Jump to content

Sign in to follow this  
MrGreen

Always too fast in descent

Recommended Posts

Hi. 

 

I've recently started using the NGX again after a 6 month hiatus and I'm running into problems. 

 

Every time I'm descending the aircraft picks up speed compared to the VNAV DES TGT SPD and I get the "DRAG REQUIRED" message on the FMC, usually when descending through FL250-150.

I've tested not using the speed brakes and I'll hit 290IAS or more, compared to my target of 260IAS.

 

I use PFPX for the flight planning and ASN as weather. I add the winds from the DESCENT column of the briefing to the DES FORECAST page and also ISA DEV and QNH, and they seem to reasonably match whats in the simulator at the time of the descent, still I'm always overspeeding. 

 

Here are two screenshots that might help in identifying the issue (they are quite large so I linked them instead of embedding):

 

http://s2.postimg.org/fsmnx3hk9/DESFORECST.jpg

 

http://s12.postimg.org/f1k6n7p71/DRAGREQ.jpg

 

I have the FSUIPC dynamic friction mod installed but since it shuts off automatically above 30kts, and affects only ground friction, I can't imagine it would be the culprit.  

 

What am I doing wrong here? I realize that you might have to use the spoilers once in a while, especially if ATC is involved but to me this seems like the VNAV is calculating descent paths that are too steep for the aircraft to handle.

 

STAR for this flight was RASMU 3F to Malmö Sturup Airport (ESMS), but since it happens on almost every flight I doubt it has anything to do with this specific STAR. Cost index is "6".

 

 

Thanks in advance.

Share this post


Link to post
Share on other sites

but since it happens on almost every flight

 

Sounds like (and based on the one screenshot - looks like) you're getting on the runway, advancing the throttles to about 40%, hitting TO/GA and leaving your hardware throttles there for the rest of the flight.

 

This is all fine, until you get to the descent phase and the A/T enters HOLD mode. In this mode, the pilots can override the idle path descent by adding some throttle as necessary. Since your hardware doesn't have servos in it, it didn't go back to idle when the A/T commanded idle thrust just prior to descending. If you bump your throttle, or the throttle spikes, the sim "wakes up" and uses the hardware position as long as you're in HOLD mode (most of the descent). Either remember to drop the hardware throttles to idle prior to T/D, or set the A/T OVERRIDE setting to NEVER in the PMDG SETUP> menu.

Share this post


Link to post
Share on other sites

Having done a few real life trips recently in 737-800, 700 and 900 aircraft, they use the speed brakes a lot in descents. I mirror that in the PMDG NGX. I plan my descents earlier than the FMC shows, usually 130-140 miles out. Keep it at no more than 2,000 fpm and then as I get to 10,000 I change the FPM to 1,500 fpm. Works pretty well. Maintains speeds and gets you to STAR required altitudes much better.

Share this post


Link to post
Share on other sites

Its only a 10 kt jump from your des speed not a biggie. Why you not in ECON des speed?

 

Once established in the descent, ask yourself frequently if you are on profile or not. Use all available sources this is better than “hoping” that you are on profile or blindly following the VNAV data.

 

 

If you are not on profile, consider what you can do to regain it:

 

 If low on profile, you can reduce your ROD by using V/S until the correct profile has being re-established. Now re-engage VNAV and continue monitoring the profile.

 

 If you are high on profile above FL100 then increasing IAS (up to a maximum speed of 330 kts) may be necessary to regain the profile. Update you descent speed in the FMC to recalculate your new descent profile. If you are still high then use the speedbrake for additional drag.

 

 If you are high on profile below FL100, you are limited to 250kts and speedbrake should be used to increase your descent rate. If you are still high on profile you should slow down and use the following:

 

 Max Speed 250 Kts = 1500fpm ROD = 250’/nm

 250Kts/Speed Brake = 1800fpm ROD

 220kts/Flap 5/ Speed Brake = 2300fpm ROD

 180kts/Flap 10/Speed Brake = 1500fpm ROD = 500’/nm

 

These give the best height loss per nm. If still high on profile then extended vectors, an orbit outside 10nm or joining a holding pattern may be your only options to avoid a possible High Energy Approach (HEA).

 

Remember, flaps are used to slow down the aircraft and speedbrakes are used to increase drag. We do not use flaps as drag devices, therefore use “Flaps to Slow Down and Speedbrakes to Go Down!”

Share this post


Link to post
Share on other sites

Agree with Vernon.  Why the 260/.76 descent?? I've never heard that from ATC.... you might get a call at some point to line up at 280 but your speeds raise the question: Why?

Share this post


Link to post
Share on other sites

Infact you are on profile its just your speed thats slipping away, 10kts isn't a biggie. You haven't got much of a TW maybe 9kts or more you have a strong crosswind. To be honest what is the worry? Just keep ontop of the speed.

Share this post


Link to post
Share on other sites

Thanks for all the quick answers.

 

 

This is all fine, until you get to the descent phase and the A/T enters HOLD mode. In this mode, the pilots can override the idle path descent by adding some throttle as necessary. Since your hardware doesn't have servos in it, it didn't go back to idle when the A/T commanded idle thrust just prior to descending. If you bump your throttle, or the throttle spikes, the sim "wakes up" and uses the hardware position as long as you're in HOLD mode (most of the descent). Either remember to drop the hardware throttles to idle prior to T/D, or set the A/T OVERRIDE setting to NEVER in the PMDG SETUP> menu.

 

 

I suspected this might be the case before, because one time the throttles were stuck at 47% N1 in the descent. Since then I set my hardware throttles to 0 during cruise. However, I think I've always had the "A/T OVERRIDE = NEVER" since that is the default, afaik?

 

 

Its only a 10 kt jump from your des speed not a biggie. Why you not in ECON des speed?

 

 

I've been able to manage the "overspeeds" without many issues, the point of my question was more to try to identify if something in my descent planning was incorrect or if a sim/outside sim issue was the cause. If you look at my two screenshots you can see that the winds on the ND correspond almost exactly with what I put in the descent forecast for that altitude.

 

The 260IAS descent speed I chose because I've been told that the two 737 carriers in Norway use  260 descent speed as SOP. Also a few of the Swedish airport STAR charts/AIP advise on keeping 260IAS or less after crossover altitude. However I might have misunderstood this, since I know Braathens company procedure were to fly 280 in the descent. Also I should add that I had this "issue" with ECON SPD. 

 

I guess what puzzles me is that if the wind speeds I put into the DES FORECAST page match the "real world" why would the VNAV path be too steep?

 

 

Infact you are on profile its just your speed thats slipping away, 10kts isn't a biggie.

Well, one thing to note is that the speed was increasing at this moment in time and would probably stabilize at 290.  Indeed the aircraft was on profile, but the profile was too steep for the aircraft to keep its speed. If this happened once in a while I could put it down to winds being different, but I've never had it not happened since I picked up FSX again. 

 

 

Maybe I'm overthinking it but I want to know if I'm doing something wrong, so that I can fix it, or whether something is wrong with my simulation. 

Share this post


Link to post
Share on other sites

I've noticed that the A/T does not go to exactly idle at the start of VNAV descend. You can try to hit F1 after entering descend and will see the N1 is actually something like 5% higher than idle.

 

What I do is press F1 at around FL290 and check thrust is reduced to idle. This should work fine. 

Share this post


Link to post
Share on other sites

I have noticed this as well.

 

I have the Saitek Throttle Quadrant and have mapped the reverser-notch (after the bottom end of the throttle range) to F1. On touchdown, after I've put the throttles in the F1-notch during the flare, the throttles are usually 30-something% and I have to cycle the F1 again to get them to fully idle. Is this a problem with my throttle-setup or something inherent in the NGX or actually prototypical?

Share this post


Link to post
Share on other sites

I've noticed that the A/T does not go to exactly idle at the start of VNAV descend. You can try to hit F1 after entering descend and will see the N1 is actually something like 5% higher than idle.

 

What are you using as evidence of this?  You cannot be using N1.  Watch the fuel flow for an indication of idle.

 

My throttle override is set to HOLD only, so I pull the Warthog throttles back to idle after TOGA and before 80 KIAS and keep them there until otherwise needed.  I've never seen either the 777 or 737 not go to flight idle at TOD (deice equip status excluded).

 

There is no problem here, if you want to descent at 260/.76 you'll need lots of speed brake. It's that simple.  In the US you rarely see a speed restriction below 280 kts; and those will be below FL230.  True, even at ECON descent you may need some drag but I never apply drag when FMS is recommending it unless I'm more than 10 kts above target and even then I may not use it if I'm coming up on a segment where I'm held at altitude where she will slow down on her own.

Share this post


Link to post
Share on other sites

I have noticed this as well.

 

I have the Saitek Throttle Quadrant and have mapped the reverser-notch (after the bottom end of the throttle range) to F1. On touchdown, after I've put the throttles in the F1-notch during the flare, the throttles are usually 30-something% and I have to cycle the F1 again to get them to fully idle. Is this a problem with my throttle-setup or something inherent in the NGX or actually prototypical?

It's inherent to the NGX due to the way flight idle has been simulated (because of an FSX limitation probably).

 

Are you selecting the 260 KIAS descent speed before descent commences? Reducing it will move the TOD and reduce the descent path angle. If you reduce the target during descent it may be harder to maintain. Also, how much fuel load do you have planned for landing? Excess fuel weight will make speed harder to control.

Share this post


Link to post
Share on other sites

There is no problem here, if you want to descent at 260/.76 you'll need lots of speed brake. It's that simple.  In the US you rarely see a speed restriction below 280 kts; and those will be below FL230.  True, even at ECON descent you may need some drag but I never apply drag when FMS is recommending it unless I'm more than 10 kts above target and even then I may not use it if I'm coming up on a segment where I'm held at altitude where she will slow down on her own.

 

Very interesting information.

How come 260/.76 requires a lot of drag? Shouldn't the VNAV simply calculate a more shallow descent path? Also since low descent speed is used to save fuel, if you have to use the speedbrakes more to make it work, doesn't that counteract the benefits?

 

EDIT:

 

kevinh, on 17 Jul 2015 - 9:59 PM, said:

It's inherent to the NGX due to the way flight idle has been simulated (because of an FSX limitation probably).

 

Are you selecting the 260 KIAS descent speed before descent commences? Reducing it will move the TOD and reduce the descent path angle. If you reduce the target during descent it may be harder to maintain. Also, how much fuel load do you have planned for landing? Excess fuel weight will make speed harder to control.

The descent speed was selected quite a distance before the TOD.

 

Is there anyway to get around the idle limitation? I thought I could solve it by binding the reverser detent to F1 but for some reason I have to hit F1 twice.

Share this post


Link to post
Share on other sites

 

 


How come 260/.76 requires a lot of drag? Shouldn't the VNAV simply calculate a more shallow descent path? Also since low descent speed is used to save fuel, if you have to use the speedbrakes more to make it work, doesn't that counteract the benefits?

 

Exactly the thinking process I had when I started with the 737 back in FS9 days but have gained a little knowledge on the way... not much but a little.  VNAV will not simply calculate a more shallow descent path because it doesn't know what you're trying to optimize.  If you want to optimize economy then select ECON descent.  The reason going slower does not less cost is because the air at upper flight levels is so very very thin.  The drag is negligible at speeds below Mach transition.  To see this for yourself, turn off fuel to the engines and maintain altitude and watch how long she takes to decelerate.  In the upper 30's this may surprise you.  Slowing down does mean you're in the air longer and that costs money no matter what your speed is so the ECON  is based on CI which incorporates the money value of time.  Watch the ECON descent speed increase with CI, and the lowest ECON descent speed is with CI = 0. If time is not valuable then use CI=0.

Share this post


Link to post
Share on other sites

The descent speed was selected quite a distance before the TOD.

 

Is there anyway to get around the idle limitation? I thought I could solve it by binding the reverser detent to F1 but for some reason I have to hit F1 twice.

I was trying this and found that if I planned for a 260 knot descent it held the airspeed well. If I reduced it after descent commenced it couldn't decelerate to 260 and I got DRAG REQUIRED messages. Fuel weight didn't have much effect on descent speed, but it would probably have affected the TOD point.

 

I don't hit F1, just let the NGX work as designed. The fact that the VC throttles aren't at idle isn't a bug, it's the way flight idle N1 is achieved in the NGX. The N1 is correct. If you hit F1 to force the throttles closed you get an artificially low N1.

VNAV will not simply calculate a more shallow descent path because it doesn't know what you're trying to optimize.  If you want to optimize economy then select ECON descent.

Yet selecting a lower descent speed certainly does affect descent path in the NGX. Not by much but it does. It's not so much about optimisation as 737 aerodynamics. The pilot is doing the optimisation (or not) by selecting the descent speed. The FMC calculates a profile based on that airspeed using the performance data it has. The TOD point is certainly affected which means the flightpath angle changes.

Share this post


Link to post
Share on other sites

My rough estimate from the real world:

 

1/3 of the time drag is required.

1/3 of the time thrust is required.

1/3 of the time no changes.

 

Ie. One third of flights don't need any fixing, 2/3 need a little help and it's usually only 10 knots or so for a small portion of the descent.

Share this post


Link to post
Share on other sites

I don't hit F1, just let the NGX work as designed. The fact that the VC throttles aren't at idle isn't a bug, it's the way flight idle N1 is achieved in the NGX. The N1 is correct. If you hit F1 to force the throttles closed you get an artificially low N1.

I knew this is related to the way PMDG bypasses the FSX limitation. The interesting thing is that when I hit F1 above FL290 N1 goes down only momentarily (and then goes back to the original N1), but below FL290 throttles actually stay at idle upon hitting F1.

 

Anyway this trick works for me so it may worth a try if nothing else works.

 

Btw I set the A/T Override to "never". Not sure if this matters.

Share this post


Link to post
Share on other sites

I knew this is related to the way PMDG bypasses the FSX limitation. The interesting thing is that when I hit F1 above FL290 N1 goes down only momentarily (and then goes back to the original N1), but below FL290 throttles actually stay at idle upon hitting F1.

 

Anyway this trick works for me so it may worth a try if nothing else works.

 

Btw I set the A/T Override to "never". Not sure if this matters.

Only PMDG can say which is the intended value for N1 at idle. I have no idea, but I tend to leave it alone. A few percent N1 near idle should not make much difference to thrust.

 

A/T Override settings will only affect things if the hardware throttle input has been left above idle or is prone to noise.

Share this post


Link to post
Share on other sites

Only PMDG can say which is the intended value for N1 at idle. I have no idea, but I tend to leave it alone. A few percent N1 near idle should not make much difference to thrust.

 

So would you recommend me not to bind F1 to the reverser detent then? I like it as an insurance in case the hardware-throttles happen to stay at a >0% level, hitting F1 will make it go 0 so autobrakes etc applies. But I don't want to break the simulation.

 

 

I did a flight yesterday with cost index 50 (descent speed 296/.792, if I remember correctly) and I never had to use drag (until touchdown ;) ). I did however do the F1 "trick" during the descent to shave off a few % of N1. 

 

I'm not sure if the F1 thing actually had any effect, and I feel it is harder for the aircraft to keep a low ECON SPD in descent compared to 280+.

 

Maybe I have unrealistic expectations, but it feels like I shouldn't have to use drag during long parts of the descent (5000-10000ft) if my planning is correct and there are no unexpected changes during the descent (new speeds, different routing, sudden changes in weather). Am I wrong?

Share this post


Link to post
Share on other sites

What am I doing wrong here?

I don't think you are doing anything wrong. The 737-800NGX will pick up speed during descent if the vertical profile is maintained. I do two things to try and keep the speed under control. Press F1 at FL290. Enter tailwinds in the descent forecast no matter what PFPX says. I set the tailwinds at 180 degrees opposite the direction of the STAR. If a STAR if heading 245 degrees I set the winds at 065 degrees increasing the velocity of the tailwinds as the plane descends. It might not be real world but it keeps the NGX under control. In fact thrust is usually required during the descent.

Share this post


Link to post
Share on other sites

So would you recommend me not to bind F1 to the reverser detent then? I like it as an insurance in case the hardware-throttles happen to stay at a >0% level, hitting F1 will make it go 0 so autobrakes etc applies. But I don't want to break the simulation.

I don't see how binding the reverse detent to F1 can create a problem. You can't break the simulation like that. If pressing F1 helps the descent then that's good too.

Share this post


Link to post
Share on other sites

I never knew about pressing F1 at FL290 ? So do i need to place my thrust levers to idle after a vnav des ?

Share this post


Link to post
Share on other sites

I normally place the TLs in the idle detent prior to the start of a descent just so they're in more or less the right place later on, but that's all.

 

You don't need to press F1 at FL290. If you do, you will get a lower N1 (and thus a steeper descent) but this is unrealistic as it is effectively circumventing PMDG's simulation of Flight Idle.

Share this post


Link to post
Share on other sites

I normally place the TLs in the idle detent prior to the start of a descent just so they're in more or less the right place later on, but that's all.

 

You don't need to press F1 at FL290. If you do, you will get a lower N1 (and thus a steeper descent) but this is unrealistic as it is effectively circumventing PMDG's simulation of Flight Idle.

 

Thanks simon. I always place my TL after reaching cruise. So is this a FSX limitation where PMDG has done something to flight idle?

Share this post


Link to post
Share on other sites

Exactly -- as I understand it, PMDG's flight idle simulation effectively moves the thrust levers off the backstop slightly (I imagine there must be an FSX limitation that prevents commanding a higher N1 without thrust lever movement).

Share this post


Link to post
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.
Note: Your post will require moderator approval before it will be visible.

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.

Sign in to follow this  

×
×
  • Create New...