Source: 2019 Q2 Beartracks, Jared Yates
As any home airplane builder can tell you, sometimes we win, sometimes we lose. Sometimes we make a part once, and it works out great. Sometimes we make the same part 5 times and still aren’t happy with it. Here’s my story of trying out an autopilot installation. I place it in the latter category. It’s a failure in the sense that I didn’t accomplish what I set out to do in the beginning, but it is a success in that I have more data. Sometimes we don’t know that a road is a dead end without driving a few miles.
It all started with the ADS-B mandate. We were happy with the 100-series Dynon products: a D180 on the left, and a D100 on the right. Information could be shared between the two, and the systems were always very robust. As the ADS-B deadline loomed, we didn’t have a firm compliance plan, until Dynon announced that they would be willing to buy back the older units if we purchased their newer Skyview system. This became a very attractive option, partly because their offer price for the buyback was more than I had paid for any of the units in the first place. Also, the Skyview does have some cool extra features, including autopilot integration, and even automatic control of electric trim. Other features are added periodically, because this is still their flagship system, unlike the stagnant D-series. The job went well, spanning the last two months of 2016. We flew with and liked the new system before installing the autopilot, and in the following months, installed the aileron and elevator servos. At the time, these were the only two servos the system supported. There are several ways to install servos in the Bearhawk, but I decided on the under-seat area, as inspired by Bob Triplet. The pitch servo went under the front two seats, with the arm in-line with the elevator cables. The aileron servo went under the front right seat, clamped to the front of the big truss at station C-S. Here are some photos of the installation.
The first photo shows the pitch servo, with the stick in the full-aft position.

This is looking in from the right front door, with the seat removed. The bracket is .050” aluminum, with two –12 unpadded Adel clamps around the tube at the top, and a length of 1/16” extruded aluminum angle at the bottom. The angle is attached to the floorboard.
The next photo is in the same position, but instead of looking across the cabin, it is looking forward.

The roll bracket is also .050” aluminum, but it is supported by unpadded Adel clamps all around. Three –16 across the top, one –16 on the longer diagonal tube/bottom and one –12 on the nearly vertical, outboard tube. I had to lengthen the bolts where the right lower aileron cable joined the stick, and where the upper elevator cable joined its bellcrank. These longer bolts gave a place to secure the rod end bearings for the pushrods. The aluminum tube for the pushrods is 3/8” diameter, .083” wall 6061-T6. One foot was sufficient for making both. The rod ends are McMaster Carr part number 60645K812, with 1/4×28 threads and a 3/16” ball ID. Four are required, along with plain jam nuts.
A few more pictures of the installation:












All of this worked well enough, mechanically speaking. I was able to calibrate the servos and do some test flying. The pitch control was fine, but it became immediately obvious that the roll control was not going to be able to steer the airplane by using ailerons alone. To be fair, there is not a consensus on this point in the Bearhawk community. Some builders adjust aileron rigging to reduce adverse yaw, and some help the system by using their feet when required. But to have the autopilot capable of maneuvering the airplane, at least in our case, aileron alone wasn’t going to work. For those who haven’t flown an airplane with adverse yaw, here is how it goes. If we are in level flight, feet on the floor, and move the control stick to the left, the airplane will bank to the left, with a slight movement of the nose to the right. If we increase the left aileron, the nose goes further right. If the autopilot attempts to turn the airplane left and sees the nose go right, it puts in more left aileron, making the nose go further right, until the servo starts to slip. The airplane just flies along uncoordinated, not getting any closer to the desired heading or bank. We were at a decision point, about to give up. A good price had come along on a third servo, and I had purchased it based on rumors that Dynon was possibly expanding to support a yaw axis. I called them to ask how that was going, and if they needed any beta testers. The support guys said the third axis feature was about to be released very soon, and that I should hold out if I could, so I did.
Meanwhile, I worked out a mechanical installation for the yaw axis. This one is not as easy as pitch and roll, because the system is not a closed loop. In the aileron or elevator system, you can find any point in the control cable, anywhere in the airplane, and make inputs to that single point. If you pull the cable, the controls go one way. If you push the cable, the controls go the other way. However, the rudder system is like the flaps. If you pull a cable, depending on where you are and which way you pull, it may deflect the surface, or it may just make slack in the cable. The servo, with its single point of contact to the cable, needs to have a closed loop. My first thought was to use a two-ended control horn, something what Garmin designed for their RV-10 yaw damper. Dynon doesn’t sell such a horn, though it seemed easy enough to bolt a piece of aluminum on top of a regular one-sided horn. Before I could get to that, a better plan came along. I would close the rudder cable loop by using a capstan-style servo and two pulleys. The servo’s axis of rotation would run longitudinally, and it would be mounted aft of the cabin area. The capstan servo is the same as the arm servos, but instead of an arm, it has a little cable drum fitted with 1/16” diameter cable. The ends of those cables are clamped alongside the primary rudder cables, using nylon blocks. The cables pass through pulleys, which are attached via a thin piece of 4130. I welded bushings onto the ends of the 4130 strap so that the pulley bolt could attach. The servo has a bracket made from .050 aluminum, which is clamped with unpadded adel clamps to the fuselage tubes. There is a single clamp, which allows the assembly to pivot a little vertically, finding its best angle on its own.

If you think this would be a pretty difficult installation post-covering, you are right. It’s miserable in fact. Making things worse, I had what turned out to be a bad wire connection near the servo, which led to me having to take the servo out and put it back in several times. Sometimes it would calibrate and test correctly, and sometimes not. The culprit was a bad d-sub pin, but once it was fixed, the servo started working reliably.
Then came the flight testing. While Dynon had been so nice to release the software update that would enable the yaw servo, they didn’t release much of any documentation about what the settings would do. As months passed, I would talk with their sales folks at the airshows. I learned that the brains behind their autopilot program was Lawrence Doan, and that he was the one who would be able to help me get it all sorted out. I felt a little like Dorothy in the Wizard of Oz. Everyone told me about the great wizard, but nobody could put me in touch with him. I wrote him directly and never heard back. I’m sure there are lots of builders who wish to talk with the wizard, and he’s probably very busy. The reality became “nobody gets to see the wizard, not nobody, not no-how.” I’ll skip describing a year or two of back and forth with various Dynon folks who were happy to help at first, but never came through to deliver as promised. They have been understaffed/overwhelmed with their Type-Certificated program, among other things. And who can blame them for not wanting to spend a lot of time fiddling with a single installation when hundreds of builders are backed up out the door to buy well-vetted systems for their RVs.
In the absence of documentation about what the settings were supposed to do, I spent some time in the air just turning knobs and fiddling with them. I was able to get the system working well enough to sort-of fly the plane, but I never could get the rudder inputs good enough to pass the “passengers don’t get nauseous” test. I suspect that the yaw algorithms were designed more as yaw dampers for the RV-10 crowd than as true three-axis systems for a plane that requires foot-skill to fly. While I’m positive that a system could be created that would fly the Bearhawk, I don’t think Dynon has created it yet, and they don’t seem to be motivated to do so any time soon.
Contrary to my experience, there are several happy Bearhawk owners using Dynon and other autopilot systems. Some only use them for straight and level flight. Some supplement with foot input. Most are probably not as picky as I am about how well the autopilot flies or doesn’t fly the airplane. I’m totally spoiled, with over 10,000 hours in airplanes that have extremely robust autopilot systems, developed with industrial-grade R&D budgets. Some with auto-land capability. It isn’t fair for me to expect the same from an E/A-B grade system. It’s a case of “the princess and the pea.” But after a few years of trying, all the while carrying around 15 pounds of weight, I finally decided to give up and remove the system. If we are to succeed in building airplanes, we must have gumption and persistence, but there are also times when we have to pause, disregard sunk cost and completion bias, and take another route. This is part of the process, and part of what makes the hobby a life-long challenge. Education and Recreation. Winning some and losing some. It’s all part of building an airplane.

Source: 2016 Q3 Beartracks Russ Erb
Do I Need ADS-B?
A lot has been written lately about ADS-B. It’s all over the aviation press, but do you care? If you currently fly an airplane without a transponder, you probably don’t. You may skip to the next article. It’s okay. I won’t mind.
The FAA has mandated that as of 1 January 2020, all aircraft flying in airspace that currently requires a transponder will be required to be equipped with ADS-B Out. In case you’ve forgotten, that’s basically within 30 nautical miles of a Class B airport, inside Class C airspace, or anywhere above 10,000 feet MSL except below 2500 AGL. Only ADS-B Out will be required. ADS-B In is completely optional, but won’t have full functionality without ADS-B Out. In short, the ground towers will not transmit traffic data to you unless the tower is pinged by your ADS-B Out. Otherwise it won’t know you are there. Learn more at one of the many online resources, such as http://www.freeflightsystems.com/ads-b-university or the Air Safety Institute online course “ADS B for General Aviation” on the AOPA website.
Why Did I Choose To Equip Now?
Well, it certainly wasn’t because of the FAA’s recently announced rebate program. That wasn’t announced until after my installation was complete.
A minor reason for installation was to save money. I was paying a monthly subscription to XM to receive weather in the cockpit. Living in the desert of Southern California, this was very cost-ineffective, as about the only time there was any weather to display was on the trip to Oshkosh. ADS-B In provides free in-cockpit weather (FIS-B), but only when airborne. Much like VORs, the signal from the transmitting towers is blocked by ground clutter unless you are close to the tower. That was okay, because on the ground Foreflight can receive weather as long as it has a cell phone signal.
My primary reason for installing ADS-B now is that I am afraid of mid-air collisions with traffic I never saw. I have good reason to be concerned about this. Since I started flying in 1989, my success rate in seeing other aircraft in flight, whether identified by someone else on board or called out by ATC, is well below 50 percent. I have flown all the way from Southern California to Oshkosh and back and never seen another airplane outside of the traffic pattern. I know I wasn’t the only one flying at the time. One year flying into Copperstate I saw an airplane through the right side window flying away from me—after passing uncomfortably close right below me. I never saw him coming and I bet he never saw me. I decided that it was worth the money to get better traffic info to lower my stress level when flying. ADS-B In was relatively cheap by comparison, but by itself was not much better than nothing. It would show you the traffic that had already equipped with ADS-B Out, but not the thousands others that hadn’t. For full functionality of ADS-B In, I had to have ADS-B Out to ping the towers to show the Mode C only traffic. It’s still not a perfect solution, but it is the best available.
The Easy Way
If you haven’t started installing your avionics and you haven’t built any antenna mounts on your fuselage yet, you have it easy. You can pick any ADS-B solution available and modify your aircraft as needed to install the equipment. Easy peasy.
The Hard Way
In my case, I had to do it the hard way. I had to modify an existing installation in a flyable airplane. This was not as simple as sliding out one avionics box and sliding in another. There are some ADS-B Out solutions that are intended to be simple add-ons to work with your existing transponder. These are typically 978 UAT boxes, and, while I didn’t check, I assumed they would need their own antennas added to the airplane. On a sheet metal fuselage, that can be as simple as drilling a new hole in the skin and adding the antenna. However, the Bearhawk is a fabric covered fuselage. You can’t mount an antenna on just fabric. All of the antennas currently on my fuselage have their own special antenna mount built up from the fuselage under the fabric. For me, any system requiring an additional antenna was immediately rejected.
Prior to the ADS-B install, my primary panel mounted navigator was a Garmin GNS-480, the rebranded version of the UPSAT CNX-80. This was a WAAS capable GPS, but it needed a software update to allow an ADS-B transmitter to recognize it as a valid position source. Garmin had claimed to be working on an update for years, but it was a low priority. This update was finally released in 2015, and its release was part of my decision to do the upgrade now. The upgrade was easy, taking about 15 minutes at the avionics shop with the box still in the airplane.
My existing transponder was a Garmin GTX-327. It was in perfect working order, but was not upgradeable to serve as an ADS-B transmitter. As for somewhere that would display ADS-B In data, I had two choices. I had a Garmin Aera 510 mounted to the panel that had previously been used to display XM weather. I also had an iPad mini mounted to the panel running Foreflight.
ADS-B Out Selection
My intent was to pick an upgrade path that worked with as much of my existing equipment as possible. For ADS-B Out, the Garmin GNS-480 received the required software update. For the transmitter, I decided to replace the Garmin GTX-327 transponder with a Garmin GTX-330ES. There were other transponders available, and some were cheaper. I decided to go with the GTX-330ES because the front panel was identical to the GTX-327 that I had, which would minimize any relearning to use it. While the front panel looked identical, it was not a simple swap. The tray for the GTX-330ES was about 4 inches longer than the tray for the GTX-327. Thus, the trays had to be swapped, which required some modifications behind the panel. The wiring for the transponders was not identical, so I had to get Approach Fast Stack to make me a new cable from the hub to the transponder. The deal was sweetened, as Approach Fast Stack gave me a nice trade-in credit on the GTX-327, as they have a healthy demand for GTX-327 transponders in some other common installations.
ADS-B In Selection
Again, the easiest solution was to use as much of my existing equipment as possible. I could get the Garmin GDL-39 which would display on the Aera 510. I could also get the Stratus 2S which would display on the iPad mini in Foreflight. I chose the latter because the iPad screen was larger and had a better resolution. The Stratus 2S is a dual-band receiver, so it receives air-to-air ADS-B signals directly on both 1090 and 978 Mhz. It will also receive traffic (TIS-B) on either frequency and weather (FIS-B) on 978 MHz. I avoid depending on battery power when I can (I’ve had range anxiety since childhood), so I mounted the Stratus 2S under the panel and wired it into ship’s power, using the same cigarette lighter socket and dual USB charger that powers the iPad. The Stratus is set up to turn on when power is applied and turns itself off within 2 minutes of losing external power if the sensed ground speed is zero. Because the Stratus is under the panel, its internal antennas may be blocked. I placed a remote GPS antenna on the glareshield under the windshield where the XM weather antenna had been. I mounted a remote ADS-B antenna on a mount on the left interior wall by the rear seat. The mount is attached to the stiffener tube that comes out of the left sidewall. This may seem like an odd place to place an antenna in a steel fuselage cage, but it was the largest opening I could find that was accessible. It must work, as I have picked up ADS-B towers as far as 100 nautical miles away.
Testing
Since I had changed out the transponder, my transponder certification (14CFR 91.413) was obviously invalidated. I took the airplane to our local avionics shop, and while certifying the transponder he also confirmed that the ADS-B Out signal was transmitting properly. He also fixed some of the settings that I had wrong due to ignorance.
After having the system checked out, I flew the airplane with the system turned on. After the flight, I went to http://www.faa.gov/nextgen/programs/adsb/ and sent the FAA an e-mail with the requested in-formation. They went into their records and very quickly sent back a test report showing that my system was working properly.
Sure enough, it turns out there are airplanes out there that I never see. At least now I have a better chance of finding them, and even if I don’t, I can use the display to be sure I avoid the traffic.
Even so, it’s still important to look out the window. I saw a Cessna pass in front of me a mile or less away and it never showed up on the traffic display. I suspect he had no transponder or it wasn’t on. Those airplanes won’t show up on any traffic system other than your eyeball.
If my panel looks different than yours, that’s because it is. I increased the height of the panel by 2.4 inches to get all of the stuff in. I also increased the height of the firewall to keep a straight line from the nose bowl to the spinner. That also means my windshield is cut different, and on, and on, and on…

Source: 2015 Q4 Beartracks
Builders Eric and Brent of Charlotte NC send this photo of an engraved plate they made to decorate their Bearhawk project. Most of their initial fabrication work is complete, and now they’ll disassemble the airframe and prepare for covering, paint, and final assembly. Readers may recall the article about their all-metal Bearhawk from earlier this year.

Source: 2012 Q1 Beartracks, Dave Lenart, Bearhawk N942VT
I had to make a decision about a new GPS for Victor Tango, since the one I installed during the build had become obsolete. I had originally installed a Lowrance 2000C in my panel and had it perfectly mounted next to my Dynon EFIS, nice job if I do say so myself. I really like the Lowrance and have used it with great success; even on our trip to OSH last year the little unit did a wonderful job of keeping Laura and I on the red line drawn on the chart. Very shortly after I purchased the unit Lowrance made the corporate decision not to support their line of aviation GPS units any longer. What this means for those of us with the unit is that there are no updates available and as airspace and airports change it will gradually become less and less accurate. So the search was on for an upgrade, but which one? There are many choices out there and opinions differ from the tried and true Garmin units to the latest in IPad applications. I almost went with the IPad along with the WingX program, I know some pilot friends that like the setup. But I thought it might be actually too large and cumbersome in the Bearhawk, especially being a stick airplane. I’m sure there are some good ways to use or mount the IPad, but it wasn’t for me. The Garmin units are always good but I’ve never liked the small screen size of the 496 style unit and the six and seven series units are a big chunk of cash, although again I know they have their faithful for good reasons. Last summer at OSH I wandered by one booth in the “B” building that had a huge crowd of people around it every time I tried to get close, it was the Adventure Pilot folks that were demonstrating their IFly 700 GPS. It’s based on a large touch screen with GPS information overlaid on a sectional chart. The screen itself is 6” wide by 3 ¾” tall, the unit is about an inch bigger than that overall. Since last summer they have upgraded to the IFly 720 which adds to the original a sunlight readable screen, wireless update capability and road navigation function so you can use it in the truck on the way home. Now, I know everyone has their favorites and so be it, I am also not advertising for the Adventure Pilot company in any way but just giving my opinion on this GPS with the hope that it may help someone else who is gathering information prior to their purchase, so don’t get all offended anyone but I really like it. I say this a little surprised myself because for the first couple of flights I hated it! Well, that’s a little strong, but I was so used to my old simple screen that the sectional overlay seemed like just too much information. That, and the inevitable learning curve of any new device with its differences, some intuitive some not. But overall I am very pleased with this GPS unit. Why?
1/ The screen size is excellent. There’s plenty of real estate to look at and the zoom function makes it so that you can focus in on a small area with great detail. I was at first disappointed with the fact that the unit is only oriented in landscape without the option to rotate to portrait but the company says there will be a downloadable upgrade soon to accommodate that. For my panel it will work better vertical and I’ll mount it flat on the plate I made when the download is accomplished. Within the system you can select an optional screen that gets rid of the sectional and gives you a greatly simplified screen if so desired.
2/ Charts. The unit came pre programmed with all U.S. charts and the upgrades are easily made wirelessly by connecting at a wireless zone. As you fly along the next chart automatically scrolls under the plane icon. This will be great for long cross countries or if you live on the edge of a chart somewhere.
3/ Street navigation. This function has been nice to have, so much so that I cancelled my OnStar subscription in the GMC. I don’t take the GPS out of the plane after each flight, but if we’re going on a road trip or I’m going to an unfamiliar job it comes out easily and mounts in the truck via a Ram suction mount.
4/ Price. Here’s the kicker. The unit cost $749. I call that a deal considering its capability and the fact that it can be connected to an ADSB receiver for weather and traffic down the road.
The picture shows the mount I made using parts of a Ram mount and a plate to cover the existing hole for the Lowrance, it works good for now but after the portrait download is made I’ll be able to mount it flat on the panel again. There is no external antenna and I’ve not lost signal so far within 100 km of my home airport at Lebanon NH. There are lots of neat features like full airport information available by holding your finger on the airport in question for a second or two but you can find out more on their web site if you wish. Laura and I will be heading over to the Maine coast soon for some ‘lobsta’ and I feel better about having an updated GPS.