reported degraded dynamic stability when flying between 300 and 400 KCAS in AAR gains

[DummyCatz]

Hi,

Good job with the new FM/FLCS update. However I'm experiencing a greatly degraded dynamic stability when flying between 300 and 400 KCAS in AAR gains (AAR door open). This may apply to gear down too.

Testing method: Open the AAR door and stay within 300 to 400 KCAS, apply a step input to make a pitch disruption. Observe the well developed phugoid.

Expected behavior: The dynamic stability should at least be inline with that of airspeed < 300 KCAS. This degradation of dynamic stability is not recorded in the Flight Characteristics section in -1.

F16 FLCS TO-LDG gains phugoid.trk

Edited June 9, 2022 by DummyCatz

[BIGNEWY]

posts merged we will investigate. 

thanks

[DummyCatz]

8 minutes ago, BIGNEWY said:

posts merged we will investigate. 

thanks

Hi, I don't think the inertia coupling bug in the roll axis, and the dynamic stability bug in the pitch axis, are the same thing.

[BIGNEWY]

Hi 

I have split the topic again and we will investigate, but do you have any evidence this is not correct as it is or just a feeling? 

thanks

[DummyCatz]

8 minutes ago, BIGNEWY said:

Hi 

I have split the topic again and we will investigate, but do you have any evidence this is not correct as it is or just a feeling? 

thanks

Hi, it's based on control quality guidance and a bit of common sense in the flight control system domain:

1. With reference to MIL-STD 1797A, the F-16 is rated Level 1 in control quality, and a well developed phugoid to such extent is a very unwanted characteristics, that could immediately lower the control quality score rated by pilots.

2. This unwanted characteristics, if exist IRL, should be recorded in the Flight Characteristics section of T.O. -1.

3. The AAR gain is designed to be flown from 0 to 400 knots airspeed, and in this region of airspeed, the pitch-rate response should be consistent.

4. This behavior does not exist in the previous version of DCS F-16, which exhibits a very well dampened pitch-rate response. Considering this is not on the update/bug fix list, we need explanation from ED on which DCS version exhibits the correct behavior.

[Stackhouse]

1 hour ago, BIGNEWY said:

Hi 

I have split the topic again and we will investigate, but do you have any evidence this is not correct as it is or just a feeling? 

thanks

We are seeing this even in BFM servers with no damage. Seems almost like an abrupt FLCS failure or a behavior when in the previous update would only occur under damage. Video linked

 

Edited June 9, 2022 by Stackhouse

[BIGNEWY]

11 minutes ago, Stackhouse said:

We are seeing this even in BFM servers with no damage. Seems almost like an abrupt FLCS failure or a behavior when in the previous update would only occur under damage. Video linked

 

 

please include the track replay if possible

thanks

[BIGNEWY]

Reported I will ask the team to take a look. 

thanks

[StandingCow]

Has anyone tried A/A refueling since the FM changes?  A buddy and I did (or in my case attempt to) A/A refueling and it felt like there was a dead zone which made minor pitch corrections really difficult to accomplish, you'd suddenly be pitching up or down way further than expected?

[Red Rooster]

Hi there, 5 of us from our sqn also experienced this last night - the plane was very twitchy most of the flight but refuelling was most challenging.  Is this behaviour correct?

[Tom Kazansky]

I verry much hope there are no dead-zones implemented. Those would be poison for precise flying while AAR.

[WHOGX5]

2 hours ago, Red Rooster said:

Hi there, 5 of us from our sqn also experienced this last night - the plane was very twitchy most of the flight but refuelling was most challenging.  Is this behaviour correct?

I found the roll rate a bit hard to handle at first as I'm used to making bigger corrections, but now that my muscle memory has adapted to it, it's very controllable even without curves or saturation. The DCS F-16 has been very sluggish ever since release so it's only natural that it'll take some flight hours to adapt to the new FM update. The F-16 is actually twitchy and snappy now as it should be and once you get used to it you'll be very grateful for this update.

[Aquorys]

Just tested it, feels like it always did. The F-16 doesn't like to go slow though, so if your tanker is doing some 250 knots or so, expect it to be rather unstable in pitch. Around 300+ should be good. Rolling typically pitches the Viper up a little bit, you might want to trim down a little bit while refueling.

Gains are different if the refueling door is open (in DCS as well as on the real aircraft). The dead zones may come from the joystick, even quite high-end joysticks for home use don't feel as accurate as the stick or sidestick in real aircraft or the professional simulators used for pilot training (which might also explain why the "real ones" are orders of magnitude more expensive that what you have at home).

[Dragon1-1]

3 minutes ago, Aquorys said:

Gains are different if the refueling door is open (in DCS as well as on the real aircraft). The dead zones may come from the joystick, even quite high-end joysticks for home use don't feel as accurate as the stick or sidestick in real aircraft or the professional simulators used for pilot training (which might also explain why the "real ones" are orders of magnitude more expensive that what you have at home).

Actually, the real F-16 does have a bit of a "dead zone" (not really a zone in that the stick doesn't move), which is a small breakout force that needs to be applied before the aircraft responds. It's tiny, but it's there, a force sensing stick that could be actuated by a fan blowing on it would be no good. OTOH, certain high-end bases like Winwing's SuperLibra can be configured without any dead zone at all. It's quite achievable for a cam-based design with a sufficiently powerful spring. I believe FSSB also has either a very small deadzone (similar to the real Viper's) or none at all, it's a force-sensing design, though with weaker actuation forces than the real thing.

I think the price of the real hardware is more due to the certifications and general ruggedness of it, as well as the fact that control forces in real aircraft tend to be high, so everything needs to be made to not only withstand them, but to never ever fail even after years of being constantly put under stress. The real Viper stick also uses LDVTs as sensors, which are also not cheap, and neither are ruggedized switches they use (hats, in particular, are to the tune of 400$ apiece). Precision isn't the problem, that's easy when you're making something that's designed to sit in an indoor simulator at 1G, and which, if it breaks, doesn't hurt anything but the user's wallet. Pro simulators simply use the same components as real aircraft.

That, and pro equipment often costs an arm and a leg just because manufacturers know companies and especially governments will pay that much, not because of any good reason ("research grade" is even worse, and don't get me started on anything "medical grade"). 

[Aquorys]

1 hour ago, Dragon1-1 said:

turers know companies and especially governments will pay that much, not because of any good reason ("research grade" is even worse, and don't get me started on anything "medical grade"). 

Wait until you see "military grade" 
Although, even then, sometimes there is a good reason for it. E.g., have you ever heard of the RAD750? It's a microprocessor that is used in avionics, e.g. for rockets, satellites and radiation hardened equipment. Those sell for about $200,000 per piece. Functionally it's the same as a PowerPC 750, which is what you find in a PowerMac G3 from ~1998, and you can buy those for about $50 per piece.

[SickSidewinder9]

At first I thought it was very different after the patch, and I had some trouble, but a lot of it may have been in my head as I very quickly resolved it.  (Switching back and forth between F-16 and A-4 can produce some interesting results in my flying)

It did seem like trimming nose high for low speed produced some odd behavior like flyups.  Again, it could have been me.

Edited June 10, 2022 by SickSidewinder9

[Dragon1-1]

32 minutes ago, SickSidewinder9 said:

It did seem like trimming nose high for low speed produced some odd behavior like flyups.  Again, it could have been me.

In pitch, you trim the Viper for G. In other words, you don't. Neutral pitch trim should be exactly 1G, so unless you mess with it, you're effectively always in trim. Viper is the only aircraft I know (though 5th gens and other advanced FBW aircraft are likely the same), that uses roll trim more than pitch, and which does not need any trimming in normal operation.

If you try to trim nose high, you're basically trimming it for positive G, which at low speed will produce a flyup. I think this is more or less what the real aircraft would do. Nose position is controlled with the stick 100% of the time.

[SickSidewinder9]

Interesting

[Aquorys]

1 hour ago, Dragon1-1 said:

Viper is the only aircraft I know (though 5th gens and other advanced FBW aircraft are likely the same), that uses roll trim more than pitch, and which does not need any trimming in normal operation.

I guess you'd want to trim roll if you have released a weapon and have an asymmetric load as a result. I do, cause I'm not a big fan of having to push the stick all the time.

Wrt. other FBW aircraft, some do more auto-trimming than the Viper, e.g. virtually anything that comes from Airbus/EADS - all the modern civilian jets like the A320/330/340/350/380, but also the Eurofighter or military transport aircraft like the A400. Malicious gossip has it that a perfectly normal Viper FBW behaves somewhat similar to a degraded Airbus (one that's flying in so-called "alternate law", cause some systems or sensors aren't working).

Edited June 10, 2022 by Aquorys

[Dragon1-1]

Yeah, I'd expect Eurofighter to have even better FBW, Viper's roll channel is particularly dumb, and it won't trim out a weapons release, so trim is used there quite a bit when you're dropping single LGBs or JDAMs. I don't think I've ever flown an Airbus. Sure I programmed it once or twice in Flight gear, but there wasn't much actual flying past rotation.  These things literally do everything for you, including landing if you let them. That's one reason I prefer vintage jets for that kind of sim, a Boeing 707 or a Tu-154 have none of that and they actually require some effort to fly.

[dannyq8]

How has this post been marked "correct as is" when in this post it is being investigated?

Seems to be the same issue

 

[StandingCow]

2 hours ago, dannyq8 said:

How has this post been marked "correct as is" when in this post it is being investigated?

Seems to be the same issue

 

Yea, it feels like sometimes threads are marked "correct as is" without much thought, not saying that's what is actually done but it feels like it.   That thread you linked could possibly be related to this "issue" if it is actually even an issue,  hard to know.

And before this thread was marked correct as is... was the team asked about it?  Was the f16 tested with the fuel door open after the FM changes?  So many questions...

[StandingCow]

Seems like something is indeed incorrect:
 

Quote

STRIKER — Today at 10:45 PM

Thanks, as USAF flight test engineer that flies online with real F-16 evaluator pilots, etc., I can tell you it's not accurate unfortunately with the FLCS refueling door pitch gains. I hope they fix it soon so we can AR once again.

 

[Sinclair_76]

13 hours ago, Dragon1-1 said:

I believe FSSB also has either a very small deadzone (similar to the real Viper's) or none at all,

 The FSSB does have a break (out) force adjust (BFA) up to 0%-10% of the total maximum force in 4 levels.

 

 

Edited June 11, 2022 by Sinclair_76

[Tom Kazansky]

As a user of the stable version of DCS I just did AAR with the viper and can say there is no deadzone and no lag even/especially on the smallest inputs. I saw very precise and "calm" response. I thought this would be correct as is.

Has this behavior really changed with the latest open beta?