I dropped by the hangar last night after our EAA meeting to try and further diagnose why the G5 wouldn’t power up. I called Stein on Saturday and he suggested pulling the backup battery, but I decided to just try powering it up again without changing anything. Low and behold, the unit powered up just fine! The backup battery was dead, so apparently it was keeping the unit in an odd state. A quick call back to SteinAir this morning confirmed that this is a known issue and Garmin has a fix on the way (if not already available).
I received my new Garmin G5, so it’s time to pull out the old TruTrak Gemini.
The G5 installation kit includes a clever mounting ring that attaches behind the hole using the standard instrument mounting holes and has a hole for an alignment pin at the top and a threaded hole at the bottom.
With the mounting ring installed, installing the G5 just takes a moment using a 3/32″ Allen key through the lower hole to secure the unit.
Fortunately, I had enough slack in the pitot and static tubing to connect to the new fittings. Even the wiring was pretty trivial as the d-sub connector that went to the Gemini had only four wires (power, ground, RS-232 in, and a dimmer connection). For my installation, I only needed the first three of these since the G5 has a photocell and provides automatic dimming. I did need to rearrange the other three wires and change the GTN-635 serial format, but that was all straightforward.
After doing the ground config and a vibration test, I shut down the plane to get ready for a quick test flight. Unfortunately, when I went to start the plane up again, the G5 wouldn’t power up. After diagnosing all of the connections and finding nothing wrong, I left for the day.
I’ve never been very happy with the TruTrak Gemini PFD. The AHRS is iffy and it would often show in incorrect attitude.
After seeing the Garmin G5 at Oshkosh and hearing how happy people were with them, I decided it was a good time to replace it with something better. I was pretty sure the G5 would fit, but I was concerned about it covering up the labels for the two push/pull cables below and whether the backup battery would interfere with the angled brace behind the panel. Fortunately, my buddy Greg had a G5 he hadn’t installed yet and he let me borrow it to confirm it would fit. Here I’m just holding the unit in the hole from the rear. You can see that it fits flush against the instrument panel, but does cover the labels below.
Though it’s tight, it does clear everything behind the panel!
From above, you can see that there’s nearly 1/4″ of clearance between the backup battery and the angled support.
I wrapped up the heater valve swap by reconnecting the SCAT tubing and control cable and then reinstalled all of the interior aluminum covers and seat pans. Before reinstalling the cowl, I checked the breather vacuum valve in case I needed to clean it. I’ve been checking this every oil change and it never seems to get worse than this.
Before reinstalling the interior, I needed to swap the ADS-B module. Dynon recently contacted me about swapping this because they apparently found a hardware defect and they want to replace all of the units in the field. Although I don’t relish removing all of the interior and baggage wall to get to this, I don’t mind since I really appreciate how proactive Dynon is about resolving issues like this.
The new unit is identical from the outside, so it’s a trivial swap (unlike when I swapped from the SV-ADSB-470 to the SV-ADSB-472).
Dynon released version 12 of their software which adds auto-trim to their autopilot. To enable this, I needed to rewire the trim to go through the Dynon Autopilot Control Panel instead of through the Vertical Power VP-X. This entailed several hours of laying on my back under the panel rerouting wires.
The Autopilot Control Panel needed power and ground to run the trim servos. Fortunately, I still had a few extra power pins on the VP-X, so that was a quick addition. I then needed to reroute the four wires from the trim switch and the two pairs of wires to the pitch and roll trim motors. These eight wires all had d-sub pins or sockets, so I built a new harness that extended the wires from this point.
After finishing up the rewiring and upgrading the software, I ran the trim calibration routine and then flew Jenn to lunch in Half Moon Bay to test it out. The auto-trim worked great and kept the plane in trim the whole flight.
I stopped down at the airport with the kids and wrapped up the new avionics installation. All that was left was installing the nutplates and doing the wiring. I previously just had a single network cable connecting the two displays, but I needed to replace that with three network cables (left SkyView to knob panel, knob panel to autopilot panel and autopilot panel to right SkyView). Since I had run the previous network cable between the two screens along the sub panel, it was long enough to cut and install new connectors on to make two of the three cables I needed. Fortunately, I had enough scrap network cable in my box of wires to make the third.
After hooking everything up, I did a network config to allow the SkyView system to find the new boxes and did some ground tests to make sure everything was working properly. Since I had the kids with me, I couldn’t take it up for a test flight, but I’m sure these will be a nice improvement to the panel.
I cut the holes for the new avionics boxes and test fit them into the holes. Cutting these holes in place was a real pain. The nibbler I had wouldn’t cut 0.063″ thick sheet, and I didn’t have anything else that would work well. I ended up drilling a series of 1/4″ holes around the perimeter and using a file to open the hole up enough for the avionics to fit. It was quite a bit of iteration and it ended up taking me nearly 2 hours to get them cut. That was still way faster than pulling the panel out of the plane though. With everything installed and aligned, I drilled the mounting holes through the holes in the face of the avionics.
Afterward, I pulled the new boxes and the GTN, then used a scrap miniature #6 nutplate to drill holes for the rivets that will hold the nutplates in place. I still need to countersink the holes and rivet them in place, but I’m beat tonight. It shouldn’t take more than a couple more hours to finish up the nutplates and then fabricate a few new network cables to tie the boxes together. The autopilot control box (on the right in the picture above) also includes a trim controller that essentially does the same function as the VP-X that controls the trim now. In the future though, Dynon plans to add auto-trim to their autopilot, so I’ll eventually rewire the trim to go through this box.
I put the plane back up on jacks and slowly enlarged the mounting holes up to #19 for a #8 screw while adjusting the wheel pants to be perfectly aligned with the longitudinal axis of the plane. They were off a little bit when I started which is probably due to the epoxy/flox shims slightly shifting the position of the pants from when I first drilled them. Since my brackets are stainless steel, I needed to use a fair amount of pressure to drill these, so I backed them up with some wood to prevent them from flexing in as I drilled.
After drilling these out, I pulled the wheel pants and gear leg fairings off the plane and to the plane back off the jacks so it’s ready for flight again.
Dynon recently released two new boxes that work with the SkyView system: one to provide dedicated knobs to control the altitude bug, heading/track bug and barometric pressure and another to provide a set of dedicated autopilot controls so that you don’t have to navigate through various menus to control it. I’m mounting these in my radio stack with the outer edges flush with the other boxes in the stack. This leaves enough room in the center for another small device if I want to add one. I laid out for the holes, but it was too late to start cutting into the panel. Once I have these mounted, hooking them up to the SkyView network will be pretty straightforward since I can just daisy chain them between the displays.
The inclinometer on my SkyView has always shown nearly 1/2 ball out to the left, even when the fuselage was absolutely level. After doing some research, it looked like the only way to adjust this was to shim the ADAHRS until in showed a centered ball with the fuselage level. I used my shop crane to raise the tail to a level attitude and put my aircraft jacks under the jacking points on the wings to hold the plane rigid in the roll axis and to make minute adjustments to the lateral level of the plane.
After getting the longerons level, I used the wing jacks to adjust the lateral angle until my digital level showed 0.0 when placed across the seat back support. I then zeroed the smaller digital level I have and placed it on the ADAHRS. Sure enough, it showed about 0.5º angle with the left edge low (exactly as a left ball would indicate).
I removed the attachment screws holding down the ADAHRS and fabricated some small aluminum shims to fit under the left edge. After some experimentation, I found that I needed a 0.048″ thick shim to center the ball. Now the ball is nearly perfectly centered. I’m really surprised how sensitive this is. Varying the shim thickness by just a few thousandths made a visible difference in the position of the ball. Given that my digital angle is only accurate to 0.1º, it’s really hard to get this accurate since I can now move the ball from one vertical mark to the other while still reading 0.0º on my digital level.
While I had the plane in the level attitude, I was curious how much ground clearance I have from the propeller tip. This is the best case without any further gear compression. If I were to bounce a wheel landing or raise the tail too far, this clearance could diminish significantly.
Greg also stopped by with his Windows laptop so we could make a few adjustments to my VP-X. My Windows laptop is acting up (surprise, surprise), so I haven’t been able to use it to connect. We configured the Wig-Wag function, set up intermediate stops on the flaps (and configured the VP-X to use them instead of requiring me to hold the flap switch until they reach the position I want), and turned off current monitoring on the fuel pump. I still need to adjust the flap airspeeds, but I forgot at the time.