I got an AN3-6A bolt from a friend while waiting to get a replacement from TAF, so I was able to put together the seat locking mechanism for the seat.
EDIT: after reading the Sling 4 instruction manual, I now believe the cable goes on the other side, so I’ll move it down before I close it up and rivet it in.
Now having figured out the complete assembly, I also assembled the mechanism for the second seat, but unfortunately, the steel cable assembly for the second seat is too long, so I put in an order for a replacement.
So one mechanism assembly completed, the other on hold.
Then I completed assembling the second seat itself. When I built the other seat I noticed that I was missing some screws for the hinges and put in an order to get the missing screws, but I also remembered that I got some various metric screws from boltdepot a while back and luckily I had some countersunk M4x12 screws . Ialso found that there’s a typo in the instruction manual, which says they are M4x10, but the part number is HW-CAS-412-X-X-0 and they are actually M4x12. So I was able to put together the side hinges with those.
Now I just need some upholstery to make them a bit more comfortable to sit on. I ordered the upholstery a few weeks ago, so they should arrive in a few weeks hopefully.
It’s time to replace the now empty box of parts in the Garage with another full one. I moved on to the Fuselage box to get started with the interior of the Fuselage assembly. First order of business was finding all the parts for the seats.
After a bit of digging I found all the parts for the seats based on the inventory checklist in the box.
Once I had all parts in order, I started laying out the headrest based on the manual.
Then I clecoed it all onto the seat back and started riveting from the seatback.
Following the backside, riveting the front side of the headrest.
With the headrest completed, time to make it look like a seat. I checked out the detail diagrams for the hinges and put everything together. The holes of the hinges needed a tiny bit of enlarging which I did using a simple hand deburring tool.
Once I had the side hinges completed, I cleoed and riveted the bottom hinge of the seat in place.
Almost looks like a seat:
Assembling the bolt mechanism
The only thing left is the bolt mechanism that allows the seat to lock in place inside the Fuselage.
Unfortunately, this is where I found out I was missing something. The bolt that allows the pulley to pull the cable wasn’t quite long enough. It should be a AN3-6A bolt, but as it turns out, it’s a 5A bolt, so it’s missing the mark by just a hair.
Also I only got one bolt, set of washers and screws for the hinges, which (if it was the right length) is only enough to complete one seat, so I put in an order for the few parts and will give them a call in Torrance to see if they can send they few screws so I can complete the seats.
Here’s a picture of test fitting the mechanism with the slightly too short bolt:
After a couple of weeks of taking some time off active building while life happens and figuring out and planning out some things, I’m back to actively riveting on the airplane.
Time to finish the Pitot Tube after having cut the inspection panel hole and running the wiring a few months ago and lots of learnings about new tools, from the wiring, to flaring the tubes and connecting AN fittings.
I’m using the Garmin GAP 26 heated & regulated Pitot Tube (GAP 26-20). This version automatically controls whether the Pitot Tube needs to be heated using a regulator controller that is mounted next to the Pitot Tube and will only apply heat if needed based on outside temperatures. This basically will allow me to always turn on the Pitot Tube in my panel as part of my standard operating procedures and the regulator will control whether it actually needs to be heated to prevent icing.
After studying the installation manual to make sure I install it correctly, I measured the tube and had to figure out how far I have to cut it in order to fit.
Fitting the Pitot Tube
The Pitot and Angle of Attack (AOA) mast are over 12 inches out of the box, and that won’t fit. I made an initial guess and cut a bit shorter, but I was still too long so I made a series of shortenings and test attaching the fittings until I had it dialed in.
In the end, I had the masts cut down to right around 8 1/2 inches. The Garmin manual says to keep a minimum of 8 inches between the probe and transition to non-metalic tubing, so I had a little bit of margin.
Time for flaring the tubes. The AN fittings use a 37 degree flare, so I got a Rigid 377 flaring tool and a metal tubing cutter to cut the tube. Before doing this on the real Pitot Tube, I made some test flares on a spare tube I bought from Aircraft Spruce.
After I had that dialed in, I did another test fit to get the length correct, accounting for the bend towards the tube and then mounted the fittings.
Final fit and connecting to the nylon tubes inside the wing after having cut the nylon tubes.
Installing the regulator and wiring
With the Pitot Tube itself installed, time to finish the regulator and wiring that controls the heating of the Pitot Tube.
As I explained previously, I plan to mount the regulator unit onto the inspection panel plate, so I did some measuring and orienting to make sure the twist action of the round plate wouldn’t interfere with the wires coming out of the regulator.
Here is the final orientation that I figured out would work best (the screw will mount to the bottom, so the wires will come out the top):
I contemplated between screwing or riveting it on, but I figured that it’s unlikely that it will need to be changed out frequently, so I riveted it onto the plate.
The last part was to create a connection from the regulator to wires I ran through the wings. I used some weatherpack connectors for this, which create a waterproof connection and a solid crimp, similar to the Delphi GT 150 that Midwest Panel uses to connect the wiring harness.
Final completed connection between the regular, the Pitot Tube and the wires running to the center.
Installing the plate to the wing
The last and final part of the installation was to mount the inspection panel plate onto the wing. I did this last to prevent scraping up my arm while I had to do all the mounting inside the wing, since the backing plate that holds the plate in place has a series of pokey corners that love to eat airplane builder blood.
First I lined up the plate with the wing and then did the match drilling of the holes:
The I dimpled the plate and the matching holes on the wing and riveted it in place.
And here is the completed and closed up Pitot Tube and Inspection Plate that holds the regulator:
It’s been a while in the making after a few requests over the past several months, so I finally took the time and do a walkthrough of my Workshop where I’m building my Sling TSi.
Apart from walking through my garage workshop setup and a bunch of the tools I’m using throughout the build, I’ve also given a small update on my current tasks. I’m waiting for the balance tube to finish off the ailerons and I’m currently finishing up the installation of the pitot tube, after running the electrical wiring the other day.
I’ll make a separate post on the installation of the pitot tube when I’m done, but here’s a preview picture of the first fitting to figure out the length of the tubing:
I figured out where I want to run the wires a while back after some tinkering and I am using one of the strut channels for the length of the wing, except the very end at the wing-walk where I had to make one curve down the bottom.
The hardest part was figuring out a good way to come out the bottom where the wing-walk is, since the strut channel doesn’t go through there. On the last picture above, you can see when I finally managed to grip on to where I want the loom to come out of with the help of some duckbill pliers, which were a suggestion from my EAA chapters Technical Counsellor when he visited a few months ago.
Running the wire
With the question of where to run the wire solved, onto actually running the wires.
I am installing the Garmin GAP 26 Heated/Regulated Pitot Tube, which comes with a Regulator that needs to be installed next to the Pitot tube and controls whether the Pitot tube actually needs to be heated.
For this, there are three wires to run – two for the power and one for the discreet output, which integrated into the Garmin G3X Panel to show when the Pitot Tube is actually heated.
I ran the three wires through some braided sleeving to give them some extra protection and make running them through the wire channel easier in one go.
With that out of the way it “just” took a lot of back and forth, more use of the Endoscope and the thin arms of Juliana and repeated shouts of “push, push” and together we managed to run the wire all the way. She cheerfully pronounces “Congratulations, it’s a wire” as it came out the other end.
With most of the preparations out of the way and half of the skins riveted, I took one more session to finishing the Elevator.
There was only one extra part I had to do for 3 of the rivets that were on the bottom edge. In order for them to fit correctly, I had to shorten the rivets so they wouldn’t protrude out.
Aside from that, I just went to town and pulled the rest of the few hundred rivets.
The last part on the riveting side was the front lip.
One of the holes on the lip has to be enlarged to fit a grommet for the wiring for the Elevator Trim Tab. I enlarged the hole using my step drill bit and then installed a snap bushing.
The last part was to install the center balance counterweight. I did some test fitting with this, but the AN3-13A bolts that one of the versions of the manual that I have mentioned are too short, so I’ll check with the factory on the proper length.
With that being said, the general assembly of the Elevator is completed:
Timelapse of the complete construction of the Elevator
With the Elevator construction completed, I’ve also finished my video timelapse of the process:
The process is pretty straightforward, but there are a lot of rivets to be pulled, so it’s a lot of repetition, so I spread the work out over a few sessions.
I first did the half of the bottom, both left and right side, and then finished up the left side completely, followed by the right side.
After I finished the entire bottom half of the Elevator I flipped it around and put a small padding onto the Trim Tab control so it can’t dig into the Elevator skin.
The two day workshop helped in explaining principles of airplane electrical systems, wiring and bringing everything together to design and build out your avionics.
Aside from a lot of good learnings and explanations, there was also a couple of hands-on exercises to get familiar with crimping, soldering and connecting things.
The first exercise was to hook up a headset jack to a PM1000 intercom system. This was very handy, since whether you decide to build all your avionics or not, you will most definitely have to do the headset connectors.
I forgot to take pictures of the process, but here’s the finished headset jacks with the soldered connection. This included learning to ground the shielded wire, soldering the actual headset jack and doing some d-sub crimping for the intercom connector.
Also shown in the image above is the result of the second exercise, a crimped BNC antenna connector. This part, I was already familiar with from hooking up my NAV Antenna in the Rudder a few months ago.
The final exercise was to create a small electrical circuit. This includes a “master” switch, circuit protection in the form of a fuse and a dimmable “cabin” light that is tied behind the master switch. Aside from the practical application of the exercise, it also tied together a lot of the theoretical parts of the workshop and was a great finish for the weekend.
Here’s the finished working circuit in action:
I also ran into two other Sling Builders, Richard Howell, who recently started building a Sling 2 and Skip Jones, who is also building a Sling TSi. Now we just need to all finish building our airplanes and then we can be a chapter of Sling Pilots in the Pacific Northwest.
One of the things that the factory forgot as part of the change from the Sling 4 to the Sling TSi was the required access to the internals of the wing next to the Pitot tube.
In the Sling TSi design, some inspection panels were removed including the one next to the Pitot tube. By itself, if someone was building the wings from scratch, that might be fine as long as the builder installs the Pitot tube beforehand and doesn’t anticipate to ever need to access it, such as if using an unheated Pitot tube.
However, I am going to use the Garmin heated & regulated Pitot tube, which not only requires wires to be run to the Pitot tube, but also that I need to mount the regulator unit next to it.
Since I ordered the quickbuild, I ordered it with this specification, but unfortunately the factory didn’t receive the Pitot tube from their supplier in time and shipped my kit without installing it.
So after I received my shipment and inspected everything, I realized that installing this after the fact wasn’t going to be easy, particularly with the lack of a hole in the wing. Unfortunately the factory also forgot to run the wiring to the pitot tube, which creates a whole second issue, for which I’ve been working on a solution.
I informed the factory a while ago and also gave Matthew, one of the other TSi builders a heads-up since he hadn’t started on the wings yet. The factory realized their mistake in the plans and promised to come up with a solution and send me instructions and a plan.
Drafting a plan
While I was waiting for the factory to come up with a plan, I actually started drawing up my own plans to fabricate the entire inspection panel myself and used it as an opportunity to learn and use Solidworks, which I can use for free as part of being an EAA member.
Since I had the chance to chat with Mike Blyth at Oshkosh for a while and we chatted about my build, I mentioned that I was still waiting for the factory to come up with a plan for the inspection cover and he promised that he’d check on the progress when he got back to South Africa and indeed, two weeks later, I got an email with the plans.
The factory plans, in keeping with the other inspection access panels, uses the same flush round inspection cover that is used to access the Flaps and Aileron connecting rods.
Since I am still busy with other things in the build and haven’t actually made my own panel yet, I am going to go with the factory plans that they have drawn up for me.
Cutting a round inspection access panel hole
There is just one difficulty to overcome – the access panel is round and large and I don’t think there are 143.4 mm drill bits I can buy in Home depot.
Since this is quite an operation, I decided to get some practice with the tool on a piece of spare aluminum and also made a video of it, since I figured that it might be helpful for other builders in the future.
I started by marking out the circle using a drafting compass. It’s been quite a few years, but luckily I still remembered how to use it and how to find the center of the circle again by making two marks. Proof that you may indeed use what you’ve learned in geometry class sometime in life, even if it’s 15 years later. After that I clamped the piece of metal on the edge of the table.
I drilled the center pivot hole to 1/8th of an inch, which makes the pivot sit in the hole and then measured out the starting hole for the drill, make a starter hole and then used a step drill bit to upsize the hole until it aligned with my marked circle.
After that, I set up the drilling tool with the pivot and made sure that the outside of the cutting bit aligns with my circle and then attached the drill and went to work.
Here are a couple of pictures of the first circle I cut – note that briefly I had the pivot point jump out of the hole, which caused me to waver a bit which you can see towards the bottom where it’s not perfectly round, so make sure the pivot continues to stay in the hole.
Annotated video of the process of cutting the hole
It’s been a busy few weeks after returning from Oshkosh, so it’s been a while since I’ve written an update.
Since I spent the whole week at Oshkosh, I had a lot of time to figure out various bits and pieces that will go into the airplane and talk to the various vendors. I also finally got to meet Adam and Steve from Midwest Panel Builders who I’m working with for the Avionics.
It was great to get a feel for various things including the control sticks – I will go with the Tosten grip.
Another thing I was looking at is the TCW Control Valve servo for the vent shutoff valve that the TSi has. The standard design for this is a physical shutter that I’m not a great fan of, as I feel it looks a bit out of place in the panel. So I was looking for alternative options and Adam mentioned this servo as a possible option. After I checked it out, I think indeed it might be a nicer solution and I will try to make it work. In fact I might also use it for the heating control.
Aside from figuring out things for my own build, I also attended a lot of seminars, saw a lot of cool airplanes and the airshows and met a lot of interesting people and stories.
I also spent a lot of time lurking around the Sling tent and checking out various details of the completed Sling TSi that they had on display there. It was nice to see how a lot of the small details to the plans for the fit and finish were in this plane, which didn’t exist in Wayne’s TSi yet.
I was also able to meet a few of the other Sling builders at the Sling Ding meeting and had a long chat with Craig and Austin about our various builds. And I got to meet John & Marta King who I have to give credit to getting me through the ground school of my private and instrument training and ran into a few other people including Angle of Attack, Aviation 101, Jason Miller, JP The Candourist and Mike Patey and Draco the bush plane.