Sign in to follow this  
ark

L35A Pitchs Up at 29000ft

Recommended Posts

A friend of mine has the FSX L35 ver 2.6 on a Win8 system. When he climbs through 29000ft using the AP SPD mode, the L35 suddenly pitches up to max VS on the VS dial and the airspeed falls off towards a stall. He tried deleting and reinstalling the L35, and it happens both in the DVD and Steam sims.  Anyone else experience anything like this or have any suggestions what the problem might be?

Thx,

Al

Share this post


Link to post
Help AVSIM continue to serve you!
Please donate today!

I just finished a flight and the same thing happened to me using FSX.

 

Gerry

Share this post


Link to post

I just finished a flight and the same thing happened to me using FSX.

 

Gerry

Gerry,

What Op system are you using, and does this happen without any 'fix' for the SPD light bug in ver 2.6?

Thx.

Al

Share this post


Link to post

I found the the Mach hold to be a bit aggressive when changing attitude in order to maintain the mach number.

Up to now, I have only seen excessive pitch down in a Mach hold climb which temporarily resulted in a descent, but not the other way round.

 

Maybe the auto pilot needs some more fixing so that it won't descend during a climb or vice versa when trying to chase the assigned IAS or mach.

Because I found the mach hold to be generally a bit unstable with regards to attitude, I usually do a V/S climb after FL290.

 

Depending on how you did your SPD light fix (I assume you did it the obvious way), it should not have anything to do with that at all.

Share this post


Link to post

Hi Dave,

My friend experiences the pitch up problem every time he climbs in SPD mode, both in FSX DVD and Steam. This happens with and without any fix for the IAS/MACH - VS mode light problem.

I haven't seen this pitch up problem yet despite trying climbs in SPD mode at different speeds and power settings, thus different VS rates. I do sometimes see the FD 'jump' up a tiny bit at the 29000ft boundary, but it immediately returns to where it was. I run Win7 and my friend Win 8 so was wondering if that might be a factor although I don't see why that would be.

 

Al

Share this post


Link to post

Hi Al,

 

I'm using Windows 7.  I did do the "fix" for 2.6, but I believe this sudden pitch up happened once before the fix.  I usually use spd hold during climb.  I haven't tried v/s through 29,000 but will do so next flight.

 

Regards,

 

Gerry

Share this post


Link to post

I had overall smooth transitions from SPD to MACH during beta, however the aggressive pitchup occured one or two times (out of 20-30 flights). I was unable to determine the cause of them, so it's not related to the OS or installation, it must be something within the sim.

Share this post


Link to post

From what I can discern by looking at the xml code is that when SPD mode is engaged, the current VS, IAS and Mach# are recorded -- these are the AP reference values, When crossing 29000ft the current applicable speed value, depending if you are climbing or descending, is compared to the reference IAS or Mach#  and an offset VS value is added or subtracted to/from the reference VS value. So for example, if the airspeed when crossing 29000ft is high, increasing the required VS will cause the aircraft to pitch up a bit (and so the FD to move up a bit) until the airspeed (and eventually the FD)  return towards the reference value. With constant power, pitch is used to control airspeed. It is this feedback mechanism that seems to be failing for those who are experiencing extreme pitch/FD changes.

 

Al

Share this post


Link to post

If your theory is correct, why do you suppose that the feedback mechanism is failing for some and not others as we are all using the same code?

 

Gerry

Share this post


Link to post

If your theory is correct, why do you suppose that the feedback mechanism is failing for some and not others as we are all using the same code?

 

Gerry

Wish the heck I knew the answer to that! If may have something to so with the operating system. So far the only person reporting that the pitch up happens frequently (or every time) is running Win8. It others are using Win8 and not having this problem, it would be good to know.  It could be some kind of timing issue, although FSX XML files typically are executed 18 times/sec.

 

From the AP code I looked at (and I am very much an amateur at this), at the 29000ft boundary for every 1 knot of airspeed, or 0.001 of Mach#, that the current airspeed differs from the AP reference values set when SPD was activated, 400ft/sec of VS is added or subtracted from the reference VS to force a return to the reference airspeed. If a much larger VS delta was erroneously being added or subtracted, maybe that would account for the observed behavior. But the fact that the pitch up (haven't heard of a pitch down issue) only happens occasionally for some is what makes me think of timing.

 

But this is all very much speculation.

 

Al

Share this post


Link to post

I had overall smooth transitions from SPD to MACH during beta, however the aggressive pitchup occured one or two times (out of 20-30 flights). I was unable to determine the cause of them, so it's not related to the OS or installation, it must be something within the sim.

Matthias,

Check my logic (I use the term loosely) here.

 

If you set SPD mode at low altitude, say at 1500ft, with an IAS of 200kts, and then climb through 29000ft, at the 29000ft boundary your true airspeed has changed quite a bit although your IAS is still 200kts. Likewise, your Mach# has changed quite a bit from the time you set SPD hold. So when the AP compares the original Mach# to the Mach# at 29000ft, there will be a significant difference and so the a/c will pitch up to make the necessary correction. However, if the AP were to update the Mach# value as you approached 29000ft, it would not see much difference at 29000ft and so less correction would be required which might make the 29000ft transition smoother in general.  Same idea holds regarding updating IAS approaching 29000ft when descending although the altitude change would usually be less between when you set SPD hold and reach 29000ft, so less change in IAS.

 

So if you haven't tried it, to reproduce the aggressive behavior you saw just a few times you might try setting the SPD hold at low altitude and see what happens. However, all this does not explain why some see the aggressive behavior and some just see the FD pop up a bit as a speed correction takes place at 29000ft.

 

Al

Share this post


Link to post

I will do some testing tomorrow and see what i can do. From previous beta testing the pitch would just adjust a small amount to hold the mach speed as it should. So i was thinking it could be weather as i test with no wind. But the code i wrote is designed to hold the recorded speed at the time it is turned on or passing through 29K. 

 

If i can't fix this issue the only other option is to go back to holding IAS only. 

Share this post


Link to post

From previous beta testing the pitch would just adjust a small amount .........

 

Yes, this is what I usually see when climbing through 29000ft (the FD pops up a bit and then settles back somewhat), but as I understand it, what the a/c is doing here is adjusting the speed to match what the Mach# was when SPD hold was selected which could have been, for example, down low at 2000ft, etc. The a/c TAS has been increasing throughout the climb as IAS is held fixed, so it would seem to maintain a smooth transition when climbing through 29000 ft you would want to hold the current Mach#, not return to what the Mach# was at 2000ft.  Likewise, on descent, you would want to hold the current IAS as you pass through 29000ft for the rest of the descent. Or so it seems to me. I suppose the question is what is the intent when climbing through 29000ft -- to return to the initial 2000ft Mach# or to maintain the current Mach# for the rest of the climb? Perhaps Gary Stewart could provide some real world insight.

 

The other issue of why some, but not others, often see a 'violent' pitch up and major drop in airspeed when climbing through 29000ft is very strange. When this happens it seems as it the AP feedback system (PID?) has failed. From discussions with my friend, the bigger the adjustment required when crossing 29000ft the more likely the violent pitch up, at least in his case.

 

Thanks very much for looking into all this -- much appreciated.  And thanks for the L35A, it has become the a/c I fly 95% of the time.

 

Al

Share this post


Link to post

After testing passing through 29K 10 times with wind and no wind the pitch was smooth and the system works as it should. I never once had any issues. So for now all we can do is wait to see why just a few users have an issue. So far most users, me and the beta team are no having issues. 

 

I think if your computer is not updating fast enough then this could happen because it can't adjust for the new recorded pitch when changing to mach.

 

 

The SPD mode under 29K records the IAS at the time you engage it.

The SPD mode above 29K records the MACH at the time you engage it.

 

When your passing through 29K in a climb it's holding the IAS and then disengages the SPD mode for 2 seconds then re engages the mode. This records the new MACH and it now continues to hold the MACH.

 

If it pitches way too much then it's using the MACH speed from the time you engaged the IAS and did not update the new MACH speed. So again this would be a personal sim issue. 

Share this post


Link to post

Very interesting, I understand better now how SPD Hold works.

Thanks for looking into the issue and the feedback.

Al

Share this post


Link to post

I had overall smooth transitions from SPD to MACH during beta, however the aggressive pitchup occured one or two times (out of 20-30 flights). I was unable to determine the cause of them, so it's not related to the OS or installation, it must be something within the sim.

 

Matthias,

 

I had the aggressive pitch up happen to me yesterday for only about the 4th time since flying with ver 2.6 (which I fly regularly). I use FSX on a 4GHz system that runs smoothly most of the time. Here's one possibility for the randomness at which these pitch ups seem to happen.

 

In looking at the code it seems 5 ft altitude ranges are used to turn SPD hold on and off at 29000ft. If  the a/c is in the altitude range 29001 to 29005, SPD mode is turned off. If the a/c is in the range 29006 to 29010, OR in the range 28995 to 29000, SPD mode is turned on. The 29006-29010 range is for climbs, the 28995-29000 range is for descents. As explained above,  it is when SPD mode is turned on after being off that the reference IAS and Mach# are ‘updated’.

 

So, for example, assume we turn SPD mode on at say 1500ft and start climbing to 32000. The IAS at 1500 is now the reference speed value. When we reach the 29001-29005 range SPD mode is turned off, and then turned on again when we hit the 29006 to 29010 range at which time the reference Mach# is updated for the rest of the climb. On a descent, assume the SPD mode is turned on at 32000 and stays on until we hit the 29005 -29001 range at which time SPD mode is turned off. It then is turned on again in the 29000-28955 range and the reference IAS for the rest of the descent is captured. Very clever, but I wonder if these 5 ft altitude ranges may account for the random behavior some see at 29000ft. 

 

If the a/c VS is say 2000ft/min at 29000ft, it is only in one of these 5ft altitude ranges for 5ft x 60sec/2000ft = 0.15 seconds or 150 millisecs. If the VS is 3000ft/min, then the a/c is only in a 5 ft range for 0.1 sec or 100millisecs. As I understand it, the XML code is typically run 18 times/sec or every 55.5 milliseconds. So it seems that if the sim happens to be 'busy' when the a/c is in one of the 5ft ranges, there is a chance the SPD mode won’t be turned off in the 29001 to 29005 range, or maybe not turned on again in one of the ‘on’ ranges, and thus the Mach# or IAS, depending on which way you are going, won’t get updated.

 

I think this could certainly show up as a random problem, and the randomness of it makes it very difficult to test. One experimental idea to try to reduce how often these pitch ups happen would be to increase the size of the SPD hold 'on/ off' windows to say 10ft, although I don't know if there would be unintentional consequences to that. Since for some reason my friend seems to experience the pitch ups quite often (to the extent he prefers not to use ver 2.6),  I will ask him if he will try the larger SPD hold on/off windows to see if it helps or not.

 

Best,

Al

Share this post


Link to post

Yay, that sounds reasonable! I am running at 4.3GHz and with the Lear it's usually very smooth. On slower PC's that problem may occur more often.. I will try to make some tests tomorrow with a light airplane and different rates of climb. I think you have a really good point here!

Share this post


Link to post

 

I tried a few climbs and descents and it seems to work great so far.  And you can easily see the SPD Hold light go off,  and then on again as the a/c goes through the 100ft windows which is a nice way to confirm that nothing has been 'missed' by a busy sim. If the random pitch ups still happen but the lights go off and on as expected, then we know something else is going on.

 

Thanks very much!

 

Al

Share this post


Link to post

What rates did you fly? I just tried with deliberate high power setting resulting in around 3000fpm and got the pitch up.

Share this post


Link to post

What rates did you fly? I just tried with deliberate high power setting resulting in around 3000fpm and got the pitch up.

Well that's interesting. I just tried it again 3 times with at VS in the 3500ft/min range with no problems. I could see the IAS/Mach lights cycle on and off as expected.

I assume you saw the lights cycle as expected as well. If so, I'm thinking there is something in the feedback system that is not getting initialized or reset correctly after the Mach# is updated at 29000ft. Maybe another timing issue.

 

Al

Share this post


Link to post

It switched directly from IAS to MACH without obviously being off for a noticable time. I will do further testing when I find the time.

Share this post


Link to post

As I recall, the target Mach# (or IAS) is maintained by adjusting the VS, and when these aggressive pitch ups happen, it is as if the a/c is trying to maintain a new and very high VS. So I wonder if 'sometimes' the updated VS is not getting set correctly for some reason.  I will see if I can use any of the logging capabilities in FSUIPC to see what the target VS value is at any time.


It switched directly from IAS to MACH without obviously being off for a noticable time. I will do further testing when I find the time.

I think you should see a definite 'lights off' time.  Did you restart the sim or reload the a/c after loading the new Cab file into the panel.MAIN and/or panel.GNS folder?

Al

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this