Jump to content

FD during ILS approach not accurately showing the requested movements


Gábor
 Share

Recommended Posts

This is the only plane where the FD during ILS cannot really help to stay on the LOC. Somehow here I always go a bit off. I did thousands of ILS approaches during my sim flights, but here it is always a bit off. Somehow the FD is not enough sensitive here...

Please fellow pilots, share your oponion on this to see what is the situation here.

Link to comment
Share on other sites

The ILS radio simulation feels accurate to me, compared to real life.

Being a bit off is not unrealistic, especially in gusty / turbulent conditions - being way off is called an unstable approach, and if every ILS was flown "on rails", airplanes would never need to go around.

This simulation brings radio reception + avionics processing to you so you can see what it really feels like to fly that kind of approach. The more you practice the better you'll get at anticipating what the FD is commanding and you can stay ahead of those little oscillations to stay stable. Good luck! 

  • Like 2
Link to comment
Share on other sites

The flight director is just an aid, you have the raw data there in front of you so there should be no reason to go a bit off on the approach. The FD behaviour in this add-on closely mirrors that of the real world avionics. Perhaps if you share some screenshots I can expand on my explanation a little bit. 

Brgds

  • Like 1
Link to comment
Share on other sites

I know what OP means and I think this is mainly an effect of the challenger having an atrocious "feel" to it at high AOA. It just isn't pleasant to fly at close to vref

13 hours ago, Gábor said:

This is the only plane where the FD during ILS cannot really help to stay on the LOC. Somehow here I always go a bit off. I did thousands of ILS approaches during my sim flights, but here it is always a bit off. Somehow the FD is not enough sensitive here...

Please fellow pilots, share your oponion on this to see what is the situation here.

so my trick is to mostly fly ILS by raw data in the last 4nmi or so, because the FD encourages use of pitch to control vertical speed which is a BAD idea 

pitch for speed, throttle for v/s, and let the FD zip around, it's ok,  don't chase

PS
Try using command bars instead of V flight director
that way you can more or less focus on lateral guidance and let the pitch command slip 

Edited by lordauriel
  • Like 1
Link to comment
Share on other sites

15 minutes ago, Graeme_77 said:

Pitch for speed is not really a good way to fly a jet! Thrust for speed, pitch for glidepath. It's not a Cessna ;-)

For the OP: Blind FD landing with self-induced LOC and GS deviation. https://streamable.com/qq4nq7
 

If I do not watch FD I am nicely land....

What I learned from an A320 Instructor captain, that you have to continously change your view from FD to sight and vice versa. I keep my eye more on FD for sure. Ok, I will use more raw data....

Link to comment
Share on other sites

10 hours ago, Graeme_77 said:

Pitch for speed is not really a good way to fly a jet! Thrust for speed, pitch for glidepath. It's not a Cessna ;-)

For the OP: Blind FD landing with self-induced LOC and GS deviation. https://streamable.com/qq4nq7
 

Hi, 

 

Since first flight with the CL650 i noticed a slightly different behaviour of the FD, i'm slowly getting used, but I still have trouble on a "weird" SID departure because i easily overshoot lateral path. On approach, even ILS or RNAV or whatever, i know where the runway is so i can precisely decide how much input give to the ailerons, but on a departure it's a little bit tricky for me since i feel like the FD is following me instead viceversa. Feels like the FD is a bit lazy and not in the right mood to direct me. 

what i'm referring to is clearly shown on you video in the precise moment when you disengage AP and pull the column, the horizontal line of the FD is rising quit a bit. From 0° up to 5° nose up the FD is following you instead of remaining "on glide path" or at least remaining fixed, then it start to follow your oscillations up and down. That's what i'm referring at, but like i said before, on approach, if not severe imc condition, i can survive, but not on a departure whatching the vertical lines rubbering around my path. 

The SID i'm practising more is the BEG6Q out of LIME RW28 and the FD, expecially the vertical line induce me to fly like an alpin skier. 

This is just what i noticed about the CL650 FD, and i'm not saying it's wrong modeled or what else, i'm just saying that i clearly saw FD's on the sim market that more suit my aviation ignorance 

 

 

Edited by kecm80
  • Like 1
Link to comment
Share on other sites

On 2/9/2022 at 11:17 PM, kecm80 said:

when you disengage AP and pull the column, the horizontal line of the FD is rising quit a bit. From 0° up to 5° nose up the FD is following you instead of remaining "on glide path" or at least remaining fixed, then it start to follow your oscillations up and down.

Edit: if you’re reading this thread for the first time, you should skip my incorrect assumption of what was going on and read below for some great in-depth analysis.


The reason for this is because when the nose is pulled up initially, the airplane is on glide path so the FD has nothing to command. It’s essentially saying “whatever you’re doing, keep doing it”.

Until the airplane actually gets off glide path, all the FD pitch command is going to do is “stay the course” which could mean adjusting the airplane’s pitch several degrees, for example, during a flap configuration change. 

Let’s say you’re on glide path and you add flaps and lower the nose to maintain descent rate. You wouldn’t want the FD to stay at the original pitch command, you’d want the FD to follow the nose. 
 

Edited by rjb4000
  • Like 2
Link to comment
Share on other sites

3 hours ago, kecm80 said:

Hi, 

 

Since first flight with the CL650 i noticed a slightly different behaviour of the FD, i'm slowly getting used, but I still have trouble on a "weird" SID departure because i easily overshoot lateral path. On approach, even ILS or RNAV or whatever, i know where the runway is so i can precisely decide how much input give to the ailerons, but on a departure it's a little bit tricky for me since i feel like the FD is following me instead viceversa. Feels like the FD is a bit lazy and not in the right mood to direct me. 

what i'm referring to is clearly shown on you video in the precise moment when you disengage AP and pull the column, the horizontal line of the FD is rising quit a bit. From 0° up to 5° nose up the FD is following you instead of remaining "on glide path" or at least remaining fixed, then it start to follow your oscillations up and down. That's what i'm referring at, but like i said before, on approach, if not severe imc condition, i can survive, but not on a departure whatching the vertical lines rubbering around my path. 

The SID i'm practising more is the BEG6Q out of LIME RW28 and the FD, expecially the vertical line induce me to fly like an alpin skier. 

This is just what i noticed about the CL650 FD, and i'm not saying it's wrong modeled or what else, i'm just saying that i clearly saw FD's on the sim market that more suit my aviation ignorance 

 

 

Yeah, this is a good explanation. FD following us... This is what I meant. Thanks.

Link to comment
Share on other sites

13 hours ago, rjb4000 said:

The reason for this is because when the nose is pulled up initially, the airplane is on glide path so the FD has nothing to command. It’s essentially saying “whatever you’re doing, keep doing it”.

Until the airplane actually gets off glide path, all the FD pitch command is going to do is “stay the course” which could mean adjusting the airplane’s pitch several degrees, for example, during a flap configuration change. 

Let’s say you’re on glide path and you add flaps and lower the nose to maintain descent rate. You wouldn’t want the FD to stay at the original pitch command, you’d want the FD to follow the nose. 
 

that's a good explanation, thanks! 

Link to comment
Share on other sites

On 2/9/2022 at 11:39 AM, Graeme_77 said:

Pitch for speed is not really a good way to fly a jet! Thrust for speed, pitch for glidepath. It's not a Cessna ;-)

For the OP: Blind FD landing with self-induced LOC and GS deviation. https://streamable.com/qq4nq7
 

 

On 2/9/2022 at 10:17 PM, kecm80 said:

Hi, 

 

Since first flight with the CL650 i noticed a slightly different behaviour of the FD, i'm slowly getting used, but I still have trouble on a "weird" SID departure because i easily overshoot lateral path. On approach, even ILS or RNAV or whatever, i know where the runway is so i can precisely decide how much input give to the ailerons, but on a departure it's a little bit tricky for me since i feel like the FD is following me instead viceversa. Feels like the FD is a bit lazy and not in the right mood to direct me. 

what i'm referring to is clearly shown on you video in the precise moment when you disengage AP and pull the column, the horizontal line of the FD is rising quit a bit. From 0° up to 5° nose up the FD is following you instead of remaining "on glide path" or at least remaining fixed, then it start to follow your oscillations up and down. That's what i'm referring at, but like i said before, on approach, if not severe imc condition, i can survive, but not on a departure whatching the vertical lines rubbering around my path. 

The SID i'm practising more is the BEG6Q out of LIME RW28 and the FD, expecially the vertical line induce me to fly like an alpin skier. 

This is just what i noticed about the CL650 FD, and i'm not saying it's wrong modeled or what else, i'm just saying that i clearly saw FD's on the sim market that more suit my aviation ignorance 

 

 

So, I'll chime on this topic.  If you watch the FD's pitch bar behavior beginning 0:09 in the video, you will see that as the pitch attitude increases from 0 (at 0:09) to 5 NU (0:12) you see that the FD pitch bar follows the pitch attitude increase.  At 0:09, both the FD pitch command bar and the aircraft symbol pitch attitude is 0 NU.  As pitch attitude was increased from 0 NU to 2.5 NU, the FD pitch bar followed it also to 2.5 NU.  At about 3 NU in aircraft pitch attitude, the FD pitch bar stopped moving and following the pitch attitude increase of the aircraft symbol.  This is not the correct behavior of the flight director system.  

The FD command bars pitch and roll command are generated by the flight director compute relative to attitude changes required to satisfy lateral and vertical references, in this case displacement of the localizer and glideslope needles from the centered positions.  The FD does care what the actual attitude of the aircraft is at any given time.  in fact, you can change attitude of the aircraft in pitch and roll instantaneously away from the FD bars, those bars would not move.  You can literally pitch up and down rapidly through the FD bars, and as long as you did not displace the localizer or glideslope needles from their centered position, the FD command bars will not move.  

This movement of the FD bars with the changing aircraft pitch and roll attitude is a common mistake that I see in flight simulators (to that mean PC type, not the real Level D's). It makes using the FD difficult for PC pilots because they are not used to also watching the raw data, and then get into the habit of chasing the FD.  The video is a very good illustration of this behavior. 

In contrast, watch the FD roll bar commands for the same period from 0:09 to 0:12.  The minute the roll begins, the FD roll bar begins moving back to the right.  It does not follow the roll of the aircraft and await a localizer displacement before moving.  In roll, the FD roll bar is behaving correctly.  if the FD pitch bar were behaving the same way, when the increase in pitch attitude began at 0:09, the FD pitch bar would not have moved, but remained in the same pitch attitude calculated to satisfy the current displacement of the glideslope needle from the centered, on glidepath position. 

Just to provide my background.  34 years flying business jets. 3 1/2 years teaching in the Learjet simulators. 17 years in Collins Proline series, and 13 years in the Proline 21.  I have done these little tests in the Level C and D simulators, and in the real airplane.  The FD pitch command bar and roll bars, or the "V" bar does not change with pilot-initiated pitch and roll changes.  They only change, i.e., move in response to computed commands to satisfy the reference target. 

On another note, I saw above the comment about flying pitch for speed and power for glide path.  It is always for coordinate change of both.  However, as a technique, in the Challenger 300/350 with its big, sailplane type wing, I have found it easier to trim for speed, i.e., VREF, and then make small thrust adjustments to follow the glideslope/glidepath.  I don't use this technique as much on the Lear 75, which is more straight wing jet.  But sometimes, you have to try other tricks. 

I believe that if you tweak the flight director pitch command bar programming so that it does not follow pitch attitude changes, you will go far in addressing these concerns. 

Just one pilot's view.

Thanks,

Rich

Edited by richjb
  • Like 2
Link to comment
Share on other sites

Hi Rich,

Thank you for the detailed and well presented explanation of your concerns with the Flight Director. When you mentioned your background I thought there must be a typo for your Proline 21 experience but then I did some of my own mental math from my flying and realized how quickly time has passed. Your concerns are noted and have been passed to Totoritko for review and evaluate for revision. I didn't see anything specific in the PL21 manuals that talked specifically about FD behavior logic. Do you have a specific reference for the kind of behavior you're talking about? This implementation was a clean room ground up build of the PL21 which surprised the heck out of me when I first learned that fact. It does mean there are cases where unless someone with your experience who has done these kinds of nuanced test speaks up issues may go unnoticed due to a lack of highly informed qualitative information which otherwise wouldn't be actionable.

Thanks again!

 

Link to comment
Share on other sites

1 hour ago, CaptCrash said:

Hi Rich,

Thank you for the detailed and well presented explanation of your concerns with the Flight Director. When you mentioned your background I thought there must be a typo for your Proline 21 experience but then I did some of my own mental math from my flying and realized how quickly time has passed. Your concerns are noted and have been passed to Totoritko for review and evaluate for revision. I didn't see anything specific in the PL21 manuals that talked specifically about FD behavior logic. Do you have a specific reference for the kind of behavior you're talking about? This implementation was a clean room ground up build of the PL21 which surprised the heck out of me when I first learned that fact. It does mean there are cases where unless someone with your experience who has done these kinds of nuanced test speaks up issues may go unnoticed due to a lack of highly informed qualitative information which otherwise wouldn't be actionable.

Thanks again!

 

Let me first check my math.  Started flying the Collins FMS and Proline 4 avionics in the Falcon 2000 in 2004.  We flew that aircraft for five years at which time we upgraded to the Challenger 300 with the Proline 21.  We upgraded to the ProLine 21 Enhanced on our two CL300s in 2016, and in 2018 we added a new CL350 to the fleet that had the Proline 21 Enhanced with IRU.  Doesn't look the same as the CL650, but functionally it is all very close.

Let me say as well, and as I have said on other post, you folks have nailed the Proline 21 and the complexities of a business jet.  The Hot Start CL650 is unprecedented!   It's hands above anything else in the flight sim market.  I don't care about MSFS anymore.  I haven't touched P3D since Christmas.  

I can't say that I have specific reference documentation for what I described as that would be in the engineering manuals that are not typically available to operators or pilots.  That said, I will see what I can find out for you on the subject. 

What I described is typical of all the flight director systems that have flown, be it the old mechanical FD 108/FIS 84 & FD 109/FIS 85, the Honeywell Primus (Lear 45), Collins, or the Garmin G5000 I'm flying the Lear 75.  Speaking of Lear 75, I am in school next week.  I can see if I can shoot a video with similar profile off path to what was shown on the sample video above.  

Let me continue looking at this for you.

Thanks!

Rich 

Edited by richjb
  • Like 1
Link to comment
Share on other sites

41 minutes ago, richjb said:

Let me first check my math.  Started flying the Collins FMS and Proline 4 avionics in the Falcon 2000 in 2004.  We flew that aircraft for five years at which time we upgraded to the Challenger 300 with the Proline 21.  We upgraded to the ProLine 21 Enhanced on our two CL300s in 2016, and in 2018 we added a new CL350 to the fleet that had the Proline 21 Enhanced with IRU.  Doesn't look the same as the CL650, but functionally it is all very close.

Let me say as well, and as I have said on other post, you folks have nailed the Proline 21 and the complexities of a business jet.  The Hot Start CL650 is unprecedented!   It's hands above anything else in the flight sim market.  I don't care about MSFS anymore.  I haven't touched P3D since Christmas.  

I can't say that I have specific reference documentation for what I described as that would be in the engineering manuals that are not typically available to operators or pilots.  That said, I will see what I can find out for you on the subject. 

What I described is typical of all the flight director systems that have flown, be it the old mechanical FD 108/FIS 84 & FD 109/FIS 85, the Honeywell Primus (Lear 45), Collins, or the Garmin G5000 I'm flying the Lear 75.  Speaking of Lear 75, I am in school next week.  I can see if I can shoot a video with similar profile off path to what was shown on the sample video above.  

Let me continue looking at this for you.

Thanks!

Rich 

Hey Rich,

I was on the GIV for a couple years before covid hit then jumped to start a masters in education, would have loved to fly a Falcon, I've heard they are great planes to fly. Thanks for the kind words. We took an initial look and found some areas to do further investigations in some of the PID controllers. If you have some spare time a video of the specific case would be very useful from the sim. We will continue investigate refining the pitch FD behavior, thanks again for the feedback. Best of luck at recurrent.

Edited by CaptCrash
Link to comment
Share on other sites

35 minutes ago, CaptCrash said:

Hey Rich,

I was on the GIV for a couple years before covid hit then jumped to start a masters in education, would have loved to fly a Falcon, I've heard they are great planes to fly. Thanks for the kind words. We took an initial look and found some areas to do further investigations in some of the PID controllers. If you have some spare time a video of the specific case would be very useful from the sim. We will continue investigate refining the pitch FD behavior, thanks again for the feedback. Best of luck at recurrent.

Thanks CaptCrash,

I'll see what I can do!

The Falcon is about the sweetest business jet to fly, alongside the Sabreliner.  They both had no flying vices.  The Falcon's systems were another story.  They were a nightmare.  Three or four switch changes just to move fuel from wing tank to the next.  Only airplane I've ever flown where after a V1 cut and continued takeoff, if raised the flaps and forgot to tie the busses, and the left engine failed, you would lose the left ADC and be without an airspeed indicator to fly V2.  Left me shaking my head! 

Never had a chance to fly the Gulfstreams but have friends who do.  The consistently use the phrase "bulletproof" to describe that airplane! 

Link to comment
Share on other sites

7 minutes ago, richjb said:

Thanks CaptCrash,

I'll see what I can do!

The Falcon is about the sweetest business jet to fly, alongside the Sabreliner.  They both had no flying vices.  The Falcon's systems were another story.  They were a nightmare.  Three or four switch changes just to move fuel from wing tank to the next.  Only airplane I've ever flown where after a V1 cut and continued takeoff, if raised the flaps and forgot to tie the busses, and the left engine failed, you would lose the left ADC and be without an airspeed indicator to fly V2.  Left me shaking my head! 

Never had a chance to fly the Gulfstreams but have friends who do.  The consistently use the phrase "bulletproof" to describe that airplane! 

There isn't much that can ground a Gulfstream...for better or worse lol. The GV and 550 are probably the best of the new generation, once GA built that bigger wing and reworked some of the systems it really became a pleasure to fly from what I hear, some of the IVs were getting very long in the tooth. I've heard Dassault have a way with systems integration which can be...intriguing.

Link to comment
Share on other sites

2 hours ago, CaptCrash said:

There isn't much that can ground a Gulfstream...for better or worse lol. The GV and 550 are probably the best of the new generation, once GA built that bigger wing and reworked some of the systems it really became a pleasure to fly from what I hear, some of the IVs were getting very long in the tooth. I've heard Dassault have a way with systems integration which can be...intriguing.

"intriguing"???  You're being polite! :-) 

Link to comment
Share on other sites

Hi CaptCrash, et al,

This morning, I flew v1.3.1 (need to download the new version) and looked closely at the flight director behavior.  When basic pitch modes like ALT, VALT, FLC, and VPATH are active, the flight behavior appears to be correct in that the FD pitch bar does not move and follow aircraft pitch changes.  In other words, you can fly the airplane symbol up and down through the FD pitch bar, and bar does not initially move and also follow the FD pitch bar as is shown in the video when an APPR LOC and GS are active. 

I flew the ILS 1R at KICT and the FD pitch bar exhibited the same behavior.  As shown in the video, if an aircraft pitch change is made, the FD pitch bar initially follows that pitch change, e.g., pitch the nose up a few degrees, and the FD pitch bar follows it for a few degrees until the GS begins to deflect.  I spoke to a friend who worked at Collins for a number 15 years and trained us on the Collins equipment about this and show him the video.  He said that was not how it works.  The FD pitch bar should remain steady and initially not move with the pitch change, and in fact, in anticipation of the GS needle moving downward it should deflect slightly downward.  What the HS CL650 is doing with the respect to the ILS GS and the FD pitch command is not correct. 

Next, I flew the RNAV (GPS) Y 1R at KICT.  The approach was flown with LPV Level of Service (LOS).  Initially, I used VNAV to descend through the T-Legs from BACAY.  When tracking the VNAV in VPATH vertical mode, the FD pitch bar behaved as expected.  In other words, you could change the aircraft's pitch attitude through the FD pitch bar without that command bar also moving.  Naturally, as deviated off the VNAV path, the FD pitch bar would command a return to the path.  Approaching the FAF (CUTIK), I selected APP Mode and the vertical guidance changed to VGP.  In VGP mode, the FD pitch bar behaved like it does in the ILS GS mode, and as shown in the video.  The FD pitch bar moves with aircraft pitch changes.  As noted, this makes it very hard to track the FD command bars and track the VGP path. 

Next, I came back and again flew the RNAV (GPS) Y 1R at KICT.  This time, I downgraded the approach to LNAV/VNAV LOS by selected the ARRIVAL DATA page, selecting the APPR MODE to "RNP", which selected the GP MODE To "BARO".  (see screen shots attached).  So instead of flying LPV SBAS generated angular vertical guidance, the FMS would be using Baro-VNAV, linear guidance just like in the en route phase, but with smaller scaling.  On this approach with LNV and VPATH as the FD modes, the FD pitch bar behaved as expected.  The FD pitch command bar did not move and follow with aircraft pitch changes, but rather remained steady until there was a vertical path deviation at which it showed a command correction to return to the path. 

It appears that the flight director issue is in both the ILS GS and LPV VGP vertical modes, that the FD pitch bar will initially follow a pitch change before it computes that maybe it shouldn't have, and then as the GS or VGP moves from the centered, "on glideslope/glidepath" position, the FD pitch bar initiates a command correction to command a pitch change to the return the aircraft to the glideslope/glidepath. 

In both cases, as the aircraft pitch changes, the in GS or VGP, the FD pitch bar should not change and initially follow the aircraft pitch change.  It should not move until the GS or GP needle begins to move off path, or according to my source from Collins, it should make a slight anticipatory command opposite of the pitch change to return the aircraft to GS/GP.  At the very least, the FD pitch command bar should mimic the behavior in VPATH or the other basic FD pitch modes like ALT.  These easily demonstrated in level flight with ALT engaged, in a descent following a vertical path and in VPATH mode, or on an RNAV approach using RNP and BARO set in the ARRIVAL DATA page (ps...kudos for getting this page right!).  

I hope that this will help with the troubleshooting. 

Thanks,

Rich 

 

Arrival Data Page.PNG

ARR Page.PNG

Edited by richjb
  • Like 3
Link to comment
Share on other sites

  • 2 weeks later...

Soooo....with an empty leg, good clear VFR weather, and an understanding co-pilot, I did a little flight direct experimentation trying to replicate the video in Hotstart in the Challenger 350....not 650, I don't fly that model.  Same Collins Proline Enhanced system.   

I was to say the least surprised with APPR LOC and GS active, when the aircraft was pitched up smoothly but rather swiftly, that the flight pitch bar initially followed the pitch change for about 2.5 degrees (one mark on the pitch ladder) much like what you saw in the Hotstart CL650 video.  I was able execute a slight pitch up and then re-stabilize several times.  In case anyone from FAA is reading, I was stable speed throughout the approach with only a slight GS deviation and ceased the activity to stabilize at 500' all the while in clear VFR conditions to meet stable approach criteria.  

I have good friend who worked in pilot training and customer support in Collins for many years. He has trained me on the Collins equipped aircraft that I have flown in the last 20 years.  He agreed with me that the Hotstart CL650 was not how he expected the flight director pitch bar to behave.  I have shared my real observations with him but have not heard back yet.  It also does not agree with our understanding of how flight directors work, especially with the old mechanical FD 108 and FD 109. 

I may ask around further. 

For now. it appears that Hotstart's CL650 flight director is behaving as it would in the aircraft.  I'll stand corrected on my comments above. 

Kudos once again to the folks at Hotstart!

Rich

 

Edited by richjb
  • Like 2
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...