Fuselage

Tonight I picked up with the intent of working towards getting the elevator control system reinstalled. In all probability this wouldn’t be the last time, but to my way of thinking everything from the bell crank back at least ought to be able to stay in place. First up, though, I needed to finish up work back where the ADAHRS units were, specifically routing the pitot/AOA lines and securing them for good.

This was kinda fun, all in all I think I removed and replaced the backboards back in the tail at least twice. I needed them out of the way to initially route the lines back from the seat floor area, but then of course needed them back in place to crawl back into the tail again – except there was one issue, with the way the lines were going to route up the bulkhead, the next-to-aftmost board had to be left out. I could only hope that I wouldn’t need to be that far back.

After putting the boards back and collecting the hardware and tools I needed for the adel clamps, I squirmed back into place to where I could reach the ADAHRS. One thing I wanted to do before buttoning things up back here was to double-check the electrical connections, and this turned out to be a good call. Both the D-sub network connectors were snugged down (I figured the most likely thing would be that I didn’t tighten the thumbscrews), but when I gave the OAT connectors a pull test…a wire came out of one of them. I was kind of expecting to have a bare wire come out – which would have been really fun – but it turned out the pin just hadn’t been seated in the connector. I just had to pull the connector and reinsert the pin, which would have been dirt simple any other time, but here it took about 15 minutes thanks to the cramped space.

Finally, I got back to the pitot lines, first inserting them into the connectors, then pulling the slack back forward and installing the two adel clamps. This, too, was made rather enjoyable thanks to the location. In some ways it was more fun than the connector – instead of lying on my back and reaching over my head, as it were, I was on my side and trying to use a screwdriver in one hand and a wrench in the other.

Once the adel were in place, I crawled out, removed the backboards again, pulled all the slack forward, and secured the lines to the aft bottom skin where I added a zip-tie pad – this will ensure the lines stay well clear of the rudder cable. The lines are visible in the lower right of this photo, and the red silicone tape can just be seen at the zip tie pad:

Next, I went to work on the control system. I decided to basically work through the entire fit procedure again, to make sure everything was still adjusted correctly. This meant installing both sticks on the control column on the bench, adding the pushrod between them, and making sure both sticks were parallel. Then the sticks came out again and I got the control column back in the fuselage, and the sticks and intermediate pushrod reattached yet again.

Next were the other two push tubes. A while back I disassembled these and gave them a coat of epoxy primer, after the rattle can primer got scraped up a lot just with the previous fitting work. The smaller forward push tube just gets set to a specified length, so I readjusted the rod ends, tightened the jam nuts, and added torque seal for good measure. This was about the time I decided I wouldn’t be installing the tubes tonight – I’d like for the torque seal to dry first, But I did pull down the larger aft tube as well, and reset the rod ends to where they were before (I took notes on this). This should be the proper length, but I’m not going to torque-seal these rod ends until I’ve verified the length…which will happen another day.

So that was the night’s work. I should be able to finish double-checking this stuff this weekend, and get it installed for good. The only complication I need to consider is the autopilot pitch servo – I need to review those installation instructions and see about getting that permanently installed at the same time I do the bell crank.

So today I picked up where I left off last night, starting with getting the forward elevator push tube loosely in its place under the seat floors. Then I got to spend some time reviewing the annoyance of getting this whole system together. For whatever reason, the manual tells you to 1) slide the forward tube into place, 2) bolt the bellcrank to the forward tube, and 3) slide the bellcrank into place. Problem is, the bellcrank can’t go into place from beneath, which is how it would have to be inserted if this procedure was followed. So instead it has to go from above, and then the attach bolt has to be positioned and tightened in the very tight space between the baggage ribs. This was not much fun before the turtledeck was in place, and will probably be even worse now.

Anyway, before I could do any of that, I needed to take a look at the autopilot pitch servo and see how that would fit in with the rest of the procedure. Turns out that the way the servo mounts, it needs to be installed before the bellcrank. I also needed to drill the hole in the bellcrank for the servo linkage, as well as assembling the linkage itself. Finally, there was the consideration of the optional bracket on the servo to limit movement of the arm. Limiting the range of motion is important because if the servo linkage can go over-center, it’ll lock up the controls, which would be bad for fairly obvious reasons. Generally, the control stops and linkage setup should primarily prevent this, and the bracket is a backup.

After thinking all this over, I concluded that I was going to have to install all this stuff twice if I was going to get the limiting bracket set up correctly – that is, installing the servo, bellcrank, and aft pushrod, moving the elevators full down (putting them closest to the over-center position), and checking the alignment of everything. This led to the first of many trips in and out of the tail today. The real fun with this is that the backboards can’t go in place because they block access to stuff I’m working on, so I was left to work using a few random pieces of scrap wood placed across various ribs.

Now, the first problem was with the interference between the limiting bracket and the servo mount bracket. The install kit warned of this interference, and I figured the first temporary install would help me decide how to maybe trim the limiting bracket to fix the problem. Well, turns out the interference was way more than I expected, to the extent that I’d have to basically destroy the limiting bracket to make it work. So I decided not to install it – realistically, given that the elevator stops are beefy structural pieces, any event that allowed the controls to move past those limits would probably already be catastrophic.

The good news here was that I didn’t have to fit up everything, but I still had to pull the servo again to remove the limiting bracket. Then I got to climb in again, reattach the servo, and then start to really think about getting that forward tube and bellcrank put together. Well, that latter part was what broke me for the night. The rod end bearing on the tube installs with a thin washer on either side, which makes for some really fun coordination. Back when I did this before, I glued the washers inside the bellcrank to simplify this, but whatever I used (I think it might have just been superglue) gave up the ghost today. I made one brief attempt to just get them inserted manually, but there was just no way that was happening with the tight confines.

So I pulled the bellcrank out again and embarked upon gluing the washers in place again. This time I used epoxy, in the hopes that it’ll hold a little better. The fun part was figuring out how to clamp the washers in place while they dried, without also epoxying the clamping stuff in place. I ended up putting a short #10 screw through each side of the bellcrank, with a nut to clamp it all in place. Judicious application of the epoxy kept there from being a ton of squeeze-out that might glue everything together. If this was a real structural bond I’d have wanted to be more liberal with the epoxy, but here I just need those washers to stay in place while everything is assembled.

So that’s where it sits tonight – tomorrow when the epoxy’s dry, I should be able to finish getting the put tubes and bellcrank installed, and hopefully confirm that the aft tube length is right and the servo arm lines up as it should. Then I guess I’ll be gettin the rear avionics shelf installed again as well…maybe. Not sure if I should wait on that or not. In any case what I’ll really be moving on to soon is getting the vertical stab attached and then probably looking at the fiberglass work back there.

Tonight I picked up with the bellcrank yet again. Saturday’s epoxy had the washers nicely held in place, so I crawled back into the tail and went to work. First up was attaching the forward push tube to the bottom of the bellcrank, which actually went together surprisingly easily given the tight quarters.

Next was the center bearing on the bellcrank, which was where things got more interesting. Back when I first installed this, I found that I couldn’t fit the specified washer stack around the bellcrank – one regular and one large washer on each side just wouldn’t go in the space. So I just removed one of the regular washers and then it all fit fine, and I think at the time I figured that it was a “put what fits” type situation.

Tonight, though, as I tried installing all this, I ran into a problem: the large washer on the left side interfered with the forward attach bolt for the autopilot servo. There was simply no way to get it to sit flat where it needed to be. When I crawled back out of the airplane, originally I was just figuring on replacing the large washers with regular ones. Generally large washers are placed around bearings to help capture the bearing – that is, if the bearing were to fail, having it be able to come out of its housing could lead to a system jamming. The large washers, though, are often omitted if the bearing is surrounded by large flat surface that serves a similar purpose. It seemed to me that the two mount pieces might work well for this – but I decided to double-check the plans first.

Turns out I got this wrong the first time around. Instead of the large AN washers I used, the plans specify special super-thin large washers – which also have a slightly smaller OD. Small enough, it turns out, to not interfere with the servo bolt, or so I discovered during my second trip into the tail. This trip got even more fun when, as I tried to weasel the last washer into its gap and push the bolt through, I decided I needed a long, thin tool to get the job done. Not wanting to climb out and back in again, I summoned help from inside, by calling Josie using my Apple Watch. Sometimes technology really is super useful…

Finally, I connected the servo linkage to the bellcrank, then climbed out and finished the job by attaching the forward push tube to the control column. Then it was time for a control check, mainly to verify freedom of movement but also to double check that, with the sticks full forward, the servo linkage was nowhere near going over-center.

The latter definitely checked out:

Freedom of movement was a different matter. While I could move the sticks through their full range of motion, there was a weird sort of “bump” feeling a bit aft of the neutral position. Some investigation revealed the issue: one of the rivets holding the two bellcrank halves together is rubbing on the servo mount bolt head, the same one that was problematic with the washer earlier. The contact is light, not enough to block the stick movement, but enough to be noticeable.

It’s hard to get a photo but hopefully it’s somewhat visible here:

I think I can fix this by just reworking the washer stack around the bellcrank. If I move the regular washer on the right side over to the left, it should offset the bellcrank just enough to provide the clearance I need. If that doesn’t work, then I guess I’ll need to pull the bellcrank out and maybe replace these six rivets with flush equivalents. We’ll see how that goes some other day.