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

[Marc_B]

Went up today and feel like I have everything squared away for using VNAV into RNAV approach.  I can verify constraints, arm both VNAV as well as APR, and given the lateral course is GPS for an RNAV approach, the VNAV will function all the way through initial step downs smoothly all the way to the FAF** to capture GP.

But ILS approach is different due to switch to Localizer (green needles) that throws some wrinkles as VNAV will NOT function unless it's following a GPS course.

I'm currently set up for Auto-switch CDI.  So step downs aside, I have to have the APR armed on the GFC500 by the time the auto-switch occurs (Garmin just says within 1.2nm from course - See below - Garmin pic and link).  If I don't have APR armed, then when the auto-switch occurs the AP reverts to Roll/Pitch mode.  With APR armed, the transition (from GPS to LOC) is smooth and lateral course is followed correctly.

https://support.garmin.com/en-US/?faq=bx9OhUS9XL0rij0JBrpZ36  (Describes when the GTN (Xi) auto switches the CDI)

EDIT: *** see picture below; the Auto Switch CDI seems to occur not at 1.2nm, but specifically when the GPS sequences the waypoint to the FAF as the active leg (in this example as soon as WELDS to BUFFS becomes active, the CDI switches.  ***

BUT SPECIFICALLY DISCUSSING VERTICAL NAVIGATION ON ILS COUPLED APPROACH: 

KGXY ILS 35 approach has IAF WAVDI (@/abv 7000), WELDS (@/abv 7000), BUFFS - FAF (6600)...if you're at 16,000 feet flying in from the east direct to WAVDI, the GTN750Xi programs VNAV in the box to set the bottom of descent to cross WELDS at 7000 (even if the constraint is listed at or above).  No problem so far.  The issue is that the auto switch occurs AT WELDS so the VNAV no longer functions past the switch.  So what happens is that the auto switch CDI effectively disables VNAV (** vs. when on an RNAV approach you can still follow VNAV with the GFC500 all the way to the FAF--or one waypoint prior to FAF if Transition to Approach is disabled).

 

So my question is, what is the optimum method/setup with the GTN VNAV leveraged with the GFC500 VNAV automation for ILS/LOC/VOR?

1. Should the CDI auto switch be DISABLED?  (Edit: personal preference/works either way) if the unit didn't auto-switch could you continue inbound on the GPS overlay and just make sure you switched by the FAF?  (Edit: yes, but CDI would need to be switched by the FAF, and VNAV falls off as soon as you switch CDI)  But if you have APR armed it's expecting green needles, right...so would it be able to fly GPS overlay (prior to the FAF) for the LOC without reverting to roll/pitch? (Edit: With manual CDI it will fly GPS overlay, but as soon as you switch CDI 1) if APR armed the LOC captures immediately and AP sequences correctly, or 2) if APR is not armed, AP reverts to Roll/Pitch.)  But even if so, you'd have to make sure you 1) switched CDI, then 2) armed APR...if you armed approach prior switching to the LOC, would the ILS approach loaded would make the autopilot revert to roll-pitch?

FROM AFMS for GFC500:  "If the Course Deviation Indicator (CDI) is greater than half scale deflection, the autopilot will arm the LOC mode. The pilot must ensure that the current heading will result in a capture of the selected course. If the CDI is within half scale deflection, the autopilot will enter the capture mode when the APR key is pressed."

2. If CDI auto switches and VNAV won't work on green needles, then would the optimum procedure be to just set the IAF as an AT CONSTRAINT?  (i.e. cross WAVDI "at" 7000 feet)  Then aircraft would be level at 7000 feet when the switch to LOC happened and if APR was armed then Autopilot would remained coupled and turn inbound to intercept LOC.  Then you'd have to manually descend to 6600 feet for FAF vs. stay at 7000 and capture the GS at 7000 prior to FAF.  BASICALLY, it seems like you can't/shouldn't expect to use GFC500 VNAV within 1.2nm from the Localizer Course, or within 15 nm from the FAF on a straight in approach.  (EDIT: for some reason, the GTN is treating WELDS at or above 7000 as "at 7000"; so in example above, VNAV descends to 7000 crossing WELDS and immediately switches CDI and captures LOC + ALT7000 with GS armed.  VNAV falls off at WELDS.)

Does anyone have any tips for using VNAV with ILS approaches?  @donkaye @midlifeflyer @PT20J @PilotX 

 

Yes I realize my question is somewhat academic as ATC would most likely be giving me step downs, rather than using VNAV; and if ATC is vectoring me in Heading Mode then VNAV is out the window anyways.  So maybe my question is just "Do you have auto switch CDI enabled?"

Edited December 19, 2023 by Marc_B
added annotation to approach picture and answered questions in red

[PeteMc]

It's been a long day, consider me dense....  I'm missing what you're asking and why you care about VNAV on an ILS, so maybe rephrase it?  

I get that you may be using VNAV to get down to WAVDI and may be you even need to continue the descent to WELDS.  But after than I'm lost on the VNAV.  At this point you've switched to the ILS, you fly straight at 7000 until intercept and follow it down.  Crossing BUFFS you confirm you're crossing at 6600ft.  

Again, long day, so everyone feel free to dump on my since my heads not in it at the moment. 

[PT20J]

I haven’t tried this (I cannot remember the last time I flew an ILS), but based on my understanding of the GTN, I don’t believe it should cancel VNAV when it switches from GPS to VLOC. The GTN GPS is always active and that information is what is displayed on the GTN. (That’s why you get the warning message not to use it for navigation when selecting a non-GPS approach).

Changing from GPS to VLOC only changes the source for the external CDI (and autopilot). Switching the CDI from GPS to VLOC arms the autopilot to capture the localizer and glideslope but doesn’t change the lateral and vertical modes until localizer and glideslope capture. So, I would not expect switching the CDI to VLOC to affect any vertical mode (including VNAV) until GS capture.

Skip

[Marc_B]

@PeteMc Simplified issue is that VNAV works differently on ILS approach than it does on an RNAV approach; and this is due to Enhanced VNAV not working unless navigation source is set to GPS.  The issue is when the GFC500 has an "error" it reverts to roll/pitch mode and stops doing what you wanted it to do (so you better pay attention).  So the issue is that on the ILS procedure that I was flying there were 2 VNAV altitude constraints that were active but were unavailable for use as VNAV won't work with Green Needles.  Plus the Enhanced VNAV profile in the GTN did not account for this and placed the BOD at a location past where the CDI would autoswitch.  Just trying to have a better understanding of how to use VNAV in general and with the approaches I fly specifically.  I find that GFC500/GTN VNAV is one of the more complicated and least explained mode in the pilot guides/AFMS.

But basically wondering 1) how others have used VNAV into an ILS/Green Needle approach, and 2) do others have CDI auto switch enabled or not?

2 hours ago, PT20J said:

So, I would not expect switching the CDI to VLOC to affect any vertical mode (including VNAV) until GS capture.

Enhanced VNAV will not function if any of the following conditions are true.

    - Navigation source is not set to GPS

    - VNAV not enabled on VNAV profile page

    - No waypoints exist in the flight plan with appropriate altitude constraints

    - Lateral mode is not actively engaged to GPS

    - OBS mode is active

    - Dead reckoning mode is active

    - Parallel Track is active

    - Aircraft is on the ground

    - Aircraft is between the FAF and MAP of an approach

https://support.garmin.com/en-US/?faq=W7RRdCYWC04hI8kxfraxf7

[PT20J]

I will have to try it in the airplane. But I tried it on the GTN Xi simulator and VNAV worked all the way to the FAF even after it switched CDI source to VLOC. 

If you watch the GTN it is always showing magenta and using the GPS. When the CDI changes from GPS to VLOC coming out of the DME arc, the GTN scale changes from TERM to LNAV and VNAV TOD/BOD continues to display on the map.

What PFD are you using: G3X, G5, GI 275?

[Marc_B]

@PT20J G500TXi, G5 + GFC500, GTN 750Xi. 

[PeteMc]

2 hours ago, Marc_B said:

Simplified issue is that VNAV works differently on ILS approach than it does on an RNAV approach;

Okay, I get what you're saying about where it switches.  And I saw what @PT20J said about the simulator showing the VNAV still on the GTN screen, so it will be interesting to see what happens when he is in his plane. 

I'll grant that maybe there is an issue with this Approach, but I'm still saying the VNAV goes out the window on an ILS.  VNAV is a calculated linier descent, while an ILS may not be linier if there is any interference (terrain, structures, etc.) with the transmitted (RF) Glide Slope.  Sure the paths should be very close, but once you're on the ILS, the VNAV is irrelevant.  (But still get you think it is switching too soon.) 

 

[midlifeflyer]

If I’m following your description correctly, using autoswitch, and APR armed as soon as cleared for the approach, I would be happy with VNAV bringing me to 7000, the GPS navigating the arc, the switch to VLOC approaching localizer interception at WELDS, and maintaining 7000 until capturing the glideslope. 

Autoswitch enabled is a separate question. I leave it enabled. I typically switch manually and treat autoswitch as a backup to my failure. I have accepted that manual switch means AP drops out of NAV mode, so verification is SOP.

 

[RoundTwo]

31 minutes ago, midlifeflyer said:

If I’m following your description correctly, using autoswitch, and APR armed as soon as cleared for the approach, I would be happy with VNAV bringing me to 7000, the GPS navigating the arc, the switch to VLOC approaching localizer interception at WELDS, and maintaining 7000 until capturing the glideslope. 

(Then again, it what I would do manually, so it may not be what you want it to do)

I’m running a very similar set up as @Marc B, with G3X, G5, GTN750, GFC500, so I’m going to try this out.

The issue I see is when I load and activate a DME arc ILS approach in GP or the GTN trainer, is that the first altitude constraint is the IAF on the LOC, not the beginning of the arc. Knowing this now, I would probably add an along track WPT, before the IAF at the beginning of the arc, to get VNAV out of the way before becoming established on the arc.

Next time I’m out, I’ll run over to KGAD for the ILS 24 to try this out IRL.

[midlifeflyer]

One thing to be aware of is that the Jepp database depicts this approach differently than the Garmin database which means the AP will also fly it differently. I ran it in the trainer with the Garmin database. Here's what I did and it did. I wouldn't expect the real thing with compatible AP and PFD to work any differently.

I started at ZERNA at 7200. Constraint for the arc was to cross WELDS at or above  7000. NAV mode with GS enabled (which pretty much emulates an autopilot set at VNAV with transition to approach  enabled).  I set the altitude in the PFD at 6600, the FAF altitude.  Entering the arc, VNAV calculated a TOD to bring me down to 7000 at WELDS and maintained 7200 until then. Nearing WELDS, still in pink needles, the turn to final was anticipated. VNAV continued to bring me down until it autoswitched to green needles and the GS armed. At that point, VNAV terminated, GS armed, the system "saw" we were below glideslope, and the AP maintained altitude until capture. No surprises there.

I didn't run it with the Jepp database but I loaded it to take a look. The big difference between the two is that where the arc to the FAC in the Garmin flight plan is WAVDI → WELDS → BUFFS

with the Jepp database it's WAVDI → D170Q → Intercept → WELDS → BUFFS. That involves

• a separate "waypoint" for the arc (D170Q)
• a heading leg for a 30° intercept to final (Intercept)
• no altitude constraint for WELDS

I didn't "fly" it, but I would expect the Jepp to fly it very differently too. Rather than turn anticipation for WELDS, I would expect it to maintain 7000,  turn wherever it need to before WELDS to create that 30° intercept. Since I didn't fly it I'm not sure at what point it would autoswitch,

 

 

[Marc_B]

Since KDEN Bravo is south of this approach I haven't tried a few other options I've been curious about but that are within the "Shaded Box" of auto switch.

1) when does the CDI auto switch if you were flying direct WELDS as your final approach fix coming in from the south?  And if VNAV was descending you from "X" down to 7000 feet at WELDS how would this interfere with your descent?  Say you were on the extended FAC direct to WELDS, then surely the GTN wouldn't try to capture LOC until WELDS right?  This should be similar to DME arc I think...when the GTN changes to "arm" the capture of LOC, then the auto switch happens and VNAV disabled.

2a) if you come in from the north and fly the course reversal at WELDS when does the CDI auto switch?  So in this case again descending on VNAV to 7000 feet at WELDS...the hold would be entirely within the shaded box for auto switch...but if you were holding at WELDS I think that it would "hold it" in the pink.  But I suspect as soon as the GTN sequences to WELDS inbound, that this "arms" the auto switch as it's looking to capture the LOC.

2b) if you're coming in from the north and plan to fly the course reversal I think the solution is to program constraint to cross "At" 7000 when you cross WELDS the first time (which would defeat the purpose of losing altitude in the hold), otherwise I suspect the GTN VNAV default will treat it as descending to 7000 at WELDS inbound...at what point on the inbound leg of a course reversal does the GTN "arm" the LOC capture...at WELDS? approaching WELDS?

The only thing that I could find about the auto switching behavior in the pilot guide is same as the Garmin link I posted above.  But I think 1) if the GTN is in a configuration to join a LOC course 2) LOC is armed in the GTN (approach active), then 3) it will switch somewhere in the 2.4nm x 15nm box.

[PT20J]

I see what’s going on. Although the GTN Xi can provide GPS guidance including VNAV all the way to the FAF, the GFC cannot use it after the CDI change. Unlike intercepting the localizer in HDG, where the GFC remains in HDG after the CDI source changes to VLOC, changing the CDI source from GPS to VLOC causes the GFC to go to default ROL and GFC VNAV is unavailable  

So, the work around would be to disable CDI auto switch and let the autopilot fly  onto the intermediate segment with the CDI in GPS and then somewhere before the FAF switch to VLOC, APR, and whatever vertical mode you prefer.

I wonder how the GFC 700 and 600 handle this?

Skip

[donkaye]

This won't answer the above questions but:

1.  My plane lives at a Class C Airport, KSJC, where I have a hangar and that has an ILS.  It is Notam'd "OUT" at least 20% of the time while work is being done to the runways.

2. Given a choice, and that's close to 100%, I don't use an ILS anymore except maybe to practice and teach.

3. I  don't trust "Autoswitch", so I personally have disabled it.  For example, with autoswitch enabled,  running the ILS into KSCK from IPDEW with the Hold in lieu of PT on my "own Nav", having been told to "Cross IPDEW at 3,000, Cleared ILS 29R Approach" and with both VNAV and APR armed, the plane properly crosses IPDEW and then descends to 1,800 to cross IPDEW inbound at 1,800.  As the plane turns inbound to intercept the LOC, the CDI autoswitches and that's it for flying the ILS on AP. The plane goes to "la la land" not knowing what to do. 

4.  In 3 above with autoswitch disabled, I fly the PT in NAV mode using VNAV and once at 1,800 inbound switch the CDI to VLOC and push APR.  As long as you remember to flip the NAV frequency to the ILS (the GTN places it in the standby register automatically when you load the approach), the GS and ILS are flown perfectly.  I use the KISS principle.

[PT20J]

I think @donkaye has a great point. The problem with automation is that it is very disorienting if it doesn’t do what you expected. And, it is easy to fail to get something right, like setting the altitude bug before selecting VNAV or not making the localizer frequency active before an ILS approach, that will cause the automation to fail to do what you expect. The more things you have to do manually, the more you stay “in the loop.” The best use of automation, IMO, is to avoid task saturation by letting it do some of the work. Just because it can do something doesn’t mean you should let it.

In Don’s example, he knows that APR must be selected after changing the CDI source to VLOC. But, if it auto switches, he’d have to notice that it switched and then select APR or it goes into la la land (ROL). And, unless you memorize the diagram and description in the manual, you don’t know exactly when this will happen.

Skip

 

[midlifeflyer]

1 hour ago, Marc_B said:

Since KDEN Bravo is south of this approach I haven't tried a few other options I've been curious about but that are within the "Shaded Box" of auto switch.

1) when does the CDI auto switch if you were flying direct WELDS as your final approach fix coming in from the south?  And if VNAV was descending you from "X" down to 7000 feet at WELDS how would this interfere with your descent?  Say you were on the extended FAC direct to WELDS, then surely the GTN wouldn't try to capture LOC until WELDS right?  This should be similar to DME arc I think...when the GTN changes to "arm" the capture of LOC, then the auto switch happens and VNAV disabled.

2a) if you come in from the north and fly the course reversal at WELDS when does the CDI auto switch?  So in this case again descending on VNAV to 7000 feet at WELDS...the hold would be entirely within the shaded box for auto switch...but if you were holding at WELDS I think that it would "hold it" in the pink.  But I suspect as soon as the GTN sequences to WELDS inbound, that this "arms" the auto switch as it's looking to capture the LOC.

2b) if you're coming in from the north and plan to fly the course reversal I think the solution is to program constraint to cross "At" 7000 when you cross WELDS the first time (which would defeat the purpose of losing altitude in the hold), otherwise I suspect the GTN VNAV default will treat it as descending to 7000 at WELDS inbound...at what point on the inbound leg of a course reversal does the GTN "arm" the LOC capture...at WELDS? approaching WELDS?

The only thing that I could find about the auto switching behavior in the pilot guide is same as the Garmin link I posted above.  But I think 1) if the GTN is in a configuration to join a LOC course 2) LOC is armed in the GTN (approach active), then 3) it will switch somewhere in the 2.4nm x 15nm box.

First, the DEN TRACON I remember (I flew out of KAPA for 20 years) was extremely helpful so I think you can get them to give you what you want.  I've gotten vectors to  final on that approach. Direct to the WELDS IF, cleared straight in is even easier for them, whether practice or for real., especially since WELDS is 1,000 below the Bravo shelf.  Aside from that there's no reason you can't use the Garmin trainer to check this stuff out. As I mentioned, the funky autopilot emulates VNAV transition to approach when GS is turned on.

But, generally speaking, autoswitch from GPS to LOC will occur when the PFAF FAF becomes the active waypoint. Any active VNAV will terminate at that point, the GS will activate and altitude will hold until GS capture.

And I agree completely with @donkaye on his two main points. 

31 minutes ago, donkaye said:

2. Given a choice, and that's close to 100%, I don't use an ILS anymore except maybe to practice and teach.

3. I  don't trust "Autoswitch"

On the ILS I always include a fully coupled ILS to the missed when giving transition training and most times for IPCs precisely because I think it's the most complicated thing we have in approaches these days.  For real flying, "We like the RNAV" when the ILS is offered or advertised. 

And while I don't trust autoswitch eithers, I don't disable it. I treat is as a backup to my failure to manually switch.

[donkaye]

1 hour ago, midlifeflyer said:

And while I don't trust autoswitch eithers, I don't disable it. I treat is as a backup to my failure to manually switch.

The problem with that from my point of view is that you don't have control of when it switches.  Things can happen (or don't happen) after the switch that is unintended.

[midlifeflyer]

1 hour ago, donkaye said:

The problem with that from my point of view is that you don't have control of when it switches.  Things can happen (or don't happen) after the switch that is unintended.

Perhaps. That's why pilot technique and assessments of feature utility differ. In may case, since my manual switching is invariably at or before the time it would autoswitch, I don't see an issue. Do you have an example scenario where it might  switch when I didn't want it to? That would definitely be a problem.

[Marc_B]

I appreciate the help and safety of automation when I understand what it's doing and what it looks like when it fails.  I'm sure that's why all of us who have solid autopilots would probably be flying coupled approaches in actual instrument conditions (rather than hand flying).  With Enhanced VNAV and specifically coupled to Green Needle approach I'm trying to explore what it's doing/expecting and how it fails.

But GPS is potentially delicate and certainly the scenario exists with planned or unexpected GPS outages; so I like being comfortable with Green Needles, especially since it seems to be the mode that is rife with more quirks and complications.

I also have to say that blowing through a FAC is overall easier (for me) to catch sooner, vs. not switching to Green Needles but everything else is working as intended.  Even when Roll/pitch defaults in the AP, I don't think that I've gone farther than 1 dot deflection.  I imagine both have Pros and Cons, so just trying to explore what everyone else has found and figure out what makes the most sense for me.

Regarding AP failure modes: I'd rather have autoswitch turn off VNAV and level off at an altitude but still follow the approach, than have the autopilot revert to roll-pitch and require me to catch it and figure out how to fix it.  Hence, I like autoswitch and am okay with following this to make sure it happens rather than having to push more buttons.  Usually when the autopilot reverts to roll/pitch, I find that most efficient path is to turn off the autopilot, reestablish course, troubleshoot why that happened, and then re-activate AP. 

 

[RoundTwo]

5 hours ago, midlifeflyer said:

One thing to be aware of is that the Jepp database depicts this approach differently than the Garmin database which means the AP will also fly it differently. I ran it in the trainer with the Garmin database. Here's what I did and it did. I wouldn't expect the real thing with compatible AP and PFD to work any differently.

I started at ZERNA at 7200. Constraint for the arc was to cross WELDS at or above  7000. NAV mode with GS enabled (which pretty much emulates an autopilot set at VNAV with transition to approach  enabled).  I set the altitude in the PFD at 6600, the FAF altitude.  Entering the arc, VNAV calculated a TOD to bring me down to 7000 at WELDS and maintained 7200 until then. Nearing WELDS, still in pink needles, the turn to final was anticipated. VNAV continued to bring me down until it autoswitched to green needles and the GS armed. At that point, VNAV terminated, GS armed, the system "saw" we were below glideslope, and the AP maintained altitude until capture. No surprises there.

I didn't run it with the Jepp database but I loaded it to take a look. The big difference between the two is that where the arc to the FAC in the Garmin flight plan is WAVDI → WELDS → BUFFS

with the Jepp database it's WAVDI → D170Q → Intercept → WELDS → BUFFS. That involves

• a separate "waypoint" for the arc (D170Q)
• a heading leg for a 30° intercept to final (Intercept)
• no altitude constraint for WELDS

I didn't "fly" it, but I would expect the Jepp to fly it very differently too. Rather than turn anticipation for WELDS, I would expect it to maintain 7000,  turn wherever it need to before WELDS to create that 30° intercept. Since I didn't fly it I'm not sure at what point it would autoswitch,

 

 

How are you enabling GS on the trainer?

[donkaye]

40 minutes ago, midlifeflyer said:

Perhaps. That's why pilot technique and assessments of feature utility differ. In may case, since my manual switching is invariably at or before the time it would autoswitch, I don't see an issue. Do you have an example scenario where it might  switch when I didn't want it to? That would definitely be a problem.

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

[Marc_B]

12 minutes ago, RoundTwo said:

How are you enabling GS on the trainer?

Using the Garmin Aviation Trainer it has a NAV and GS button.  But I don't think that the GS mimics the GFC500 VNAV behavior exactly.  Although, I do think that the GTN VNAV calcs and flight path behavior in the trainer works as expected.

[RoundTwo]

4 minutes ago, Marc_B said:

Using the Garmin Aviation Trainer it has a NAV and GS button.  But I don't think that the GS mimics the GFC500 VNAV behavior exactly.  Although, I do think that the GTN VNAV calcs and flight path behavior in the trainer works as expected.

I wasn’t aware of that trainer, I was only familiar with the GTN trainer.

[Marc_B]

17 minutes ago, donkaye said:

the hold in lieu of PT on the KSCK ILS29R approach

@donkaye I need to explore what happens and when the autoswitch would happen with a HILPT.  I suspect that if you're suspended and holding at the published hold, that CDI would NOT switch.  But when you flew it just as a course reversal, or when you unsuspended your hold, when inbound IPDEW became active I suspect it would switch.  So my suspicion is that the CDI would switch on the inbound turn to the FAC..I think this would auto switch well before reaching IPDEW.

EDIT: added pic of where I think Auto Switch CDI would occur (red X) so this is where I think that VNAV would fall off if doing the HILPT for a course reversal.

[Marc_B]

@RoundTwo  The Garmin Aviation Trainer is MUCH better and allows you to use more than just the GTN!

https://www8.garmin.com/support/download_details.jsp?id=12373

[PT20J]

26 minutes ago, Marc_B said:

But GPS is potentially delicate and certainly the scenario exists with planned or unexpected GPS outages; so I like being comfortable with Green Needles, especially since it seems to be the mode that is rife with more quirks and complications.

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