Difficulty Forcing FMS out of VNAV CLB

[Hardy Heinlin]

I guess it will copy FL190 to the CRZ ALT and you will get a T/D at FL190 if it isn't already passed.

Good point. I think you are right. If you are above the altitude of the runway waypoint when you extend from it, I guess your current altitude would be your new cruise altitude; so it will go from CLB to CRZ, and when you get within the T/D range, it should shift to DES automatically. Does that sound right?

Yes.

[emerydc8]

Looking at the Atlas manual, I think this is what is meant too. I think they are assuming that you are at 3,000' and you extend from the CF, which also has an altitude of 3000'. If you were higher than 3000', I think your cruise altitude would be whatever altitude you are at when you execute it.

 Entering a new approach will allow the aircraft to transition from CRZ to the DES mode. When the new approach is entered, the altitude at the final approach course fix (CF) fix becomes the new CRZ altitude. A top of descent point (T/D) is displayed on the ND. If the T/C and T/D merge, the T/C will go away.
 If you are flying below the CF fix altitude and both the T/C and the T/D are displayed, you are in the CLB mode. You cannot get into the DES mode without first getting into the CRZ mode.

[Hardy Heinlin]

If the T/C and T/D merge, the T/C will go away.

The T/C will disappear anyway when the CRZ phase is activated.

[emerydc8]

Agreed.

[Hardy Heinlin]

PSX 10.0.9-beta5 is up and fixes the following:

• CLB-to-CRZ change in VNAV ALT now allowed by pushing MCP ALT knob.

• CLB-to-CRZ change now occurs by installing an approach under certain criteria.

• Inactive route will no longer appear on ND when a non-lateral modification exists in active route.


http://aerowinx.com/board/index.php?topic=3685.0

[emerydc8]

Thanks, Hardy! Will check it out tonight.
Jon D.

[emerydc8]

[EDIT] Hardy, I noticed your Beta update page after posting this. Does the on-approach waypoint include a runway fix as well? Could it be that I'm using an old situ? I'm not going to CRZ when I do this.

0.9.0030. When in CLB phase and when executing a modification that has an on-approach waypoint on first leg and whose constraint is not above current altitude and that lies within 150 nm, the FMC will copy current altitude to CRZ ALT and cause CRZ phase activation.

Hi Hardy,

I spent three hours checking this out using this 22R JFK-DAY situ https://drive.google.com/file/d/0B-WRh0Hf7VdZcnFYMFV5WkZVMTQ/view

(I've used this situ before in other posts too).

The cruise altitude in the box is set to 3000 and I just reset the MCP to 2000 on every takeoff, so I could see how it would work for a situation where you never reached your cruise altitude and wanted to return for a landing.

After takeoff runway 22R, I turned left HDG 140 at 1000', which put me on a downwind to RW 31L.
Once level at 2000' in VNAV ALT, I extended RW31L and executed. VNAV 1/3 correctly showed the new cruise altitude of 2000, but it stayed in CLB mode and (I think correctly) VNAV ALT. At 6 DME from JFK, I turned left HDG 040, did the LVSA and noticed that it went straight from CLB to DES. The FMA went to SPD||VNAV PTH and the VPP showed about 800' fly-up at this point. On the base and final leg, it descended at about 100 FPM to 1800', then eventually captured and flew the glidepath down to the runway okay.

On a second takeoff, I decided to take it out 9 DME on downwind. I noticed it did the same thing as before except that the VPP indicated 1600' fly-up and it descended on base to final at about 100 FPM to 1600', then eventually captured and flew the glidepath down to the runway okay. In both of these cases, I am not sure what caused it to go to DES and VNAV PTH -- the proximity of the aircraft to the glidepath, or the fact that I rolled in 500' on the MCP as part of the LVSA procedure on my base leg.

Maybe Peter (or anyone else) can give some thoughts on this, but here are mine:

When you extend from RW31L and execute, I think it should immediately go out of CLB mode -- It is not doing this. Not only should it have gone out of CLB mode, if the point that the FMC is looking at to make the descent path calculation is the RW threshold, I think the position I was in when I extended the RW fix (downwind and 4 miles from the threshold at 2000') should have resulted in going to DES right away and the VPP should have shown how far above the path I was at that point.

If I fly my base leg 6 miles out at 2000', the VPP should show that I am pretty close to being right on path. If I fly it 9 miles, it should show that I am low. In the sim on a 2-engine approach, when you extend from the RW fix, you are taught to use the 1000' fly-up point on the VPP to judge when to start your base leg. Excerpt from Atlas training material:

Extend the runway fix on the LEGS page.
 Use VNAV. Any available approach may be selected. For example, the PF may command: Put in the ILS 31L and extend the runway fix on the LEGS page. Bring up the ILS runway waypoint to 1L on the LEGS page. Put the inbound course into 6R and EXEC to create the runway extension. The runway is now the active waypoint and the ND displays distance to runway. Approach logic begins when within 25 miles of the runway; VNAV PTH is the pitch mode during speed intervene. [I think he means while descending on the path, but it should still be VNAV ALT on a downwind after extending]. If on downwind, a good time to turn base leg is when the VNAV path pointer displays 1000' fly up (1000' below path). The database altitude at the runway fix is threshold + 50'.

Does anyone have a video of an aircraft approaching T/D to get a view of how far below the path the VPP starts prior to actually intercepting the path? I know the scale itself is only 400', but there is a digital countdown on top. Where do the numbers start when approaching T/D at cruise altitude?

Maybe the situation differs in the case of being in the terminal area. If you extend from the runway and you are high already, I think it's going to go to DES right away. Once in DES, it will stay in DES and the VPP will continue to show how high you are. As you move away from the threshold on a downwind, it will show how low you are too (maybe it will display a number lower than you would see when capturing the path from a cruise altitude). The scenarios in the sim seem to always have you losing a second engine close enough to the airport that you are probably above whatever path you create and you have to go out on a downwind or base to set up for it anyway. So, maybe that's why it goes right to DES as soon as the runway fix is extended when you lose a second engine.

What I think we should be seeing after extending the runway fix and executing is CRZ, and maybe DES, depending on where you are. The FMA should stay in VNAV ALT until intercepting the glidepath on final. Right now, it is staying in CLB until DES goes active somewhere on a base leg after doing the LVSA procedure and then it goes right to VNAV PTH while being up to 1600' below the path. The aircraft should stay at 2000' in VNAV ALT until intercepting the path. Because you are within 25 miles, on-approach logic will cause VNAV to go from VNAV ALT to VNAV PTH upon intercept.

Thoughts?

Jon D.

[Hardy Heinlin]

Once level at 2000' in VNAV ALT, I extended RW31L and executed. VNAV 1/3 correctly showed the new cruise altitude of 2000, but it stayed in CLB mode and (I think correctly) VNAV ALT.

To get the CRZ mode and out of VNAV ALT, you must push the white MCP ALT knob.

When you extend from RW31L and execute, I think it should immediately go out of CLB mode -- It is not doing this.

It will do this when you are not in VNAV ALT.

Try your scenario with a CRZ ALT of 31000 and then extend from RW31L when climbing above 3000+. It will work as advertised.


|-|ardy

[Hardy Heinlin]

Beta 6 is up:

http://aerowinx.com/board/index.php?topic=3685.0

[emerydc8]

Thanks, Hardy. I will try the situ shortly.

To get the CRZ mode and out of VNAV ALT, you must push the white MCP ALT knob.

There should be no need to re-cruise here. [EDIT: By "re-cruise," I mean pushing the altitude knob after the new cruise altitude appears on VNAV 1/3 due to extending the RW fix.] There are three things that on-approach logic will do for you:

(1) If in VNAV ALT, the pitch mode can switch automatically to VNAV PTH when the VPP is centered; (2) The speed window can be opened on a descent and the pitch will remain in VNAV PTH to follow the VPP; and (3) the altitude can be changed by spinning the altitude selector knob from your DA through your current altitude to your missed approach altitude without it reverting to VNAV ALT.

Since we have already extended from the RW fix, we should be in on-approach logic, which should allow VNAV ALT to switch over to VNAV PTH on glidepath intercept. It is not like a re-cruise when you cannot get the altitude you want up high. There, you are not in on-approach logic, so you have to altitude intervene.

Jon D.

[emerydc8]

Hi Hardy,

I tried the situ using FL310 as a cruise altitude and 4000 in the MCP. I leveled off at 4000 in VNAV ALT on a left downwind for 31L. I extended RW31L fix and executed. It stayed in VNAV 1/3 with 4000 displayed as the new cruise altitude. I pushed in on the altitude knob and it went to VNAV 2/3 and VNAV PTH appeared. I turned base at JFK 12D and did the LVSA procedure. After a few seconds it switched over to DES 3/3, but the DES page was blank. It started a slow descent at 100 FPM to 3880. There was a VPP that indicated I was 120' low. The PROGS 2/3 VTK ERROR also indicated 120' low. I still had a blank DES page. Eventually it intercepted the path and started a descent on the VPP. It tracked the path down okay but DES page was blank the whole way.

On another takeoff, I extended my downwind to 16 miles and delayed the LVSA (specifically didn't set in 500' on the MCP). When I got to 12 DME established inbound, I rolled in 500'. It went into DES with the FMA indicating SPD||VNAV PTH, but the DES page was still blank. The VPP showed centered and PROGS 2/3 showed 0 VTK ERROR, even though it only descended at 550 FPM during the entire approach. It crossed RW31L at 2500'.

I am certain that it should stay in VNAV ALT while maneuvering for the approach in this situation (extending from the RW fix). Since on-approach logic is active here, it should only go to VNAV PTH when the path is intercepted on final. You should not have to altitude intervene to re-cruise so you can get it into CRZ and then DES mode. The act of extending the RW fix and executing should cause it to go right to CRZ and DES because the altitude at 1L (60') is lower than your present altitude (4000'). It should stay in VNAV ALT until intercepting the path.

Jon D.

[Hardy Heinlin]

Hi Jon,

you should try beta 7. It looks like you are using beta 6.

If you say it should stay in VNAV ALT until the route is intercepted, then that's a feature I haven't heard of before. I thought you wanted it to get out of VNAV ALT as soon as EXEC is pressed.

So, I'll just undo this ...

0.9.0033. Modification 0.9.0030 now also prevents VNAV ALT engagement.

... and we'll back to the previous version; but there you were unhappy with the fact that VNAV ALT remained engaged.


|-|ardy


I just modified it so that the CRZ ALT is set to the current altitude and that the CRZ phase activates -- and VNAV ALT remains engaged. This mode constellation works but it looks strange!

I'm adding the condition that the XTK error must be < 2.5 nm.

[emerydc8]

I'll try Beta 7.

I wanted it to get out of CLB as soon as the fix was extended, provided you are at or above that fix altitude (when extending from the RW fix you are always above it). It should stay in VNAV ALT until intercepting the glidepath (not the route). When it intercepts the glidepath, VNAV ALT will change to VNAV PTH provided on-approach logic is active (it will be here) and you have done the LVSA procedure.

It is confusing because there are situations where you could be level outside the CF or FF in VNAV PTH. This will happen when you are at the same altitude as what is in 1L (e.g., VNAV is happy to be there and the MCP is not constraining it). Let's say you were coming into JFK doing a VNAV approach to 31L and you had extended from the CF (ZACHS). If you were level in VNAV at 2000 outside ZACHS, this is where VNAV wants to be (2000 is in 1L); so it will say VNAV PTH and it will intercept the path and start down if you do your LVSA procedure prior to ZACHS (on-approach logic goes active at ZACHS).

[Britjet]

VNAV ALT will indeed change to VNAV PTH if on final approach logic AND intercepting from below the profile.
I think PSX already models this?
Peter.

[Hardy Heinlin]

This is now something different. This is the CRZ phase waiting for T/D sequencing. If there's no T/D in view, you might be too high, that is, you might have passed the T/D already because your high altitude requires a longer descent leg. You can't expect to get a T/D in front of you when you're 2 nm before the runway and 5000 ft above the runway. It will change to the DES phase promptly as soon as the MCP ALT is down. However, I just implemented an inhibit that prevents DES phase activation when XTK error is > 1.5 nm or intercept angle is > 90°.

When the DES phase activates, you will get VNAV PTH. The actual descent will start only when the aircraft is at or above the VNAV profile.

I'll upload beta 8 in a minute ...


|-|ardy

[emerydc8]

Hardy,

During the CRZ phase, what is telling PSX to start the descent phase? If it's a straight-line distance based on what's in 1L and you are downwind abeam the numbers when you extend the RW fix from 1L, you would be above path to start with. Shouldn't it go straight through CRZ to DES?

[Hardy Heinlin]

Beta 8 is up:

http://aerowinx.com/board/index.php?topic=3685.0

[emerydc8]

Thanks, Hardy. I can't keep up with you!

0.9.0038. Modification 0.9.0033 undone. When T/D is already passed when executing, VNAV ALT will now change to VNAV PTH only when XTK error is < 1.5 nm and intercept angle is < 90°.

Nice!

[emerydc8]

Hi Hardy,

I just tried Beta 8. It correctly goes out of CLB when extending the RW fix, but it is going to VNAV PTH too early. Is there a way to keep it in VNAV ALT until the actual glidepath is intercepted?

Jon

[Hardy Heinlin]

Nope. It needs a mile or so in advance to do the calculations. That's the best model you can get from me. One more mod and this thing will destroy the entire VNAV system that has been tuned and tested in the past 5 years. Too risky! Too many unpredictable side effects!

If it would start only at the path interception, you won't see the deviation indicator before and how the needle is coming down. There is no deviation indicator in CRZ.