I have a problem with my advanced autopilot on ILS approaches. Bringing up the ILS in the 430, the HSI shows proper green data, and localizer and glideslope indications are right on. Today I was on a 30 degree course to intercept from the right, turned on HSI and VNAV and AP. The autopilot tried to steer us more to the right, and it appeared that it would keep on turning. Punched off AP and hand flew, watching the command bars, they were commanding a climb when I was above the glideslope. I'm pretty sure I had it set up properly but need to know; if there ARE perfect indications on the HSI, is it possible for the AP to do what I have described? Or is it driven solely from the HSI data? I tried this before on the way to OSH and saw the same issue. Both times VFR, and wont try IFR until I have lots of confidence! FYI GPS approaches work fine with AP, it's just the ILS.
Autopilot on ILS - Dynon?
- Thread starter hmanvel
- Start date
Ken_Kopp
New Member
curious how far from the LOC centerline you were when first engaging the AAP? If you are outside the normal LOC it is possible to receive sidelobes which the system will see as valide LOC signals but they are erroneous and can cause interesting behavior. This is not unique to Dynon but an artifact of the LOC antenna itself.
I have over 100 ILS approaches completed with the Dynon AAP (since development began) and have only seen this behavior when locked onto a sidelobe of the LOC. The way to ensure this doesn't happen is to be sure when the LOC needle is active that it is steady and off the case before arming HSI on the AAP menu.
I have over 100 ILS approaches completed with the Dynon AAP (since development began) and have only seen this behavior when locked onto a sidelobe of the LOC. The way to ensure this doesn't happen is to be sure when the LOC needle is active that it is steady and off the case before arming HSI on the AAP menu.
I am not sure if you can do that with the DYNON system. On some systems you can select one source for the HSI and another source for the autopilot. An autopilot does exactly what you tell it, not what you want it to do. Remember the old term "garbage in, garbage out".
Like other have said this situation can also happen if you are not on the actual localizer but on a sidelobe no matter which brand of autopilot you use. Again, "garbage in, garbage out".
Another possibility is the airplane is way off in aileron trim, the roll servo disengages and the aircraft wanders off.
meljordan
Active Member
Having flown well over 100 ILS approaches while helping to developing the Dynon AP approach capability, I can confirm that it does indeed fly the approach very well as long as it is set up and the correct proceedures followed. I fly an RV6A with a Garmin 430W. Here are the steps I follow;
1. Load and activate the approach in the 430.
2. Switch and double check the LOC frequency in the 430
3. Check the 430 is in LOC mode
4. Using the CRS knob on the Dynon set the HSI course to the published final approach course
5. Establish the appropriate flight path and altitude using the Dynon HDG and ALT HLD AP functions to intercept the LOC and GS
6. Monitor the HSI to insure I have a steady lock on both the LOC and GS before arming either the HSI or VNAV AP functions, that the HSI shows an appropriate lateral needle deflection and that I am below the GS
7. If all the above is correct and before either the LOC or GS needles on the HSI begin to move, "arm" the HSI and VNAV AP buttons. Check that both buttons display a gray background.
8. Monitor the HSI and the AP modes. As soon as the HSI LOC needle begins to move you should see the AP automatically switch from HDG to HSI mode (green background) and you should get "autopilot mode" annunciation. Likewise, as soon as the HSI GS needle begins to descend, you should see the AP switch from ALT HLD to VNAV mode with another "autopilot mode" annunciation.
9. Throughout the approach careful and smooth power adjustments really help the AP attain best possible GS performance. I have established a series of power settings for each phase of the approach that I use as guidelines to help me get the airspeeds I desire.
I suggest you try a few more practice approaches and make sure you follow all the above steps. I can tell you all the above caused me to fail approaches at least once during my testing, some more times than I would like to admit.
Good luck,
Mel Jordan
1. Load and activate the approach in the 430.
2. Switch and double check the LOC frequency in the 430
3. Check the 430 is in LOC mode
4. Using the CRS knob on the Dynon set the HSI course to the published final approach course
5. Establish the appropriate flight path and altitude using the Dynon HDG and ALT HLD AP functions to intercept the LOC and GS
6. Monitor the HSI to insure I have a steady lock on both the LOC and GS before arming either the HSI or VNAV AP functions, that the HSI shows an appropriate lateral needle deflection and that I am below the GS
7. If all the above is correct and before either the LOC or GS needles on the HSI begin to move, "arm" the HSI and VNAV AP buttons. Check that both buttons display a gray background.
8. Monitor the HSI and the AP modes. As soon as the HSI LOC needle begins to move you should see the AP automatically switch from HDG to HSI mode (green background) and you should get "autopilot mode" annunciation. Likewise, as soon as the HSI GS needle begins to descend, you should see the AP switch from ALT HLD to VNAV mode with another "autopilot mode" annunciation.
9. Throughout the approach careful and smooth power adjustments really help the AP attain best possible GS performance. I have established a series of power settings for each phase of the approach that I use as guidelines to help me get the airspeeds I desire.
I suggest you try a few more practice approaches and make sure you follow all the above steps. I can tell you all the above caused me to fail approaches at least once during my testing, some more times than I would like to admit.
Good luck,
Mel Jordan
And to stress a couple of things that people sometimes miss:
-VNAV will only engage when the AP holding altitude below the glideslope. It won't sequence from ALT HOLD > VNAV if you're higher than the glideslope and try to dive onto it.
-The HSI course, when tuned to an ILS, needs to be manually spun to the published ILS course. If it's not set correctly, the AP can't possibly center the needle.
-VNAV will only engage when the AP holding altitude below the glideslope. It won't sequence from ALT HOLD > VNAV if you're higher than the glideslope and try to dive onto it.
-The HSI course, when tuned to an ILS, needs to be manually spun to the published ILS course. If it's not set correctly, the AP can't possibly center the needle.
" Likewise, as soon as the HSI GS needle begins to descend, you should see the AP switch from ALT HLD to VNAV mode with another "autopilot mode" annunciation."
This is not how my system nor 3 other Skyview systems that I have flown work. While the localizer does capture with needle movement, the GS never captures until the needle is centered. This has been a complaint of mine from day one---if it worked like the LOC, there wouldn't be an overshoot upon capture of the GS.
This is not how my system nor 3 other Skyview systems that I have flown work. While the localizer does capture with needle movement, the GS never captures until the needle is centered. This has been a complaint of mine from day one---if it worked like the LOC, there wouldn't be an overshoot upon capture of the GS.
Ken_Kopp
New Member
What would you have it do? if you lead the the VNAV activation with a power reduction by a few seconds the overshoot is almost imperceptible. Even if you do not the overshoot is in no way a cause for concern at most normal entry speeds. Considering a level approach to GS capture is for most ILS's a 3 deg intercept there is little need for early VNAV arming as the transition is very slight.
meljordan
Active Member
Yes, I misspoke in my message. The VNAV will engage when the GS needle centers, not when it begins to move. Sorry for that error in my posting.
I also agree with Ken, if you are managing your power settings the overshoot is very small.
Mel Jordan
I also agree with Ken, if you are managing your power settings the overshoot is very small.
Mel Jordan
I agree that the overshoot is controllable with good speed control, but the overshoot cannot be eliminated. The LOC does a nice job of intercepting without the overshoot. I would think the software could be tweaked to allow the same smooth GS intercept. Overshooting, no matter how small, is still an overshoot.