More beanding of the rear bow - 8 hours

 Realizing that a mirrored template was not the option to go with, I decided to make a template based on the whole turtle deck. No question this is the most accurate template I can come up with I decided to work with this.

 Again, I spent a whole bunch of time bending using my feet and legs. Spent a crazy about of time on this with little gains. Sometimes I was going backward, sometimes forwards. Just a very frustrating process all round. I think I need some help with this, I'm going to reach out to some local builders who have experience and see what their suggestions are.

A key item worth noting. It is critical when bending the rear bow, that you are able to check that the canopy frame sits flat on a table. When bending the rear bow, you ultimately introduce twist into the frame. By the end of the very long build session I realized that after you bend the rear bow, you also need to twist the whole rear bow and set the frame a flat surface to check level. Its easy to do, just put the forward bow on the ground and step on it. Then twist the whole rear bow clockwise or counter clockwise enough that it will sit flat on a table.

 If you do this, things start fitting well. If you don't twist or check twist....you will get nowhere!   

Making a rear bow template - 5.5 hours

 I've reasoned that if the pilot side of the bow is fitting perfect, then all I need to do is make a template from the pilot side and flip it over and it will show me what I need to bend for the co-pilot side.  

I will save you a lot of pain and grief....Don't do this method. It doesn't work as well as you think it world and only makes things worse. Spent about 5 hours messing with the rear bow and was getting nowhere.

Starting to fit the sliding canopy frame - 5 hours

 After some head scratching moments, I figured out one important item that I wished I knew from the beginning. Make sure at least that the forward canopy bow tubes are the same length. Mine were out about 0.158" out of perfectly level. So factoring the width of my pen line, I made some measurements and made a cut reference line with masking tape. Again, I used a hacksaw rather than an angle grinder to cut the tube as I do not want to introduce heat into the metal.

Now that I have the frame trimed, the fit is much better.

A little work with the 3/4" conduit bender and I have the front bow following the curve of the roll bar.

 

 P.S. If the one side of the canopy bow is sitting high, you want to introduce a bit of bend on the opposite side. Putting bend on the side that's high only makes it worse.

Now, I need to start work on the Aft canopy bow. I can see the pilot side fits well, however the co-pilot side needs work to fit correctly.

 

Final fitting the WD-643-SS canopy brace - 2 hours

 With the canopy brace fitted into position, I was ready to drill the holes for the AN3 bolts. I first drilled the holes to #19, and then reamed them to 3/16" 

After the holes were drilled. I began marking the lower portion of the tab so I can remove the excess material. The blue line represents approximately 3/8" edge distance while the adjacent (first) red line is the edge of the 3/4" angle, the red line on the far right is just an extra amount of material for good measure. That is my trim line. 

Fitting the WD-643-SS canopy brace - 2.5 hours

 Drilling the top most bolt of the canopy brace is going to be fun. How to drill a hole through the center point of a round tube on a receiver in the roll bar that is only half round. Honestly I don't know how I would have done this without a 3D printer.

 I designed a cap that installs over the roll bar receiver and with #40 pilot holes for the mounting bolt. Working in CAD I can align the proposed mounting holes perfectly in a round tube and then just have to project those out to suite the half round receiver. 

 

I checked the positioning of the pen marks and confirmed everything was square. Now...How to drill the hole. There is no doubt that if I drill these by hand the hole will be terrible, but there is no way I can drill them in a drill press. So the 3D printer to the rescue, I can just add on a drill guide to my cap design.

The drill guide makes a prefect and perfectly aligned pilot hole. 

Using the 3D printer again, I made a small saddle that allows me to get a nice tight fit of the brace into the receiver. 

 

Once the brace is drilled to #40 using the pilot holes in the receiver. It is extremely quick and easy (and accurate) to up-size the hole to #30, #19, 3/16" and finally 1/4" for the AN4 bolt.


 

 

 

Initial fitting of the WD-643-SS canopy brace - 1.5 hours

 Now to start the process of trimming the WD-643-SS canopy brace. I began by marking out minimum edge distances as noted on the plans (3/8").

Next comes the more tricky part...Getting the canopy brace to fit, but not cutting too much.  This ends up being a very slow process of cutting a little bit with a hack saw, then checking the fit, and then cut and trial fit. After the first few cuts, I could then get the canopy brace into the receiver on the roll bar, however I could see that I need the whole brace to be moved back so I get edge distance on both the front and back of the brace. 

 After more small cuts and fit checks, I had finally cut the bar down the right amount. I did notice though, that the weld between the tube and the tab on the brace make a fillet that interferes with the top skin preventing the brace to sit somewhat flush to the front deck skin.  I used small round files to tighen up the radius of the weld as well a slightly up up the slot in the top skin to allow a better fit. As filing removed some of the powder coat, I touched it up with some primer.

 

Ensuring square on the WD-641-PC roll bar - 2 hours

 Now that I can get the WD-641-PC roll bar bolted properly to the fuselage. I need to confirm that the roll bar sits square (or 90deg perpendicular) to the longerons. The only reliable way I can think of to get an accurate confirmation of square was to build a wood template.  Using some MDF and scrap 2x2's I made a structure that would allow me to check the entire roll bar profile for level. 

Fortunately the roll bar was quite square as is. I only was out about 3/32" at most at the very top. Just a small amount of finger pressure applied to the top of the roll bar brings it to square, so all I need to do is drill the WD-643-SS canopy brace accordingly rather than fabricate any shims.

Bolting down the WD-641-PC roll bar - 2.5 hours

 If fitting the roll bar to the fuselage doesn't make you scratch your head, trying to bolt the roll bar (tightening the nuts) will.

On my first attempt and fitting the canopy frame I can see that I have so work to do ensuring that the roll bar is 90deg to the longerons. I have a gap between the roll bar and the frame. 

Making things more tricky is that I only have access to the AN3 bolts/nus of the roll bar. I can slip AN4 bolts into the rear mount of the roll bar, but due to the canopy deck lip accessibility to get washers and nuts on is near impossible. I need to get all bolts on and tightened if I have a hope of getting the roll bar positioned properly. I spent a few hours investigating my options, looking at special tools on the net (injector wrenches, offset sockets, etc) anything that would hold a nut in such a tight/odd space. Nothing I found would work. I see some folks on Vans Airforce forums just put a large scallop on the underside of the lip of the canopy deck allowing them to use a normal socket....This was my original plan, and went as far as drilling pilot holes in the lip. But I just couldn't bring myself to cut up so much of the lip. Between damaging the final paint and messing with the strength of the part, it was not the option I wanted. 

After some lunch, and a strong coffee, I had an ah-ha moment....Couldn't I just print a socket tool? I don't know why it took so long to think of this, but it could work. Assuming the plastic is strong enough? Well, I am currently using PLA+ in the printer. PLA+ is more rigid than regular PLA and its tensile strength is about double....give or take. I'll give it a shot and see.

I started to create a socket shape to accept the AN4 nut.

 

Next, I created a shape to test if I can get it to fit up under the canopy deck lip. 

Once I had all the rough dimensions worked out, I furthered the design to allow the tool to not only hold the nut, but the washer as well. Even when so far as filleted all the tool edges and cleaned it up! 

 

Drilling the holes in the Roll Bar - 3 hours

 The process of drilling the holes in the aft deck (to miss nicking the longerons) means that the #40 pilot holes in the roll bar are not perfectly aligned with the final sized holes in the canopy deck. After all, we are drilling them to final size separately. I opened up the aft mount holes in the Roll bar to 1/4" for the AN4 bolts, but left the forward holes #40. After some checks, I can see the pilot holes do not perfectly align with the AN3 holes in the canopy deck. I will have to move the pilot holes ever so slightly. 

The best way to rectify this is to make a drill template off of the canopy deck.

Eager to use the 3D printer again... I carefully made a template to fit the mount holes in the canopy deck. (Took a couple of tries to get it perfect).  

 

Once the template is confirmed to be a perfect fit with the canopy deck, I then mounted the roll bar onto the fuselage and clamped the template onto the roll bar. 

Taking the roll bar off, you can see the #40 mounting hole is not in the correct spot. Its just off of center from the template. 

Then using my smallest round file. I slowly...very slowly filed the hole larger and moved the center of the hole in the template. Once I got aligned much better to my liking,I was in a much better position to ream the hole to accept an AN3 bolt. (3/16") 

There is no denying this was a very long and labor intensive method, but in my opinion it was the only way to ensure precision. 

Finishing the F-668 spacers - 5 hours

 One small task that I have been putting off for a while is fabricating a new F-668 space for the one that I lost last summer when the 3M wheel decided to take the small part I was buffing on a ride. (I was working on the driveway and the piece must have launched about 30' in the air and landed in the landscape bed...never to be seen from again. I have 2 perfect forward spacers, 1 rear spacer nearly finished but not drilled. and 1 spacer I have to make from scratch.

 Being that I have 1 spacer that is a perfect template, creating the 2nd spacer was super quick. Next...I need to start working on locating the bolt holes in the aft spacers for an AN4 bolt. The real pain working with these is getting them to fit up inside the lip of the aft canopy decks. The 3D printer comes to the rescue as I made a simple little tool. I can glue the spacer to the tool and then get the spacer up into the canopy deck to check hole alignments. 

This allowed easy checks for the hole positioning on canopy decks with the aft spacers and I could easily transfer the #40 pilot holes in the roll bar to the aft spacers. Next I opened up the holes in the aft spacers to 1/4" for the AN4 bolts.

Next, I then made a few sets of tools that would allow me to confirm the positioning of the #40 holes in the aft canopy deck. The tool I designed would allow me to get in under the lip, sit right against the fuselage and the longeron and I could have a #40 hole on the top that would (or should) line up perfectly with the #40 hole in the canopy deck for the roll Bar. 

 I made a bunch of these tools with various hole sizes so I could step up the hole from #40, #30, #19, 3/16", and 1/4". Checking the holes each time I drilled up a size. This took some time to do, but it made the outcome easy... and perfect. All holes, ended up as close as possible to the longerons without nicking them.