P-38 flight file follies - opinions and input sought
Guys, I am in the process of rebuilding / updating the historical 82nd Fighter Group campaign and completing the part 2 when the unit became part of the 15th Air Force. I have David Copley's permission to work with his P38 family and modify them for the campaigns.
Mav has already helped me with one problem where AI were not appearing on the runway. (Thanks buddy)
I am running into performance issues related to flying them in the campaigns. As the 82nd only flew P-38's, getting them to perform right is critical or it will become frustrating to fly the campaigns.
I thought getting some input from the gurus here might be helpful. Here are the issues and what I have tried so far to affect a fix.
Range:
The P-38s had a range of 900 to 1,100 miles. CFS2 models - DCC and the stocker can only make 450 miles. Critical issue for the campaign. Evidently the CFS2 fuel rate consumption calculation is about double what it should be. Things I have tried (working with the P-38G air files):
Changing the "Best Power Fuel Consumption" scalar in the aircraft.cfg file. I haven't found an equivalent value in the air file using Airwrench or AirEd. No results at all obtained from altering the .cfg file.
Adjusting the mixture and RPM produces minimal results, maybe 50 miles or so. Not useful.
Reducing the parasitic drag and the induced drag values in the air file, and the scalar values in the aircraft.cfg file, individually and in combinations with both. I can produce the range with about a 50% reduction in drag. However this also increases the aircraft top speed to 450 / 460 miles per hour. So this doesn't work either.
The only solution I have found workable for the campaigns is to double the fuel gallons per tank and decrease the empty weight of the aircraft by the weight of the additional fuel. This keeps the aircraft performance the same and doubles the range to 900 miles. It's not historically accurate for the aircraft fuel load but I've run out of other options so I hope it's OK with everyone.
I would like to have any other ideas or options to try. Any input guys?
Aircraft performance:
The P-38 was supposed to have great acceleration capability and climb performance recovery from slower speeds. The model has neither. it loses speed rapidly in turns and is slow to regain the speed lost when leveling out. In a dogfight, this is a problem.
The engine power is correct so I think the problem is the thrust generated by the props. It's much too low. I've adjusted the prop MOI several times as well as the prop effectiveness value and have gained some performance but not nearly enough. The AI don't have the same problem, just the player aircraft. The only solution I have found is to increase the thrust scalar in the aircraft.cfg file to 1.2 (20% increase). The downside is once again an increase in the aircraft top speed. I can adjust this with an increase in the drag values. This decreases the range, but the additional fuel load / lighter empty weight compensates for that. However, this decreases the AI top speed. In either case the player will run away from the AI at full throttle in formation.
Another factor is the abysmal throttle response. The RPM increase creeps up rather than a fast response as would be true in the real world (As a pilot, I know this to be true). More on that below.
Again, I'm hoping someone has a better idea.
Take off performance:
Due to the slow throttle response it takes a fair amount of time for the second engine to match the power output of the first engine on start up - about 35 seconds for 10% power. The P-38 had no P-factor (prop torque pull effect) on the take off roll. But if the two engines RPM is not matched you really fight to keep it straight down the runway as the left engine is pulling more power.
Further, when simply adding full power from the 10% throttle setting (once engine RPM is matched) to take off means the RPM spool up is so slow that the wing men overrun the player's aircraft. The only method to counteract that I have found is up to the player's take off procedure discipline, which is to apply brakes and run up to 40% power - taking another 45 seconds - before releasing the brakes and adding full throttle.
I have tried adjusting the engine throttle effectivity table in AirEd but it does not seem to change the RPM spool up rate much. It just decreases the RPM separation between 50% to 100% to a small amount. I have only been able to achieve a 5 second reduction in running up to 40% power while still maintaining a decent RPM separation between throttle settings.
Anyone have a way to increase the throttle response to a realistic level?
Well that's it. The P-38 air files become a kluge of workarounds to get something that works in a campaign, especially where there are any long distances involved. :banghead:
I'd appreciate any ideas that might improve any of these issues.
words from an enthusiastic amateur..
CK,
I would seek out a copy of Aircraft Airfile Manager v2.2. It's optimised for FSX, but it's full of useful notes about the various entries in the airfile, & you can even, should you wish, alter values on those airfile curves.
First a caveat. I don't use Airwrench, but base most of my tweaking on Aired & AAMv22. I've often found when converting LR aircraft that a lot of FS9 airfile values simply give you more problems than they solve. Most of what I've learnt is from observation of impacts after adjusting entries in the airfile.
Stray entries in the aircraft.cfg can also cause issues, especially with AI planes, so I like to start with a stripped down, works for CFS2 aircraft.cfg, with all the flight tuning entries set to 1.00. I then work as much as possible with the airfile, because that affects the AI flight behaviour as well as the player aircraft.
I find that some airfile entries are best left at cfs 2 nominal values (record 1204 wing incidence = the AI like it best if it's between 0 & 1 deg), (record 1101, lift entries left at the values used by the stockers).
Particular to the range issue, I've discovered record 508 & 509 can be tweaked to increase range. Lower record 508 & you reduce your fuel burn. The trade off is your speed reduces, so you need better prop tables, or lower drag entries.
The slope of record 509 & it's final value alters your speed, & also influences fuel burn.
I would also look at record 1101 angle at zero lift, as that can impact on induced drag (also the value entered in record 1204 for induced drag). If these are out of whack, you compensate with better prop tables, but then fuel burn increases. They're all interlinked.
Lot's of amend/fly/review cycles, but if I then have the speed, (plus ceiling) & range about right, only then do I look at handling. There are lots of ways to adjust flight handling within the airfile.
Some value elements of record 1101 can be adjusted, you can adjust record 320 control surfaces, the 1500+ record series can help model handling quirks, & don't forget those values in record 1500. They seem to relate to "agility". Another way of improving AI behaviour, used by Tango Romeo & Talon, was to modify the MOI numbers in the airfile, making them more responsive. So, there are different ways to recover combat handling, but only after you've got the major numbers where you want them.
I hope this helps.