Turbine RPM "oscillating" under autothrottle control.

[old_wombat]

I have the Abacus Flight Deck 4 F18 Hornet. Under autothrottle, the engine RPM is constantly oscillating above and below the steady value. Makes an already difficult landing on the carrier even more so Is there any kind of .... don't know what you'd call it .... "engine damping" parameter in the aircraft.cfg that is worth changing to try to fix this?


Steve from Mudgee.

[oebanks]

I had the same issue with my ATR42-500 and I read somewhere that your Autothrottle max RPM in the aircraft.cfg file should be increased little by little until the issue is solved. It did work for me.

[old_wombat]

Tried that. It was set to 90 percent, I increased it all the way to 99 percent with no discernable difference. The problem is more noticeable at low to very low RPMs.

Steve from Mudgee.

[jgf]

Found this in my notes: "...try experimentally increasing the fuel_flow_gain parameter in the TurbineEngineData section of the aircraft.cfg file...a larger value makes the engine spool up/down faster. This can significantly reduce or stop A/Tspeed overshoots by damping out lag induced oscillations, thus improving A/T feedback performance".

Don't know who to credit for this (copied from a thread somewhere...sometime), but it worked for my F117 that surged like an old washing machine when trying to maintain 200kts at 6000ft.

[old_wombat]

OK, I changed that fuel_flow_gain parameter for all affected planes - IRIS F14A, IRIS F14B and Flight Deck F18. I went up to a value about 12 times the default. It made no discernable difference.

Having said that, it has increased the apparent performance of all three planes, to where I can now - AT LAST !!!!! - fly the the approach slowly and manually and land on the carrier in the arrestor zone. Not every time, but reasonably repeatedly.

The question now, and it may be academic, is - how realistic is this? Are these military turbine engines really that slow to respond (using the default value)? From what I saw and heard of the engines sounds during the PBS "Carrier" show, I think the faster response might actually be more realistic. And perhaps one could argue that trying to land on a carrier via autopilot is totally unrealistic in the first place.

Steve from Mudgee.

[jgf]

... Are these military turbine engines really that slow to respond (using the default value)? ....

With the exception of the better payware (where I assume they've "done their homework" and built the flight dynamics from scratch), I think many designers either copy&paste parameters from similar aircraft or just build on an existing aircraft, editing a few parameters to get the performance they want. But there are so many interactions between engine parameters, flight dynamics, etc. that it's nearly impossible to predict how all this mix-and-match of design elements will eventually perform - there are tables in the air files controlling such things as drag vs angle of attack, or engine efficiency vs altitude, that are rarely touched. All this leads to compromised designs where easily adjustable parameters are set to get the performance the designer feels is right.

In extreme cases you'll see aircraft with half their actual weight and several times their actual power just to get them to fly according to published specs for speed, rate of climb, etc. If this was necessary, you can imagine how far off their other design parameters are. Logic would dictate that since weight and power are easily available specs, those should be set to correct values from the start and performance tuned with other parameters.

So, when you encounter some "small" flaw don't hesitate to adjust a "non-specified" parameter to get things working (in relation to a real world F-18, just what is "fuel_flow_gain"? where would we obtain this value and how would it interpret into FS values? Microsoft's SDK isn't even helpful, their description of this parameter - "Fuel flow gain constant"). So I feel use whatever value the sim will accept and produces the results you seek; for example, the pitot_static value defaults to 2.0 but I've seen this set anywhere from 1-99 in various freeware and payware aircraft.