About the time I installed the new TT VSGV (connected to my GRT EFIS) in my airplane late last fall, our local convenient ILS went down as they decided to tear up and replace the runway. Consequently, when I did the testing on my new system, I was doing it primarily with the Synthetic Approach capability in the GRT EFIS, which worked great. I haven?t had much of a need to shoot precision approaches since then, and it was only recently that the ILS came back and we started playing with it again. The latest GRT Software release gave me a push to go and do the coupled approach thing again, with the intent of tweaking the gains to get the airplane flying approaches really nice.
One of the little problems we?ve had with the previous software was a tendency for the RV-8 to porpoise a bit in pitch, particularly with an aft CG (that means, with a passenger). This happened mostly in climb, and with a little bit of turbulence upsets. I had been flying with some modified software for a few months that gave me the ability to change far more inside the guts of the algorithm than the average user should ever see, and we had sorted things out pretty well ? so I was excited to see the release version and see how it did. Well, I must congratulate Greg and the rest of the guys at GRT ? they really got it right this time! No pitch hunting with any CG that I have flown ? front to back of the envelope.
Since pitch got sorted out so well, I flew down to KLBX to try the ILS?and boy, did I get surprised. With the factory gain, I let the coupled system try and find the localizer ? and we went for a lazy tour of the local rice fields. It would cross the localizer at a 30 degree angle, then correct back slowly, crossing it at the same angle in the opposite direction. By the time we got to the runway, there was no way we were going to land. Ugly is the word. I went back around, and loaded a Synthetic Approach. Intercepted the approach, and it was rock solid all the way down! OK, so maybe I needed to tweak the Localizer gains a bit ? my instincts said ? make it more sensitive, so I changed it from 1.5 to 2.5. Around to the Outer Marker I went ? coupled it up and?.it was still abysmal. OK, how about upping the gain some more? Well that sent me off on several circuits (glad no one else was trying to use the ILS that day!), and I got the gain up to 10.0 before giving up.
I dropped a note to GRT, and very quickly got a note back from Greg, explaining in great detail what he thought was happening (fortunately, with my background, I am used to understanding guidance, navigation and control geek-speak from folks that really know what they are talking about), and the suggestion that I LOWER the gain. Yesterday morning, after flying an errand, I found myself at the Outer Marker again, set the gain to 1.0, coupled it up and?.it was almost perfect! I dropped it to 0.9, went around one more time, and declared victory. The airplane now flies itself nicely down the center of the ILS, and I can display the Synthetic Approach HiTS along with the ILS scales to confirm that what I am seeing is valid. I like comparing two different navigation solutions against each other when the weather is low.
This might help other RV-8 guys with a similar setup ? if you need to get the ILS tracking nice, don?t be afraid to vary the gains down as well as up. My testing technique was to fly about a minute on the ILS from the OM inbound ? plenty of time to see if it is going to be stable or not ? then break out and fly back for another go while changing the gain before the next attempt. Flight testing isn?t all Chuck Yeager and the Sound Barrier kind of stuff ? sometimes it is just flying ovals in the sky, adjusting things a little at a time?.
Paul
One of the little problems we?ve had with the previous software was a tendency for the RV-8 to porpoise a bit in pitch, particularly with an aft CG (that means, with a passenger). This happened mostly in climb, and with a little bit of turbulence upsets. I had been flying with some modified software for a few months that gave me the ability to change far more inside the guts of the algorithm than the average user should ever see, and we had sorted things out pretty well ? so I was excited to see the release version and see how it did. Well, I must congratulate Greg and the rest of the guys at GRT ? they really got it right this time! No pitch hunting with any CG that I have flown ? front to back of the envelope.
Since pitch got sorted out so well, I flew down to KLBX to try the ILS?and boy, did I get surprised. With the factory gain, I let the coupled system try and find the localizer ? and we went for a lazy tour of the local rice fields. It would cross the localizer at a 30 degree angle, then correct back slowly, crossing it at the same angle in the opposite direction. By the time we got to the runway, there was no way we were going to land. Ugly is the word. I went back around, and loaded a Synthetic Approach. Intercepted the approach, and it was rock solid all the way down! OK, so maybe I needed to tweak the Localizer gains a bit ? my instincts said ? make it more sensitive, so I changed it from 1.5 to 2.5. Around to the Outer Marker I went ? coupled it up and?.it was still abysmal. OK, how about upping the gain some more? Well that sent me off on several circuits (glad no one else was trying to use the ILS that day!), and I got the gain up to 10.0 before giving up.
I dropped a note to GRT, and very quickly got a note back from Greg, explaining in great detail what he thought was happening (fortunately, with my background, I am used to understanding guidance, navigation and control geek-speak from folks that really know what they are talking about), and the suggestion that I LOWER the gain. Yesterday morning, after flying an errand, I found myself at the Outer Marker again, set the gain to 1.0, coupled it up and?.it was almost perfect! I dropped it to 0.9, went around one more time, and declared victory. The airplane now flies itself nicely down the center of the ILS, and I can display the Synthetic Approach HiTS along with the ILS scales to confirm that what I am seeing is valid. I like comparing two different navigation solutions against each other when the weather is low.
This might help other RV-8 guys with a similar setup ? if you need to get the ILS tracking nice, don?t be afraid to vary the gains down as well as up. My testing technique was to fly about a minute on the ILS from the OM inbound ? plenty of time to see if it is going to be stable or not ? then break out and fly back for another go while changing the gain before the next attempt. Flight testing isn?t all Chuck Yeager and the Sound Barrier kind of stuff ? sometimes it is just flying ovals in the sky, adjusting things a little at a time?.
Paul
