the Viper is at the far end of the spectrum.

[Machalot]

15 hours ago, glide said:

Slight?  Seems huge.  Fly straight and level at 300kts with 3000pph.  Move the throttle to just before the AB detent.  Then watch the fuel flow creap up by over 1000pph.  This does not seem correct to me. 

 

Check out Table 5 of this NASA report to see how much fuel flow varies with Mach number at constant altitude and power setting. 

 

For example, looking at Test Condition A, at ~3600 ft and 85 deg Power Lever Angle (PLA, the physical angle of the thrust lever), the fuel flow rate (WFE) varies from 10,789 lb/h at Mach 0.344 to 14,654 lb/h at Mach 0.758, about a 35% increase. 

 

This is for an F110-GE-129 mounted in an F-16XL airframe, so it should be pretty representative of what we have in DCS.

Edited March 23, 2021 by Machalot

[Frederf]

F-16 engine control DEEC is semi-sophisticated. The power lever angle (PLA, well EPLA, mechanical cable is a backup) is an input into a computer and the computer ramps the engine up and down until the engine performance is equal to commanded based on feedback from sensors. Famously when going proper fast (e.g. M1.6) idle PLA is still MIL thrust schedule internally to the engine.

 

What I don't know what the DEEC holds constant, if anything, at constant PLA value. I don't think it's thrust or RPM or FF. It might be fraction of thrust between idle and mil. Someone who knows the EEC logic would have to say.

[Glide]

I think there is some reporting issues on those tables.  The tests indicate the CTC were conducted by achieving stable flight, moving the lever to 85 PLA, and measuring the acceleration during climb.  However, the tests indicate the START, OH, and END positions were all 85 PLA.  This is because of the locations of the microphones and the point in the acceleration when the jet passed the recording zone. 

 

Quote

For the CTC condition, the pilot first stabilized the aircraft at the desired altitude and just below the desired Mach number. As the airplane approached the acoustic start point, based on a radio call from the control room, the throttle was advanced to the intermediate power setting. Then the engine was allowed to stabilize for approximately 5 sec before the start of the test run. The aircraft would accelerate depending on the degree of excess thrust through the desired test conditions in level flight. Some acoustic runs were initiated directly over the center of the array with the run terminating when the elevation angle was again 15° above the horizon

This doesn't make too much sense.  You deflect to PLA 85, WAIT 5 SEC for the engine to stabilize, then start the climb?  What's happening in those 5 seconds?  Did they mix that up?  Should that have been written, "achieve level flight just below the Mach, stabilize for 5 seconds, then deflect"?  Why test a high-NPR jet but not the afterburner acoustics?

 

In my mind, the thrust levers are mechanical in operation even though it's fly by wire. If the lever doesn't move the rate of injection doesn't move unless altitude changes the flow characteristics.  Otherwise, what's the point of physical AB detents and marking idle cutoff and idle on the throttle?  I am flying along, I move the fuel flow to 3000, but my throttle is not on the mark. 

 

I think this is why folks say they don't feel the sense of speed in the game.  I think it's not the sense of speed they miss, but the sense that there's a rocket attached to your chair and it wants to accelerate.  I think if fuel flow was "stable", we would feel the jet's desire to accelerate more as we bleed off energy in turns and such.  Right now it feels like a fast airliner.

[Machalot]

10 minutes ago, Frederf said:

What I don't know what the DEEC holds constant, if anything, at constant PLA value. I don't think it's thrust or RPM or FF. It might be fraction of thrust between idle and mil. Someone who knows the EEC logic would have to say.

If there is any consistency in terminology (which is not common in aeronautics) the constant of PLA would be a fraction of delivered power (thrust times speed). But who knows. 

[Machalot]

32 minutes ago, glide said:

I think there is some reporting issues on those tables.  The tests indicate the CTC were conducted by achieving stable flight, moving the lever to 85 PLA, and measuring the acceleration during climb.  However, the tests indicate the START, OH, and END positions were all 85 PLA.  This is because of the locations of the microphones and the point in the acceleration when the jet passed the recording zone. 

 

This doesn't make too much sense.  You deflect to PLA 85, WAIT 5 SEC for the engine to stabilize, then start the climb?  What's happening in those 5 seconds?  Did they mix that up?  Should that have been written, "achieve level flight just below the Mach, stabilize for 5 seconds, then deflect"?  Why test a high-NPR jet but not the afterburner acoustics?

The text and data clearly say it's a constant altitude test.  The 15 deg elevation is probably referring to the elevation angle from the microphone array to the aircraft. 

 

 

Edited March 23, 2021 by Machalot
Split into two replies

[Machalot]

37 minutes ago, glide said:

In my mind, the thrust levers are mechanical in operation even though it's fly by wire. If the lever doesn't move the rate of injection doesn't move unless altitude changes the flow characteristics.  Otherwise, what's the point of physical AB detents and marking idle cutoff and idle on the throttle?  I am flying along, I move the fuel flow to 3000, but my throttle is not on the mark. 

 

I think this is why folks say they don't feel the sense of speed in the game.  I think it's not the sense of speed they miss, but the sense that there's a rocket attached to your chair and it wants to accelerate.  I think if fuel flow was "stable", we would feel the jet's desire to accelerate more as we bleed off energy in turns and such.  Right now it feels like a fast airliner.

Do you have a source for this theory on how the throttle works, or is it just assumptions you made up? Because if the throttle were to just set constant WFE, the delivered thrust would be lower than how it works right now as the jet increases Mach, due to not increasing the WFE with Mach. 

Edited March 23, 2021 by Machalot

[Glide]

24 minutes ago, Machalot said:

The text and data clearly say it's a constant altitude test.  The 15 deg elevation is probably referring to the elevation angle from the microphone array to the aircraft. 

My understanding is there was three batteries of tests: ground, climb to cruise, and straight and level.  My understanding is they deflected to 85 and let the jet climb to 15 deg elev in the CTC tests which were acceleration tests. 

[Machalot]

6 minutes ago, glide said:

My understanding is there was three batteries of tests: ground, climb to cruise, and straight and level.  My understanding is they deflected to 85 and let the jet climb to 15 deg elev in the CTC tests which were acceleration tests. 

No:

Quote

To satisfy dual objectives of the program, two flight tests were conducted: subsonic CTC noise deter-
mination and ANOPP predictive code validation. ...

The flight matrix for the CTC runs consisted of level flight accelerations over the acoustic array at
various Mach numbers and altitudes. ...

 

The data in Table 5 also shows constant altitude. Where are you getting the idea that they climbed?

Edited March 23, 2021 by Machalot

[Glide]

6 minutes ago, Machalot said:

Do you have a source for this theory on how the throttle works, or is it just assumptions you made up

No sources.  I am a pilot, and thermodynamics was one of my favourite classes.  The details of the DEEC are not my concern.  I am concerned with the realism of the flight model.  I am assuming there's some magic involved, but the simulation doesn't care.  Once per game loop, DCS will update the sprites with a formula that allows the fuel flow to vary with airspeed.  This results in a certain feeling of flight.  I am suggesting perhaps fixing the fuel flow variable and getting a much different feeling of flight. 

Edited March 23, 2021 by glide

[Glide]

9 minutes ago, Machalot said:

Where are you getting the idea that they climbed?

Climb to cruise.  CTC.  Note the change in Mach and exhaust volumes in this set.  They started at those altitudes, deflected, and let the jet climb to cruise where it was about 15 deg nose up.  That's my understanding of the CTC data in table 5.  Table 4 was ground.  Table 6 was straight and level.  Yes?

[Machalot]

Just now, glide said:

Climb to cruise.  CTC.  Note the change in Mach and exhaust volumes in this set.  They started at those altitudes, deflected, and let the jet climb to cruise where it was about 15 deg nose up.  That's my understanding of the CTC data in table 5.  Table 4 was ground.  Table 6 was straight and level.  Yes?

No. Where are you reading that they deflected and climbed?  I think you're just not reading the text. It says in two different places that the CTC runs were level flight, and the altitudes in Table 5 reflect this. 

[Glide]

33 minutes ago, Machalot said:

I think you're just not reading the text

Good catch.  That 15 deg was the angle to the sensor array.  That must have been a very gradual deflection to 85 as the mach increases and the nozzle volumes increase at the same time, don't they?  Edit.  Nevermind.  It's a mystery to me how they got those WFE numbers.  Fun report.  I was a hotshot data center architect in '96.  I love how they say "deck".  They probably mean a deck of punch cards.  We still had them back then. 

Edited March 23, 2021 by glide

[Glide]

Figure 9 is interesting.  WFE increases as N1 RPM increases while N2 RPM stays relatively constant.  N2 is core, so the deflection was constant.  And N1 is the fan, so while the core was pushing the mach increase, the fan speeds up and draws more fuel flow?  This makes sense, but shouldn't this increase the rate of change in the mach?, ie it should feel like acceleration.  More like a curve, less like a straight line.

 

What about deceleration?  When you slow down the fan slows down, should the fuel flow drop so quickly so far in a min radius turn, for example?  Or would you just lose the N1 fuel flow bonus and stay at full core fuel flow with 85 pla. 

Edited March 23, 2021 by glide

[Machalot]

24 minutes ago, glide said:

but shouldn't this increase the rate of change in the mach?, ie it should feel like acceleration.  More like a curve, less like a straight line.

As you know, there are two terms in the left hand side of F=ma, drag and thrust.  Dynamic pressure increases with airspeed squared, and on top of that the Cd increases as Mach increases. So there's a lot more drag to take a bite out of that increased thrust. 

 

As for the feel, any change in speed is acceleration, whether it's a linear increase or otherwise, and will feel like acceleration.  

[Glide]

You can fly those tests easily in DCS.  Start at mach .3, deflect to 85 pla, watch fuel flow until mach .75. 

 

I was watching the Grim Reapers turn tests on YT today.  I'm going to try some similar tests with the mach increase and without the mach increase.