Autopilot roll servo (7)

7 hours

With the pitch axis pretty well figured out, I removed the servo from that bracket and reused it to set up the roll axis.  Space was less critical for this one, but it had its own different challenges.

First, I knew I wanted to locate the roll servo in one of the outboard bays under the seat.  I decided on the right bay since it'll be close to the pitch servo, making the wiring easier.  There's plenty of room there, I just had to figure out how to  attach a pushrod to either the bell crank on the stick, or the right aileron pushrod.

I figured I locate the servo approximately here:

My next consideration was making sure the servo would be able to move the ailerons through their full range of motion.  This was a little uncertain without having the wings installed.  I took a look at one of the long aileron pushrods installed in a wing.  I found that the farthest the aileron could travel up and down would move the pushrod a total of about 4-3/8" in and out.

Based on that, I decided I should plan on being able to move the aileron pushrods about 2-1/4" either way from center.

The servo motor includes a travel-limiting stop that prevents the servo arm from going over-center, which can result in pretty much irreversibly jammed controls in flight.  The bracket allows the servo to rotate no more than 50 degrees in either direction.

So I needed 2.25" of travel at no more than a 50-degree angle...  Some trigonometry told me that I needed a lever arm around 3" long.  The standard crank arm that came with the servo is only 1.5" at it's longest hole, and the "long crank arm" is 2", so I decided I just needed to make a longer crank arm.

Using the crank arm I had as a template, I traced out a version on a piece of 1/8" aluminum and extended it to place holes at 3", 3.25", and 3.5".

The new crank arm is quite a bit taller than the motor itself, so I had to make sure it would still fit under the seat.  Luckily the seat slopes upward pretty steeply as it goes forward, so as long as I mounted the servo no farther aft than where it is here, the arm wouldn't hit the seat bottom.

Here's another view:

Next I had to figure out how to connect the other end of the pushrod to the ailerons.  My initial plan when I started was to just use a longer bolt and attach it to the same hole as the rod end bearing of the right aileron pushrod, at the bottom of the bellcrank on the stick.  That's why I planned on making the servo cover the whole 4-3/8" of travel.  After looking at it for a while I realized there was plenty of space on the stick bellcrank above where the aileron pushrods connect.  This would be simper to attach, and it moved the servo pushrod attachment point closer to the pivot, which means it wouldn't need to move as far to move the ailerons through their full range.

This new attachment point is about 1" closer to the pivot axis, so it doesn't move as far left and right as the aileron pushrods themselves.  Now I only need the servo to travel about 1.75" in either direction, instead of 2.25".  Doing some more math, I found I could shorten my crank arm quite a bit and still provide the full range of motion without exceeding 50 degrees of servo travel.

Now I only need about a 2.5" long servo crank arm, so I drilled more holes and test fit everything again.  It works!

Finally, I installed the seat to make sure nothing interfered with it.  I noticed that the bellcrank on the stick runs into the edges of the cutout in the seat before anything else.  I asked Sonex what is supposed to limit the ailerons' travel, and the answer was not the seat pan.  I think with both wings installed and the ailerons rigged correctly, each aileron will limit the travel of the other.  Bottom line, I think I planned to accomodate a little more travel than I needed.  Once I have the wings on I'll check again and I might be able to shorten the servo crank arm even more.

For now though, everything seems to work fine without any interference.  I drilled the roll servo bracket to the floor, and then pulled the floor off to finally clean up the huge hole I cut in the seat rib for the pitch servo.

I also riveted the roll servo bracket in place, along with some small reinforcement angles like I did for the pitch servo.

I'm calling this project done until later when the wings are installed!

Autopilot Pitch Servo (7)

7 hours

This is probably getting way ahead of what I should be working on, but it's more structural stuff inside the fuselage that I figured I could knock out now.

I'm planning on doing a Garmin G3X Touch setup for my avionics, and just like the newer Dynon and MGL systems, they make it pretty easy to add an autopilot if you want to.  All you need to do is figure out how to mount the servos to the controls, then run a wiring harness to them, and that's pretty much it.  This will add a few pounds of weight, but I want to do lots of cross country flying so I think the extra convenience will be worth it.

The plan is to install a pitch servo and a roll servo.  That will give me lateral and vertical control.  I've seen some more complex setups with servos for yaw, trim, etc., but I think just these two axes will be fine for what I'm doing.  For now I just bought one servo, since I'm just using it to design the mounting.  I'll buy the second servo later when it's time to install them for good.

Anyway, Garmin sells servo mounting kits for a few Van's models, and Dynon sells mounting kits for the legacy Sonex, but I couldn't find anything pre-made for a Sonex/Waiex-B.  I bought two generic mounting kits along with my one servo and went from there.

Here's the servo, with my hand for scale

Here's the servo attached to its mounting bracket

I'm starting with the pitch axis first.  I had a few ideas about how to connect the servo to the elevator pushrod, either under the seat, just behind the seat, in the baggage compartment, or even attaching it to the control mixer back in the tail.  I spent a while holding this servo and bracket up to various locations trying to decide what would work best.  I preferred not to put it in the baggage compartment because I'd have to make a big cover for it or risk having loose objects get tangled up in it.  I also didn't really like the idea of putting it way back in the tail for CG reasons.  So that led to me to locate it under the seat.

The center bay under the seat contains the stick and elevator pushrod, but there isn't room for mush else.  I wanted to somehow pass a linkage from the servo into that bay to either grab the elevator pushrod, or the stick itself.  I found that the round lightening hole in the seat rib was a perfect spot to pass the servo arm through.  I enlarged the hole into a rectangle, and that allowed me to set the servo in the next bay to the right, but have the arm pass through the hole right next to the elevator pushrod.

Once I got the position finalized I realized I could have kept that hole a little smaller.  The reason I had to cut a rectangle beyond the round hole is that the servo comes with a stop plate that prevents the arm from rotating too far and going over center.  It's not in the pictures, but it takes up a little extra space that I had to account for.

I wanted to attach the mounting bracket right up against the right side of that seat rib, but the flange on the rib that attaches it to the floor was in the way.  I ended up doing something that in retrospect I don't recommend, and that was I pulled the floor off and just cut a few inches of that flange off so I could set the bracket right there on top of the floor and against the rib.

I realized later what I should have done was just set the bracket on top of the flange, and made a little doubler plate out of some sheet to support the rest of the bracket.  That would have been less work and probably a little more rigid.  Anyway I put the floor back on and drilled the existing floor holes into the servo bracket, so the bracket structurally just replaces the chunk of flange that I removed.

With the mounting location decided, next I had to create a pushrod.  The mounting kit came with two rod end bearings and a piece of 3/8" solid aluminum rod.

I just eyeballed a spot on the side of the stick A-frame that I could bolt one end of the pushrod to, slightly above where the 1/4" bolt goes through to attach the elevator pushrod.  I used the pitch trim to set the elevator control to a neutral position, set the servo arm vertical, and measured the distance.  It was about 5.5".  Given the nominal thread engagement specified for the rod ends, I figured out that the pushrod should be pretty close to 3/4" short of the mounting holes on each end (or a total of 1.5" shorter than the overall distance I measured).  I went ahead and cut the supplied rod to 4".

Then I had to drill and tap each end of the pushrod.  A lathe would be the correct tool for something like this, but I don't have a lathe.  I figured out a way to get a decent result on the drill press though.  First I clamped a chunk of wood to the drill press table, and locked in the table position.  Then I drilled a 3/8" hole through the wood.  I removed the 3/8" bit, without moving anything else, and then I knew the drill chuck was pretty well aligned with the hole in the wood.  I inserted the rod through the hole, and fortunately it was a pretty tight fit.  Then I chucked in a smaller bit and I was able to drill a pilot hole nice and straight into the end of the rod.

I did that on both ends, drilling up to a #21 bit.  Then I clamped the rod in a vise and tapped both sides for 10-32 threads.

Finally I installed the pushrod!  After a little adjusting, I could move the stick through its full fore-aft range without the servo arm running into its stops.

After that I went back and added two small pieces of angle stock to reinforce the floor around the servo bracket.

At this point I thought I was done, but I realized that that pushrod was installled a little too high on the stick.  I didn't notice initially, but with the stick held aft, the bottom of the aileron bell crank interfered with the servo pushrod slightly if I tried to go full left aileron.  Pretty much exactly the same problem I had when I tried to mount the trim spring higher than standard.

I needed to attach the servo pushrod slightly lower on the stick to make room, but the side of the A-frame was not flat any lower than where I already had it attached.  I ended up making a little mounting tab that I could bolt to the A-frame, creating a new lower mounting point for the pushrod.

I had to shorten the pushrod by about 1/2", but with the new mounting point there was no more interference!

Tail Fairing Completion (1)

1 hour

Going back to finish up attaching the tail fairing.  Where I left off I had attempted to attach it with sheet metal screws, but they all stripped out way too easily.  I drilled out all of the holes and installed 6-32 nutplates, and now I can snug the fairing up nice and tight against the stabilators.  It's still not perfectly flat, but I think it's good enough for now.

 

Removable Panel (2)

2 hours

I have a feeling that I will eventually want to make changes to the instrument panel once the plane is "done," and thinking about trying to crawl into the footwell made me pretty unhappy.  So I decided I'd spend a little time and a few grams of extra hardware to make it possible to remove the panel later.

The panel on the B-model is attached to a bracket at each of the lower corners, and a flange on the underside of the glare shield. 

The brackets attach to the bottom of the panel with 8 rivets per side.  It was a tight fit but I squeezed #8 nutplates in for each hole!

The other attachment point is the flange underneath the glare shield.  It's mostly flat, except for the curved portions which are fluted.  I installed more #8 nutplates along the flat part.

For the curved area, I'll just use #8 screws and nuts.  I think reaching a socket around the back of the panel to remove those few nuts in the future will be annoying but still way easier than drilling out rivets or working on the panel with it installed!