Ok, so I started going through some things this evening.
1. My GPS software is 5.3, which Garmin indicates is the latest version
2. My Trio has the latest, confirmed by Chuck at Trio (more on this below)
3. There is zero slop in the elevator connection and none in the connection to the servo.
4. I didn't make it to opening the servo gold box, because I called Trio to get their thoughts on the software mismatch possibility.
Chuck at Trio (who has been super helpful) started going through my problems with me, and we talked about the software for a minute. He said there would be no mismatch in my software between the GPS and the Trio, apparently because my Trio is older and doesn't receive any altitude select etc commands from any type of EFIS, and relies on it's own pitot/static connections to do any type of EFIS calculations.
We then discussed VNAV servo gains, and a couple other things, but these seemed normal, or he said the defaults, especially for an RV-6, should be pretty close already (an RV-6 was used a test bed during the development of the Trio).
He then wanted me to blow or put some air past the static ports to verify that the Trio saw a climb or descent. When I told him it was going to be difficult to get under the wing to the pitot/static mast, because I have a Piper style pitot/static mast, he shut the whole thing down
He said that type of static port will never work correctly with the autopilot, because the Trio is able to detect changes of altitude to within 4 inches, and any change of the wing incidence shows up as a huge pressure change to the autopilot when the static port is located under the wing like that.
He was 99.9% certain this was my problem, and suggested I disconnect the static line to the Trio and use cockpit static pressure to test with, and in addition, since he was pretty confident that was the issue, to install the standard static ports in the rear fuselage location, in clean air.
So, I spent this evening installing the standard static ports in the fuselage, a very simple 1 hour job that so far has taken 5
Oh, well, once again too windy to fly.
So
BobTurner, looks like you were probably on the right track with the static line idea of using cabin air. The reason I didn't bother to test that first was:
1. too windy to fly, so I might as well tear something up instead
2. Made a lot of sense that that would cause that type of issue
As for the connections, when I put my GPS in config mode and checked the setting, it gives the flag to the CDI, and the autopilot comes alive like it's tracking a course, and the blue lights come on, and the display shows it's getting "GARMN" data, so I think that is all good to go there. It has also always correctly shown destination WPT and groundspeed, and the XT error from the GPS, so I think the serial is good. Also, previously, when I received it back from Trio after the upgrade, Chuck had walked me through the correct settings on the GPS for ARINC, which I confirmed are still set correctly.
So, Chuck suggested I report here about the sensitivity of the static location, and of course that the Trio was really developed and tested against a standard Van's setup for that, obviously they can't account for all the variations of what people do in practice, like my Piper pitot/static mast. He said obviously that may not explain everyones issues with the Trio, but he would definitely rank it high on the list.
I don't know if you checked into this static issue on your quest
FORANE to fix yours in the Lancair and your 9A, but might worth a discussion with Trio and describing exactly how you have your static ports mounted. Maybe you could run a dedicated static port for just the autopilot, or find a way to test it similar to the cabin air idea? Chuck said the main thing is any changes in cabin pressure (opening a vent, turning on heat, etc) would obviously affect it, but would be at least be a test.
I will test all this in flight as soon as the WX clears up again with this horrible wind we have been having all spring.