Vertical Navigation of Approach Stepdowns for ILS with VNAV - GFC500/GTN750Xi

[PT20J]

Well, I really am going to have to go flying. Like I said, I haven’t done many ILS approaches for a long time, and they’ve all been vectors to final when I have. 

Looking at my G3X Pilot’s Guide, it appears that the GFC 500 should not revert to ROL upon CDI autoswitch if APR is set before the switch. 

[Marc_B]

9 minutes ago, PT20J said:

don't forget that the GFC 500 will not couple a LOC or ILS without a valid GPS signal.

True!  I wasn't thinking about that. In that case maybe this all all goes out the window.  GPS out, then VNAV won't work either as you'd be on HDG or TRK if using AP.  But I fly the ILS 35 into Greeley enough that it makes sense to still figure out the VNAV to Green quirks.

[midlifeflyer]

51 minutes ago, RoundTwo said:

How are you enabling GS on the trainer?

The GS  button on the virtual AP.

[midlifeflyer]

10 minutes ago, PT20J said:

Well, I really am going to have to go flying. Like I said, I haven’t done many ILS approaches for a long time, and they’ve all been vectors to final when I have. 

Looking at my G3X Pilot’s Guide, it appears that the GFC 500 should not revert to ROL upon CDI autoswitch if APR is set before the switch. 

It should not. It will on a manual autoswitch but should not in an automatic one. “Should” is the operative language. Basically, the autoswitch is designed to not trigger the reversion to ROL mode but you still have to check to make sure. 

[Marc_B]

17 minutes ago, PT20J said:

the GFC 500 should not revert to ROL upon CDI autoswitch if APR is set before the switch. 

That's what I've found IRL.  If the auto switch triggers the change from GPS to LOC then it won't revert if APR mode armed.  But if you switch it manually it reverts.  Or if you forget to arm APR, then it will still auto switch but will revert to roll/pitch when it switches.

[midlifeflyer]

52 minutes ago, donkaye said:

Yes, the example I gave above with the hold in lieu of PT on the KSCK ILS29R approach.

I’ll have to take a look at that one. Thanks.

[PT20J]

3 hours ago, Marc_B said:

That's what I've found IRL.  If the auto switch triggers the change from GPS to LOC then it won't revert if APR mode armed.  But if you switch it manually it reverts.  Or if you forget to arm APR, then it will still auto switch but will revert to roll/pitch when it switches.

OK, so say you have the following setup:

1. An ILS approach is active in the flight plan and you are at the IAF. VNAV altitude constraints have been set/verified and the altitude bug is set at or below the FAF altitude and you have previously enabled Transition to Approach in the GTN configuration mode. The localizer frequency is active in the GTN nav.

2. Lateral mode is GPS (green), Vertical is VNAV (green) with ALTV armed (white). Airplane is tracking the procedure course and descending via VNAV.

3. You select APR. Lateral mode is now GPS (green) with LOC armed (white), vertical mode is VNAV (green) with ALTV armed (white).

4. The GTN autoswitches the CDI from GPS to VLOC. The autopilot intercepts the localizer. Lateral mode is now LOC (green)

What is the vertical mode annunciation?

Skip

[donkaye]

2 hours ago, PT20J said:

OK, so say you have the following setup:

1. An ILS approach is active in the flight plan and you are at the IAF. VNAV altitude constraints have been set/verified and the altitude bug is set at or below the FAF altitude and you have previously enabled Transition to Approach in the GTN configuration mode. The localizer frequency is active in the GTN nav.

2. Lateral mode is GPS (green), Vertical is VNAV (green) with ALTV armed (white). Airplane is tracking the procedure course and descending via VNAV.

3. You select APR. Lateral mode is now GPS (green) with LOC armed (white), vertical mode is VNAV (green) with ALTV armed (white).

4. The GTN autoswitches the CDI from GPS to VLOC. The autopilot intercepts the localizer. Lateral mode is now LOC (green)

What is the vertical mode annunciation?

Skip

In simple terms and "covertly" (my definition) spelled out in some of Garmin's user manuals (I saw it in one of the G1000 GFC700 documents that I can't find now), I think what Marc is saying is with autoswitch on, when the Nav source changes, the AP will not go into default roll/pitch mode, but if autoswitch is off and the Nav source is changed MANUALLY, the AP becomes stupid and will revert to default roll/pitch mode.

[PT20J]

11 hours ago, donkaye said:

In simple terms and "covertly" (my definition) spelled out in some of Garmin's user manuals (I saw it in one of the G1000 GFC700 documents that I can't find now), I think what Marc is saying is with autoswitch on, when the Nav source changes, the AP will not go into default roll/pitch mode, but if autoswitch is off and the Nav source is changed MANUALLY, the AP becomes stupid and will revert to default roll/pitch mode.

I wonder if it really makes a difference whether you switch the CDI source manually or it autoswitches? I think the key is probably that the autopilot must be in APR before changing the CDI. 

But, what I'm curious about is why it seems to cancel out VNAV when the CDI changes. It would seem to make sense for the vertical mode to remain unchanged and the GS to arm waiting for capture.

I asked Joey Ferrerya a while back why the GTN Xi STC doesn't permit Transition to Approach for the GFC 500 when other Garmin autopilots allow it. There must be a reason why they left it out. He never said why, but did say that they were actively working on a change to permit it. So, perhaps it's related to the issue that @Marc_B raised and maybe there is more software that needs to be changed to get all the logic right. 

Edited December 18, 2023 by PT20J
Clarified that the GTN Xi STC does not allow Transition to Approach for the GFC 500

[PilotX]

During high-bandwidth events (maybe low bandwidth but low proficiency as well), switching is good. Switching is good for a proficient pilot. It won't fix everything (especially if you ignore flashing MSG or other indications) but recently I had the opportunity to fly with a pilot that didn't notice that we were still on magenta until about two miles from the FAF. Approach wasn't briefed and this was an IPC. I was waiting to see what was going to happen, clear for the approach but APPR not pushed. I had hoped for a missed approach but instead I got a flurry of button pushing and frequency checking. If there hadn't been some automatic (workload alleviating tasks) functions built into the system, I bet we still would have flown the approach but it would have been even less stabilized. For reference, this was a VFR day and I don't allow pilots to use their APs when flying approaches on an IPC. In this plane equipped with a GFC600 auto switching is enabled but it has blown through the final approach course in the past so there are some trust issues. @Marc_B has put a lot of effort into understanding how is plane is configured and what it will do, I need to visit my installation and see if I have auto-switch enabled. I was always curious in my previous plane and need to be more proficient in my Bravo, which I typically hand fly VFR. 

[Marc_B]

Okay, I went out and did some testing today.  Certainly I think found something workable.  First two approaches I had Auto Switch CDI enabled.  The last two I had CDI switch to Manual.

1) KGXY ILS35 from the north with HILPT.  7000 ft with FAF 6600 bugged.  AP: NAV/ALT with APR armed = GPS (Grn) / LOC (white), ALT 7000 (Grn) / GS (white).  What I found is that the CDI switched immediately on passing WELDS [fix before FAF & first fix on the FAC).  --> LOC (Green), ALT 7000 (Green) / GS (white).  This meant that CDI for LOC was perfectly aligned and the AP maintained 7000 until it captured GS at that altitude.

2) ILS35 from the DME Arc to come in laterally at 7500 ft with FAF 6600 bugged.  AP: Nav/ALT with APR & VNAV armed = GPS (Grn) / LOC (white), ALT 7500 (Grn) / VNAV & GS (white).  Constraint at WAVDI (IAF) "At 7500" with WELDS at or above 7000 and BUFFS (FAF) at or above 6600.  VNAV captured VDI at TOD --> VNAV (green) / GS & ALTV (white).  CDI switched immediately when GPS sequenced WELDS to BUFFS (CDI was 1 dot off).  As the GTN was directing VNAV to cross WELDS at 7000 (even though constraint was at or above), the ALTV captured 7000 at the same time as the CDI switch and the end result was LOC (Green), ALT 7000 (Green) / GS (white) and VNAV DROPPED OFF.

3) ILS35 from the DME Arc coming in laterally at 7000 ft with 6600 ft bugged.  Similar AP modes as #1 above.  Turned inbound on FAC towards FAF and no switch.  Half way down the segment I had a message alerting to ensure appropriate CDI chosen.  I picked LOC1 on the G500TXi and the LOC immediately captured with CDI centered, ALT 7000 (Green) & GS (White)

4) ILS35 from the DME Arc coming in laterally at 7500 ft with 6600 bugged.  Similar AP modes as #2 above.  Captured 7000 ft at WELDS and turned inbound.  GPS (egreen), ALT 7000 (green / GS & VNAV (white). TOD shown where the GS intercept would be at 7000 ft.  Message flags for check CDI source again; Changed CDI and immediately LOC (green), ALT 7000 (green) / GS (white) and VNAV DROPPED OFF.  The GS again captured at 7000 ft before the FAF and descended appropriately.

 

TLDR:  Autoswitch CDI appears to trigger when the leg into the FAF becomes active; even on the HILPT I was on the FAC for a few nm before arriving at WELDS and it switching.  GTN VNAV logic seems to treat my ILS at/above constraints as "at" constraints (vs when I flew an RNAV 35 it smoothly descended me down to FAF altitude constraint to capture GP).  VNAV drops off as soon as you switch the CDI either manually or automatically; expect the VNAV to level you off at your fix prior to FAF + change CDI if auto switch CDI enabled.  Neither the Auto or Manual CDI switch caused mode reversion or fault if you had APR armed and were on the FAC.

Regarding Manual vs Auto Switch CDI.  This is basically personal preference.  Either works equally well.  It would be very difficult and ill advised to leave the CDI in GPS to use the VNAV leading into the FAF and I couldn't figure out a way to get this to switch automatically in that fashion...vs RNAV with transition to approach enabled this is exactly what it does.  I was happy that leaving the Auto Switch enabled still allows Altitude capture at the fix prior to the FAF and simultaneously switches to LOC.  I think it makes sense to just level off there, capture the GS further out where LOC is less sensitive and allow a longer stabilized approach along the GS.

But I still find it curious that Garmin has the logic do something different on an ILS approach with VNAV than it does with RNAV/GPS.  Why does VNAV not work following VOR/LOC?  I'm curious what the GFC600 does since that IS approved for Transition to Approach...does the 600 require GPS source for VNAV??

[PT20J]

@Marc_B, first of all, good for you to go out and experiment to really understand how all this works. 

It looks like the behavior is what would be expected when Transition to Approach is not enabled. In other words, VNAV stops at WELDS (the fix before the FAF). It crosses WELDS at 7000 because that's its endpoint -- it's not looking at the 6600 constraint at BUFFS.

Why? Who knows? I have not had much success with anyone at Garmin answering why questions. (Thirty years ago, when Garmin was so much smaller, I could get to talk to an engineer sometimes and get an answer to a why question. But those folks are well insulated from customers these days).

But, I have a hunch. The G3X, G5 and GFC 500 were developed together by the Team-X experimental group. The G3X has a VFR GPS and it's own VNAV. At the time that the GFC 500 was designed, the Garmin IFR navigator would have been the GNS 430W. This navigator had a rudimentary VNAV which is similar to VCALC in the GTN.

The GTN Xi is developed by an entirely different group. When the GTN Xi enhanced VNAV came out it had VNAV capability all the way to the FAF with certain Garmin autopilots. This appears to work on the simulator in both GPS and ILS. For some reason, the GTN Xi STC requires that this feature be disabled for the GFC 500. This information is included in both the GTN Xi Installation Manual and the Pilot's Guide. There must be a reason for this decision. Perhaps it is because it doesn't work properly on an ILS. 

Anyway, I understand that Garmin is working on a change to allow Transition to Approach to be enabled with the GFC 500. We'll have to see if and when this is released if it changes any behavior.

Skip

[Marc_B]

I added annotation edits to the original post in RED for clarity.

9 hours ago, PT20J said:

It looks like the behavior is what would be expected when Transition to Approach is not enabled.

With Transition to Approach Enabled, for ILS the behavior is 1) VNAV drops off completely when CDI Auto switches at the Fix prior to FAF, or 2) if manual CDI switch is used, the VNAV levels at fix prior to FAF and plots the new TOD at the GS intercept prior to the FAF.

 

 

[PT20J]

It seems that the whole issue hinges on enabling Transition to Approach. 

When disabled per the STC, the VNAV appears to work as described in the documentation (at least for my G3X).

Frankly, the only value I see in Transition to Approach would be for approaches without vertical guidance. With Garmin adding +V advisory glidepaths to LNAV and VOR approaches, these are becoming rare. For an approach with vertical guidance, I never descend to the GP/GS altitude on the intermediate segment because I don't see any advantage in descending lower than necessary earlier than necessary. I just remain to the IF altitude (or vector altitude restriction if on a vector) until I capture the GP/GS.

Skip

[Marc_B]

@PT20J yes and no.  I'll agree that having vertical navigation stop at the fix prior to the waypoint seems similar to "Transition to Approach" being disabled.  But that's not quite what's happening.  VNAV with use on an ILS really has nothing to do with Transition to approach.  Rather, VNAV is falling off when the CDI is changed because VNAV requires GPS source of navigation.  So what I'm seeing is the same regardless of the Transition to Approach or or off.

This is different than VNAV with an RNAV approach with Transition to approach being enabled.  In that scenario, Enhanced VNAV has a smooth descent (if possible) from enroute that goes down through the arrival/approach all the way to finish up at the FAF altitude constraint where it then captures the GP to continue the descent.  VNAV falls off when the AP captures GP.  With VNAV on RNAV approach with Transition to approach DISABLED, Enhanced VNAV again creates a smooth (if possible) descent from enroute that goes through the arrival/approach to finish up at the Fix prior to the FAF altitude constraint where it captures ALT there as the FAF altitude constraint is WHITE (reference only).  So fix prior is the "end of the line."

The Transition to Approach "turns on" the altitude constraint for the FAF as well as the 3NM Visual Approach waypoint altitude constraint.  Otherwise it's there on the GTN, but inactive as reference only.

[PT20J]

But my point is that it works as advertised with Transition to Approach disabled. Enable it and it’s anyone’s guess because we don’t know why Garmin required it be disabled. 

[jlunseth]

On 12/18/2023 at 7:27 PM, Marc_B said:

 

But I still find it curious that Garmin has the logic do something different on an ILS approach with VNAV than it does with RNAV/GPS.  Why does VNAV not work following VOR/LOC?  I'm curious what the GFC600 does since that IS approved for Transition to Approach...does the 600 require GPS source for VNAV??

I have a completely different and more manual setup, but maybe it gives me the perspective to answer what is going on. I have a KFC200, which is capable of coupling to a GS in APR mode, but not capable of using VNAV to make a descent. I have to do that manually. I also have a GTN750 Xi that controls the AP, and a standby 430 that used to control the AP buts no longer does (it is not connected to the AP by any means). (Before the 750 was installed, the 430 would Autoswitch, and now, apparently because I don’t have a G500, CDI switching must be manual.)

So think about what the GTN and AP are doing when you switch into APR mode to fly an ILS. The control signal for a GPS approach (or any other GPS direction) is a digital signal from the GPS computer. The ILS is analog. It comes from the Nav radio in the 750, not from the GPS brain. So when you switch from RNAV mode to APR, you switch off the digital signal and switch on the analog signal from the localizer. You would not want the digital signal from the GPS to control any longer, because you would have two potentially conflicting control signals operating at the same time. At any rate, and I would have to go back and check the regs, but I am pretty sure they prohibit using the GPS signal to legally fly an ILS, so Garmin does not have a choice, the AP has to use the analog signal.

What I do to make a FAF intercept altitude, if it is lower than the FAF, is to simply use the trim switch on my AP to make the descent. The way my AP works, this has the advantage of acting as a “dead man’s” switch, so if I take my finger off it, it holds altitude at that point rather than diving lower while my attention is distracted by something else going on. Or I use the lazy pilot route and just let the AP couple to the GS at the higher altitude and fly down to the FAF.

[PT20J]

I asked Bruce Williams how the GFC 600 works. He sent me this video clip. Note that the equivalent GFC 600 mode is VPATH. The GTN auto switches to VLOC after the IF. The autopilot captures TOD and follows the GTN vertical track down to the GS intercept altitude at the FAF at which point it switches to GS. I would expect Garmin to make the GFC 500 do the same thing when transition to approach is approved.

 

[Marc_B]

@PT20J I'll be interested to see how this changes with future updates!  Thanks for the video with the GFC600...certainly shows the behavior that I would expect with "Transition to Approach."  Makes me wonder why the GFC500 was not rolled out with that as well?  But I don't think we'll ever know what goes on behind the G-curtain! ha.  (and I wonder how similar or dissimilar VPATH is to VNAV?)

[Marc_B]

In the latest issue of IFR magazine, it discussed false glideslopes on an ILS and the danger of intercepting ILS glideslope from above which might lead to your autopilot capturing a false repeated glideslope.  This made me think of autoswitch vs manual switch and gave me a reason to just keep it autoswitch. 

My thought was if I'm level at the fix prior to FAF (or ATC leveled me off even higher) and I'm getting busy getting things done, talking to ATC, getting ready to land...forget to switch CDI so the AP flies the GPS overlay LOC course but GS doesn't capture and I blow though the GS without recognizing it until I'm at/past the FAF and now above the glideslope.  So I quickly change CDI, re-arm the approach...what are the chances that I'll capture a false signal?  Depending on how high you are and when you "catch" the error, maybe unlikely or maybe so... but the thought was, can you trust the needles now?  With an RNAV approach you don't have to worry about false glideslopes so you can "trust" your needles.  High on an ILS, maybe not!

The article in IFR also mentions that the window for LOC capture is broader than the window for glideslope capture and that "most systems have logic that the LOC must be captured first before the GS."  So having autoswitch engaged seems to insure that 1) I capture the LOC signal well before the GS, 2) acts to ensure that I'm less likely to fly through the GS, and 3) allows me to be monitoring the system to see the big picture rather than focused pushing a button.

I've never captured a false glideslope and now I just want to go see if I can "find" one with the AP off and flying VFR.

[Pinecone]

On 12/17/2023 at 4:29 PM, PT20J said:

Well, I really am going to have to go flying. Like I said, I haven’t done many ILS approaches for a long time, and they’ve all been vectors to final when I have. 

Looking at my G3X Pilot’s Guide, it appears that the GFC 500 should not revert to ROL upon CDI autoswitch if APR is set before the switch. 

Wow, I did not realize that you can control the autopilot from the G3X.  

That really cements my plan to put the 507 controller just to the left and above the throttle.   I don't need to look down for buttons, I can use the G3X.

 

[Pinecone]

Hmm, I wonder if there is an approach that the altitude at the last fix before the FAF is above the glide path?

I would think not.  

But that would be the only case I can think of where there would be an issue.

[Marc_B]

@Pinecone What I'm referring to is if you're held high and are far enough out you are still under the GS.  You start below the GS, but once you cross the GS prior to the FAF you're above it.  i.e. if you are at 7000 feet and using GTN and GFC500, the AP will keep you at 7000 feet and intercept the GS several NM prior to the FAF then descend.  If you don't capture at that point (APR not armed, or CDI not switched) the AP will fly you though the GS and you're now above it.  This is accentuated if (for traffic separation) ATC gives you a "maintained 8000 till established, cleared ILS approach" where they intentionally keep you higher for traffic separation.

In the picture below, the line connecting BUFFS with WELDS is not actual glideslope; the true glideslope does not intercept 7000 feet at WELDS but rather a few NM prior to BUFFS.

But that being said, is the fix prior to the FAF is set up specifically to put you under the GS/GP?  I'm not sure mountainous approach procedures that have a "dive and drive" flow to them (think Aspen, CO - KASE),  were designed to have the waypoint altitude constraints be smooth, but rather stepdowns that solely act as minimum descent altitude between segments (at or above).  That's the reason there is a GS/GP intercept shown on approaches.

However, if ATC is vectoring you onto an approach, they are supposed to bring you in under the glideslope (see below) and "where adequate radar coverage exists, radar facilities may clear an aircraft to any fix 3 NM or more prior to the FAF, along the final approach course, at an intercept angle not greater than 30 degrees,"  (7110 4-8-1 a.5.), or at least 2 miles outside approach gate (with some specified exceptions i.e. pilot request)

APPROACH GATE- An imaginary point used within ATC as a basis for vectoring aircraft to the final approach course. The gate will be established along the final approach course 1 mile from the final approach fix on the side away from the airport and will be no closer than 5 miles from the landing threshold.

 

[midlifeflyer]

1 hour ago, Pinecone said:

Hmm, I wonder if there is an approach that the altitude at the last fix before the FAF is above the glide path?

I would think not.  

But that would be the only case I can think of where there would be an issue.

One may exist but it will be rare. If I recall correctly, part of the design criteria is that the IF to FAF segment is not more than 3 degrees. 

[midlifeflyer]

13 minutes ago, Marc_B said:

@Pinecone What I'm referring to is if you're held high and are far enough out you are still under the GS.  You start below the GS, but once you cross the GS prior to the FAF you're above it.  i.e. if you are at 7000 feet and using GTN and GFC500, the AP will keep you at 7000 feet and intercept the GS several NM prior to the FAF then descend.  If you don't capture at that point (APR not armed, or CDI not switched) the AP will fly you though the GS and you're now above it.  This is accentuated if (for traffic separation) ATC gives you a "maintained 8000 till established, cleared ILS approach" where they intentionally keep you higher for traffic separation.

In the picture below, the line connecting BUFFS with WELDS is not actual glideslope; the true glideslope does not intercept 7000 feet at WELDS but rather a few NM prior to BUFFS.

 

The distance between WELDS and BUFFS is almost 7 nm. At 3 degrees, you need less than 1.5 miles to descend those 400'. Staying at 8,000 you'd still intercept the GS from bellow.

The Profile View is not to scale.

Edited January 17 by midlifeflyer