Dynon has been working on an autopilot for quite some time, but they've kept the details very close to their chest. Well, the information is finally starting to emerge, and it sounds very, very enticing. The lead-in image is an advertisement that appeared in the electronic version of the latest Kitplanes magazine.

Each servo is $750, and the interface can be via a Dynon EFIS, or via a dedicated AP74 control panel. The autopilot can be slaved to the bugs that are set on the EFIS. That means that I could add a pitch axis autopilot to my aircraft for $750, without having to sacrifice any panel real estate. The only downside is that the servos are similar to those from TruTrak, in that there is some drag on the flight controls even when the autopilot is not engaged. Those that have TruTrak autopilots claim that while the drag from the servo can be felt on the ground cannot be perceived in flight.

I'm putting any work on my pitch axis autopilot project on hold until I learn more about the Dynon autopilot. Assuming I can live with the servo drag, it is hard to go wrong for $750.

Update 28 Mar 2008 - Dynon now has autopilot information posted on their web site.

While I was in Salt Lake City two weeks ago I re-attacked the problem of using X-Plane as a simulator to help design control laws for a pitch axis autopilot, which would use electric trim commands to control the aircraft in pitch. I was stuck for awhile, but eventually got all the hooks to X-Plane working so I could read the data from the sim, display an autopilot control panel, and control the sim with the simulated autopilot. Today I did some more work, and now I've got a pitch hold mode working quite well, and am working on an altitude hold mode.

I am experimenting with some things, such as adaptive autopilot gains. The gain required will vary considerably depending on the CG, and the airspeed. The autopilot starts off with a low gain, and if the performance is not good enough it automatically slowly increases the gain until the performance is acceptable. If the gain gets too high, and its commands cause the aircraft pitch to start to oscillate, it quickly backs off on the gain until the oscillation stops. This part works quite well, at least in the simulator. It'll be interesting to see how well it works in the aircraft. This feature, if it works, should make it possible to use this autopilot on many other aircraft types without changing the software.

Originally, I was considering using the Arduino micro-controller as the basis for the autopilot. Now I am considering the Parallax Propeller micro-controller. The Propeller is interesting because it has what are effectively eight separate CPUs (they call them cogs), sharing common memory. You can dedicate CPUs to specific functions, and they can whir away without being affected by what is happening in the other cogs. For example, I would dedicate one cog to reading the serial data from the Dynon EFIS, another cog would monitor the pilot selections from the control panel and display status info on an LCD display. Another cog would do the inner loop control law, which would be a pitch hold law. Another cog would do the outer loop control laws (e.g. altitude hold, airspeed hold, vertical speed hold, etc) which would send pitch attitude commands to the inner loop control law. Another cog would act as a safety monitor, and would disconnect the autopilot if one of the other cogs locked up, or if specified airspeed or load factor limits were exceeded.

This should be a fun project. If it works, I'll share the design info with the amateur-built aircraft community.

Dynon recently released a major firmware upgrade to their products. The new firmware adds a few new features, and fixes some bugs. It might even solve my EFIS battery charging problem. Today I decided to do the upgrade.

The Dynon firmware updater only runs on Windows. They may be selling products that can only be used by people who are non-conformists when it comes to aircraft, but they only support the conformists when it comes to computers. We don't own a Windows computer, but fortunately Terry has an ancient copy of Virtual PC, a program that runs on OS X, and emulates the hardware of a PC, fooling Windows 98 into thinking it is running on a bog standard Intel powered box. It runs slowly, as every instruction must be translated from Intel code to Power PC code.

Dynon has provided several dire warnings about making sure that neither the EFIS nor the laptop losses power during the firmware upgrade process - apparently the EFIS would likely be rendered inoperative, and could only be resurrected by a trip back to the factory in Oregon. The EFIS screen displays a suitably strong warning while the upgrade is in progress.

Virtual PC does a very good job of emulating Windows, including the infamous Blue Screen of Death - it gave me a good scare today when Windows crashed in the middle of the firmware upgrade. I had depressing thoughts of having to send the EFIS back to the US for service. But, once Windows rebooted, and I made another attempt at doing the upgrade, the Dynon upgrade program picked up right where it left off and was able to successfully complete the upgrade. Now I am charging the battery, and hopefully it will hold its charge for more than a few hours.

I was at a local used computer shop this week, and stumbled across a magazine with a whole bunch of articles on making things with microcontrollers to control them. That got me thinking about my concept of a pitch axis autopilot. I did a bit or research, and determined that while there were dozens of suitable microcontrollers, only the Parallax BASIC Stamp series and the Arduino family have development environments available that run on Mac OS X.

At the moment, the Arduino seems to be the front runner, as it has some features that the BASIC Stamps don't, such as the ability to directly read analog inputs, which could be useful to allow the gains to be varied in flight during development. The Arduino also seems cheaper, as it comes mounted on a circuit board with power supply, input and output connectors, etc, while the board for the BASIC Stamps is extra.

The Arduino microcontroller has an interesting background - it is the product of an open source development effort. In other words, rather than being developed commercially, it was a collaborative design project between a number of amateurs. Despite the humble beginnings, it seems to be quite a capable device, and perhaps because of the open source heritage, it has development environments that run on Windows, Mac OS X and Linux, whereas most other microcontrollers only support Windows.

I've also discovered that there is a plug-in that allows Python to be used to read and write parametres from the X-Plane flight simulator. In theory, that should allow me to trial various control law algorithms quickly at home, and only flight test the ones that look the most promising. I'll attack this after I get flying.

My Garmin GNS 430 came back from its WAAS upgrade on Friday, making it a GNS 430W. I reinstalled it this weekend.

The upgrade also included a new Terrain display, with warnings about terrain that is too close below, or high terrain that is within 60 seconds flying time ahead. It isn't quite a Transport Category Terrain Avoidance Warning System, but it is pretty good, all in all.

The GNS 430W needs a new antenna, and I was disappointed to see that the new one, although it had the same mounting screw spacing, was wider and longer than the old one. That meant that the spacer I had made to fit between the curved fuselage and the flat antenna bottom was too small. It was a pain in the you know what to make the first spacer, and I didn't want to to invest the time now to make a new one.

The new GA-35 antenna is on the left, the old one in the middle, and the spacer for the old one on the right.

I made some little tapered UHMW plastic spacers to go go over the two outboard screws for now. I'll make a proper spacer sometime after I get flying.

Looking at the Installation Manual for the GNS 430W, it seems that there are several different antennae that could be used. One of them, the A34, supposedly has the same footprint as my original GA 56 antenna. I should have done this research earlier, and specified a requirement for the A34 antenna. I'll call the avionics dealer I use, and see if I can return the GA 35 antenna, and get an A34. That is a more interesting option than making a new spacer.

This was a bit of a crazy week, as I was in the simulator phase of my annual Cessna C550 Citation II recurrent training. I had a sim every day except Friday, and had a fair bit of studying to do. Plus I spent a bunch of time working on the on-line course for the big avionics upgrade that our Citations are undergoing. And, I had a few household repairs to make.

Today I attacked the problem of updating the firmware in my Dynon EFIS to the latest version - 3.0.0, which was released recently. The Dynon Product Support Program requires Microsoft Windows. All our computers are Apple Macs. In the past, I would install the Dynon Product Support Program on my work laptop. But, our IT folks have changed the setup on our work computers, and I no longer have the ability to install new software on it. Arrgh.

Yesterday I remembered that I had an old copy of Virtual PC hanging about somewhere, and I decided to see if it would suffice. But first, I had to make a new connection cable. The EFIS, engine monitor and GPS each have their serial connections wired to separate audio jacks. I made three cables with audio jacks that connected to 9 pin DB-9 connectors on a Keyspan 4 port USB-serial adapter, which connected to my computer. I went with the cheap and easy approach when I made those interconnect cables - I purchased three audio extension cables with an audio jacks on one end, cut audio plug off the other end and crimped the wires into the sockets on a DB-9 connector. But those wires weren't really compatible with the sockets on the DB-9 connector, and the cable for the EFIS finally failed. I purchased high quality audio jacks, which must be solder to the wires, which is what I should have done in the first place. Today I solder a connector to three AWG22 aircraft quality wires, and crimped sockets on the other end, which is what I should have done in the first place.

After making a new interconnect cable, I hooked up my 12" PowerBook computer, fired up Windows 98 on Virtual PC v5.0.4 build 2404. I was absolutely amazed to see that the Dynon Product Support Program, running on Windows 98, running under emulation on a PPC CPU, was able to find its required serial port, which was provided by OS X via a USB to serial adapter. The support program was able to connect to the EFIS, so I did a firmware backup. This took quite a long time, but it was successful.

The EFIS displays a fairly frightening screen message during any downloads or uploads. I hooked a battery charger to the aircraft battery, and connected the laptop's power adapter to help ensure no power outages.

Then I updated the firmware to the latest v3.0.0, which also took a long time, and appeared to be successful. I was initially puzzled, as I did not see some of the expected changes in the EFIS display. But a quick review of the product Release Notes indicated that I had to use the EFIS menu -> SETUP -> STYLE -> MODERN selection to change to the new display format.

This is the previous display format, for comparison.

This is the new format. The colours haven't changed - the two pictures were taken with different cameras. Note that the digital values for the airspeed and altitude are now line up with the centre of the display, similar to the presentation on transport category aircraft EFIS systems. The aircraft symbol at the centre of the attitude display is more prominent. Many, many other changes, not seen here are described in the Release Notes

Dynon added an HSI function to their EFIS systems a few months ago. I would really like to have an HSI, as they provide a much more intuitive display than the CDI that I currently have. But, the Dynon HSI is only capable of displaying GPS information from the Garmin GNS 430, not VOR or ILS information.

Two days ago Dynon stated on their forum that they are working on a solution to allow the HSI to display VOR and ILS information from the GNS 430. Yeah!! Assuming their new implementation works properly, I will almost certainly eventually replace the CDI with a second Dynon D-10A. The lower D-10A will normally be used as an HSI, but it would also provide a backup attitude display, in case the upper D-10A fails. In that case, I would use the CDI display on the GNS 430 for navigation.

This picture shows the HSI in VOR mode.

Here it is in GPS mode.

I attacked several different jobs this week, but every time I hit a wall as I needed some small piece of hardware. Add it to the list and move on to the next job. I made some progress on the B&C alternator wiring - I got the field power plug assembled, but discovered that my original plan for the routing of the alternator output cable won't work, and the cable I have is a few inches too short for the routing I will use.

I removed one of the heat muffs from the exhaust. I have two heater boxes, and two heat muffs. I had a hard time finding a good place for one of the heat muffs, and eventually settled on a location that put the muff over a slip joint. There is a definite risk of CO coming through the slip joint and getting into the cabin, but I rationalized that my CO monitor would provide warning, and I could then turn off the heat. But, I was never really comfortable with this half-baked plan. I finally came to my senses, and recognized that putting the heat muff over a slip joint was a really, really stupid idea. So, I removed it, and put out a plea for ideas to the RV-8 List at Yahoo. Two alternative locations were offered. Someone promised to send pictures of how he routed the SCAT tune to a location on the front right cylinder's exhaust.

I started the leak test on the left fuel tank today. It'll be a few days before I know whether it holds air or not.

This afternoon I installed the mounting bracket for the wing leveler. There are a bunch of possible places to mount the wing leveler servo, and I chose a location under the mid-cabin cover on the right side. I borrowed the idea from Ray Lynn. I fabricated the mount a long time ago (details - one, two , three).

Installing the mount turned out to be much more difficult than I had anticipated, as there isn't a lot of room to work in that area. I could have really used some of those folks who assemble the mechanisms in DVD players. :) But, I finally got it screwed in place.

I even got the Adel clamp installed underneath to secure the wire bundle in place. That was a real trick, as there wasn't room to use hemostats to hold the clamp closed as I usually do. Fortunately, a couple of days ago I read a tip about using safety wire to secure a clamp closed so you could get the screw in - that worked great. Now I'm having a glass of wine to celebrate a small victory.

I'll install the servo tomorrow, but I'll leave off the rod that connects it to the control column. I won't install it until I am ready to start the wing leveler servo flight testing. I don't want it in place for the initial flight testing, as there are already enough risks to deal with - I don't want to have to worry about the servo doing something strange.

Dynon has just released a firmware update for the EFIS-D100, EFIS-D10A, EMS-D120, EMS-D10, and FlightDEK-D180. The biggest change is a new round DG option - either full screen on the D-10A, or split screen on the D-100.

There is a thread on their support forum for questions.

The updated EFIS hardware came with a new Remote Compass Sensor. For some strange reason the newer model Remote Compass Sensor has a different pin-out on the connector, so I had to disassemble the connector to move two sockets to different spots on the connector. This would be a fairly easy job on the bench, but I had to do it way in the back of the fuselage, after crawling in head first. The connector was up at the top, but there wasn't enough room to lay on my back, so I had to lay on my side, and open it up, being careful to catch all 8 tiny screws, nuts and clips. Pulling and repositioning the pins wasn't too hard, and I managed to get it reassembled without losing any of the hardware.

I primed the Remote Compass Sensor bracket, bolted the sensor to it with non-magnetic brass hardware, then crawled back into the fuselage again to bolt it in place. It felt great to get that done.

Another shot of the mount.

Here you can see how far back in the fuselage the Remote Compass Sensor is mounted. It is two thirds of the way back from the bulkhead behind the rear seat to the leading edge of the horizontal stabilizer.

Last night I had a nagging worry, as I recalled that the sensor needed to be well aligned with the EFIS. I had this distant thought that I had discovered in the past that I needed a shim, and I had forgot to put it in. Today I reviewed my notes from when I fabricated the mount, and sure enough, I found I needed a shim to get the alignment correct. I reviewed the EFIS Installation Manual, and it said that the Remote Compass Sensor should be aligned within 0.2 degrees of the EFIS.

Tonight I installed my old instrument panel blank, and measured its vertical alignment with a digital level. I crawled back in the fuselage, and measured the alignment of the Remote Compass Sensor. Sure enough, it was tipped forward almost 4 degrees from what the EFIS was. Drat.

I removed the mounting bracket, which allowed me to discover that the low strength LockTite that I had used on the brass hardware didn't provide nearly as much locking power as I had expected. It is a good thing I discovered this, as I believe that the bolts would have eventually have vibrated loose.

I found a combination of washers and brass nuts that tipped the sensor the required amount. As near as I can measure, it should end up within 0.1 degrees of the EFIS. The trick now is to never measure it again. :)

Now I just need to buy some medium strength LockTite and then I can bolt the mount back in place. And I noted a chafing problem between one of the static system lines and the mount which I need to address. I will be overjoyed to finally be done with the Remote Compass Mount.