Chapter 21/24 – Pilot thigh support done

I started off this morning, which was surprisingly rain free unlike the forecast had called for, by quickly getting the thigh support cover outside.  I set it up and then shot it with 2 coats of clear.

While the thigh support cover clear coat was drying, I took the opportunity to reinstall the fuel valve bracket and the actual Andair fuel valve.

A couple of hours later I installed the dry clear-coated thigh support cover.  Not surprisingly, I’m super happy with how it came out, both structurally and paint-wise.

Here’s a close-up shot of the support screw that secures the fuel selector valve bracket to the thigh support fuel valve pedestal base.

I made an honest attempt to finish installing the Dynon Intercom but was quickly reminded that I’m completely out of D-Sub connector pins and a host of other electrical supplies.  I put together an order on SteinAir and will pull the trigger either later tonight or first thing tomorrow after I do a bit more inventorying to ascertain my current benchstock levels.

Since I couldn’t wire up the intercom, I decided to make the forward pilot seat area fuel system video that I said I would a few months back when I made the GIB thigh support sump tank video.  So here it is:

After I messed around for a few hours filming, editing and uploading the video, I then spent well over an hour finalizing all the electrical diagram updates from the recent panel mock-up wiring.

Tomorrow I plan on sitting in the pilot’s seat (I wanted to give the clear coat a good 24 hours to dry before sitting on it) and figuring out the locations for the left armrest console components.  Also, I also plan to start the final installation of the Parking Brake pull T-handle and start on the nose hatch pull-to-open T-handle as well.  In short, I have a number of build tasks I can now start with the knowledge of my exact sitting ergonomics in hand.

 

Chapter 23 – Rabbit holes

Today I had a myriad of personal things to take care of early on.  When I got back to the build I really wanted to figure out some issues that I had run across when loading up the GIB headrest with components.

I had some questions for ElectroAir on their Electronic Ignition, and when I tried to call they were already closed.  When I checked out their site’s FAQ section, I noted their new requirement for mounting the MAP sensor: they now want it mounted on the cold side of the firewall, where before they stated to mount it on the hot side.  This new placement is in line with how GRT states their MAP sensor should be mounted, but throws a wrench in the works as far as how I have my MAP system configured…. so back to the drawing board.  Literally.

I didn’t want to just write a note and deal with this later, since the information I was collecting was all right in front of me.  I spent a few hours researching configuration options, figuring out fittings, and the physical install on each side of the firewall.  The end result was a couple of respective orders for very specific fittings (restrictor & barb), and an entirely new look of my MAP system by the time I was done.  In fact, below is a shot showing my MAP system about a year ago (top) and what it looks like now (bottom diagram).

In addition, although I don’t have a pic for this, I also spent a good hour updating my firewall components and pass-thru diagram, which has changed significantly from a year ago as well.

In prep for Rough River this year, I sent a note to Mike Beasley, Long-EZ builder extraordinaire, asking if he could bring his O-320 engine baffle templates he created a few years ago.  He had said I could get a copy, so I figured now was a good time.  Mike actually had soft copies of his templates, sent me those and I immediately went down to Staples and had them printed off.

Here’s a closer shot of Mike’s engine baffle templates.

While I was at it, I loaded up pg A14 of the plans on my thumb drive to have it printed as well.  I need to dial in and refine the shape of the rather rough-cut ribs that I have in my Strake Leading Edge kit from Feather Light.

I have to say all the copies came out about as perfect as could be from what I can tell (there are tick marks on pg. A14 and it did print out spot on).  With my manifold pressure system issues put to bed, I can now move on with the build and get back to my internal cockpit configuration tasks.

 

 

 

Chapter 22 – Heat System Overview

Today I started out by spending a couple of hours updating wiring system grounds and my comm component interconnections in some of my wiring diagrams.  I also figured out my Audio Mixer requirement and updated that accordingly.

I then went down to the shop and spent a couple of hours cleaning and organizing it.

Then, since my next task required the dismantling of the mocked up heat exchanger and duct system, I decided to knock out the next oil heat system video that I had promised to do a while back . . . so here it is:

 
Tomorrow I’ll break down the oil heat & duct system and start working on it again.  I’m not that far out from mounting it into the backseat area permanently.

 

Chaps 13/17/21/22/25 – Firing it up!

And when I say “Fire,” I’m not kidding . . . read on below!

Today I started out doing some planning and inventorying for my fuel system.  Since I’m going to make my new work demarcation line at the pilot’s seat bulkhead, going forward, then I needed to figure out my fuel system in real terms, not just in my head.  I drew out the fuel system on a white board and then did an inventory to see what I had on hand and to see what I needed to order.  I was fairly pleased that I had the majority of stuff I needed on hand, but will need to order about $35 in fittings from ACS.

I also did some research on the exact installation requirements for each fuel system component.  For example, on the Andair fuel valve I could clearly see that the big red selector handle needed to come off to install it, but I hadn’t even truly read the install manual until today to find out how to remove the handle, and install the unit.  I also had some questions on the FT-60 Red Cube fuel flow sensor that I eventually got answered on the VAF site.  In addition, I wanted to double check some info on the fuel pump.

After a few hours I was happy with my progress on the fuel system and was satisfied that I actually knew the direction I was headed with that system once again.

I then my sights back onto the Atkinson pitch trim system.  Having received my reamers via UPS this AM, I was ready to drill and ream a nice round 1/4″ hole in the elevator tube to mount the control arm of the pitch trim actuator.

First, I spent a good 45 min dialing in the new configuration and physical mounting of the pitch trim actuator by redrilling the 2 mounting holes in the mounting bracket.  Not only was I moving the entire unit aft a hair, I was also rotating the front of bracket down a bit more while the aft side dropped down some as well, but not as much as the front.

Once I got the pitch trim actuator assembly newly aligned, and the mounting hole in the elevator tube was ready to be drilled, I set up the control system within the plan’s specs with the elevators at neutral and the control stick at ailerons level, which is actually about a 5° stick grip lean to the left — again, as per plans.  I then drilled out the hole in the elevator tube and mounted the control arm of the pitch trim actuator.

Once I tested the newly realigned pitch trim system, the up trim worked fine but the down trim didn’t work at all.  Apparently, if you’re prone to moments of just being dumb, then you can expect issues like this!  Apparently, in my haste to get this pitch trim actuator in this bird, I forgot to double check its current position.  Since the actuator arm was all the way in, then when I threw the switch for it to push the elevators up, it did so with no issue. But then when I went the other way, well, by the time the actuator throw ended the elevators were merely back at level . . . and didn’t go down below the level line at all.

Another issue I had was just a little battery fire, or so I thought –after I removed the actuator to set it at its midpoint position– as I was coming around the nose and had moved the actuator power wires.  Well, as I rounded the nose I could see a bit of fire right on top of the battery.  Not sure what was actually burning, I yanked the wires off the battery and kicked it across the floor so if it decided to do something violent I had the fuselage between me and it.

Well, the fire went out immediately, but there was a good bit of smoke.  Apparently one of my patch cords shorted out and literally blew apart and started burning.  It’s interesting that none of the mil spec wires had any issues, but a couple of the cheaper leads were destroyed.  A few minutes later I felt the battery to see if it was warm, and it wasn’t.  I then checked voltage on it and it was a bit low at 12.59 volts.  But then again, I haven’t charged it in a few weeks and I have been using it a lot.  An exciting little sideshow to be certain, but with that out of the way, I got back to the task at hand.

After about 10 minutes of having the garage open and a fan turned on to clear out the smoke and smell from the burning battery wires, I went to work to fix the actuator configuration by simply taking the actuator out of the nose and extending it until it was half out (1.5″ out, with a 3″ total extension).  I then remounted it, verified and reset the control system to specs, and the drilled the new hole.  When I tested it the second time around it worked like a champ (see video below).

With the pitch trim actuator install taken care of for the most part, I then turned my sights back onto getting some of these painted parts knocked out.  I wet sanded the tool box lid lightly with 500 grit sandpaper, then 1000 grit and finally 1200 grit.

As I buffed out the tool box lid I was also uploading the lengthy video above.  It took a little over an hour total to buff out the tool box lid by hand.

Again, after the Ultimate cutting compound application, I then wrapped up the tool box lid by hitting it with 3 rounds of Meguiar’s Ultimate Polish.

I then set it on top of the tool box body.

Finally, my last task of the evening was to wet sand & Simple Green wash the GPS antenna cover.

After I prepped it for more paint, I hit with 3 coats of white primer/paint to cure overnight. This antenna cover will not be a gloss color, but rather either semi-gloss or even flat.

I’m heading into the weekend tomorrow and have plans tomorrow afternoon on, but I will continue to try at get all the nose components installed –including the now prerequisite pilot cockpit area– to allow me to build the nose top.

 

Chapter 13/22/25 – Pitch trim actuator install

I actually took the first 2 pics yesterday, but since I was heading out to meet some friends, and was trying to field an influx of phone calls, I missed my window to mount the pitch trim hardpoint into the sidewall.

No worries since it just gave me more time to figure out my exact plan on how to embed the pitch trim hardpoint into the right sidewall.  Below you can see that I started drilling out the hole where the hardpoint will get floxed/micro’d/flocro’d into place.

Today I finished the pitch trim actuator assembly hardpoint install, although when I got home my phone was dead, and it being my current camera I didn’t get a few key intermediate pics… and I really needed to proceed with this install since I’m falling behind schedule on this build!

I took a scrap piece of wood, drilled a 1/4″ hole into it, then covered the outboard side with packing tape.  Not shown is the clamp and the wedge that I used to keep the head of the bolt fairly parallel with the aircraft CL, and the face of the mounting hardpoint near 90° vertical.  This pic was taken a little while after I had removed the clamp and the wedge, when clearly my phone was recharged.

While I let the pitch trim mounting hardpoint cure, I wet sanded the tool box with 500 grit wet/dry sandpaper, let it dry, then mounted it, and hit it with 2 good coats of clear coat.

While the lower tool box body clear coat was drying, I then wet sanded the tool box lid, which of course proved to be a bit more stubborn than the tool box body in smoothing out the surface imperfections.  I got to a point where I just needed a bit of clean up around the corners when of course –although being VERY careful– I broke through the black paint to the primer below.  This tool box lid truly is proving to be the problem child from hell!  So, I dried it off and hit the corner edges with 2 more coats of black paint.  I’ll let it dry a few days before wet sanding it again and hopefully getting a few good coats of clear on it!

Here’s another something that I didn’t get a pic of earlier.  Apparently, when I drew my install point target circle, the geometry of how the mounting hardpoint fit into the sidewall was different when I could actually test fit the hardpoint piece IN the sidewall.  The original hole started in the lower left corner in relation to where I eventually mounted the hardpoint, which of course meant backfilling the open area of the hole with spare Divinycell foam.

I also knew that even though I drove the position of the mounting hardpoint as far forward and up as I could, that I still may very well need to rewicker the mounting bracket to get the spacing and/or alignment of the pitch trim assembly correct.  In addition, I know this hard point install won’t win any beauty contests, but just keep in mind that this will all get sanded, floxed, and covered with 2-3 plies of BID when the nose top gets constructed.

Here’s a downward shot of the pitch trim mounting hardpoint showing the mounting face of it straight in comparison to the sidewall.  Also, although hard to tell in the pic, the face is also vertical where the sidewall tapers out slightly as its height increases.

With the micro & flox pretty much cured, I then attached the mounting bracket to test fit this sucker!

I took about 10 minutes to file down both the top & bottom edge of the swivel bracket to allow for a better fit inside the U-channel mounting bracket.  Below you can see it swiveled inboard.

And here it’s swiveled outboard.  If it doesn’t look like it’s swiveled outboard much, you’d be right!  I still need to shave down the width of the swivel bracket on the outboard side by about 0.10″ so that it has clearance to swivel outboard the required amount.

I then mounted the pitch trim actuator and assembly.  I could instantly tell that this entire assembly was sitting about an inch low.  My suspicion when mounting the hardpoint proved correct, meaning I’ll have to rework the mounting bracket assembly.  I’m still very happy with this install though since I’ve refined my target positioning from about 50 meters down to 5 meters . . . meaning that while not in it’s final, spot-on position, it’s very close and very workable!

Here’s a wide angle shot of the entire pitch trim assemble.  In his install directions, Vance Atkinson states to mount the pitch trim assembly at a point on the elevator control tube about 1.5″ forward of where the control tube passes through the instrument panel.  Right now, I’m about 3/8″ (0.375″) forward of that (so, 1-7/8″ forward of the panel).  I’m hoping since Vance had his assembly pretty much parallel with the aircraft waterline, and mine is much more aligned with the actual angle of the elevator control tube, that the added 3/8″ won’t be an issue.

[Note: As I understand it, some Strong pitch trim units installed in the same position where I’m mounting mine proved to be a little problematic, with resulting minor oscillations and difficulty in trimming the aircraft.  Of course I’m not using a Strong unit, but nonetheless I’ll keep my eye on these potential issues and work to correct them if need be.]

Ok, although not a slam dunk right out of the gate, I am –again– very happy with how the pitch trim unit install is going.  I like it up front here because although it’s tight quarters where I’m installing it, it moves weight forward, keeps the aft part of the airplane more clear of components, results in much shorter power wires, and should keep my comms clearer & more noise free due to the motor works being away from any of my comm circuits.

 

 

Chapter 22 – Taxi light cover salvaging

Today was the Big Reveal day for the taxi light cover.  Since my layup was practicing the truest form of accelerated entropy last night, I simply covered it with Saran wrap and taped it up profusely!  This morning I took off the long cross pieces of tape that I used to keep the whole layup from squeezing up the sides of the taped form, really making the cover unusable if it had.  I was now at the precipice of finding out what I had underneath of all this tape.  Would it be usable?  Would have I have to start all over again from square zero?Let’s find out . . .

Ok, so here’s the result in Airdog’s crazy layup experiment emporium.  Quite an interesting specimen I’d say, but it really does look usable.  As I’ve mentioned many times before, so much of these one-off custom jobs are ITERATIVE processes!

I grabbed my sanding block and my Perm-A-Grit tools and went to town on the taxi light cover.  I knew that I would most likely create some holes in this cover to get the shape worked out correctly . . . and I was right.  It is interesting that the resulting pics all look quite a bit like snake skin.

Here’s the opposite side.  It may be a bit difficult to see, but because I taped the top of the cover down, it resulted in 2 actions:  One negative, and one ok.  The negative action was that it created “shoulders” on each side of the cover just below the top.  My plan was to cut these off and layup a ply of BID that will bridge across the resulting hole.  The squashing of the top resulted in the 2nd action, which was that it pushed the sides just below the “shoulders” inboard more…. which was fine and actually allowed for a tighter fitting cover (i.e. takes up less room).

Here’s the sanded taxi light cover next to the taped up taxi light that I used as the cover form.

Also, here’s a shot of the inside of the taxi light cover.  Aside from scratches from the scribe I used to pry the cover off, the carbon fiber really looks nice.  Overall, this pic again has a snakeskin look to it (Marco also pointed this out when he saw it).

After assessing the cover’s usability –including cover mounting & removal– I decided that because the lips on each side clamp around the flanged base of the taxi light bulb mounting assembly, that I it was not practical to keep the forward side of the cover in place.  I cut off the forward face of the cover to simply allow the cover to be slid on from the bottom (if it’s deployed/out) or aft (retracted) side of the taxi light.  After testing this configuration I could tell immediately that I had made the right call on removing the forward cover side.  Much easier!

I then prepped it for more glass (repair layup, if you will . . .) by sanding it to shape.  This prep also included using the Fein Saw to remove the right “shoulder” just below the cover top, which left a decent-sized hole in the cover.

To keep this story line flowing, I’m jumping ahead here with my pics.  The pic below is exactly what I did above, but on the cover’s left side.  The yellow epoxy spreader is being employed as a form weight to keep the overhanging glass flat against the side lip.  Besides wrapping around the taxi light bulb mounting base to keep the cover securely in place, this side lip will get drilled in 2 places, and be held in place with two 4-40 screws.

In addition to the taxi light cover, I continued to work on securing the 2 big power cables (one + power and the other – ground) that start in the nose battery compartment and end at the firewall & starter, respectively.  Again, I had to reroute these cables up and over my rudder pedal (you can see in the pic how they get in the way if not wrangled) in an arch/ Bell curve fashion.  With the 1st added Adel clamp in place, I marked the position for the 2nd additional Adel clamp.  Note in the upper right hand corner of the pic my markings for the Atkinson pitch trim actuator mounting.  You can see that I need to get these power cables as high up on the nose sidewall as possible, but still remain clear of the pitch trim actuator mounting.

After marking the spot, I then drilled the hole in the nose sidewall.  This area was a bit tougher to drill since, if you recall, that indented area that this hole was placed has additional plies of BID to beef it up for the mounting of the pitch trim actuator.

I then performed the “poor man’s” knurling of the 10-32 threaded aluminum insert that will be used to mount the Adel clamp in this location.  I wanted to use this threaded insert since it’s a bit more robust than a RivNut.

I then whipped up some flox and mounted the threaded insert assembly (including a taped washer mounted to it with an AN3 bolt).  I then clamped it in place to ensure it would be level with the surrounding sidewall surface when it cured.

Here’s a shot about 10 hours later after it cured to about 80-90%.  The flox was still just barely soft enough to be easily removed with a razor blade (there was an entire ring of it around the perimeter of the taped washer).

Although it’s a bit slow going, I am getting through all the electrical related taskers on my list.  I expect about another week, maybe 2 at the most, before I’m ready to get back to the big chapter build items again.

 

 

Chapter 22 & 25 – Done, Done & Done!

Today I got a few things checked off the list.

First, I finished constructing a micro spreader to hopefully accelerate the finishing of the aircraft surface in prep for painting when the time comes.

Here’s a shot of the actual bottom plate working surface.  You can see I used 2 counter sunk aluminum rivets per bracket to hold the bottom spreader plate to the 4 individual brackets.  I figured 2 rivets per bracket would be enough since the primary force on the spreader plate is from the top as it pushes the micro over the aircraft skin.  If there is any issue I can of course beef up the bracket attach points.

Next, I got this pic from Marco showing my finished Nose Gear AEM box and a United Airlines bear he picked up for my little 7 year old buddy (a friend’s daughter) who loves flying.

Finally, the box Marco 3D printed for me for the Nose Gear RCU had a slight blemish on the lid, so I’ll hope he’ll forgive me for me filling it in a bit and painting the lid black.  Here’s the finished Nose Gear RCU box ready for mounting into the aircraft.  I will of course have to terminate the wires on the “B” side of the connector.

With my bridge sander, micro spreader and RCU box distraction projects out of the way, I can now get back to my list of electrical system related taskers that need to get knocked out.

 

Chapter 22 – 3D Printing AEM Box

Not me of course!  I just got these updates from Marco.  After some trials and tribulations with his 3D printer he was finally able to get a good print on both the AEM box . . .

… and the AEM box lid.

Since I pulled the trigger on the Laserware SF11/C laser altimeter I will soon have all the components required to implement Marc Zeitlin’s new nose gear Auto Extend System, with a twist of course: the backup battery emergency gear extension feature.

To get an idea of what it will look like installed, this is Marc Z’s laser altimeter installed on the inside of his Cozy gear leg cover.  Of course the Long-EZ doesn’t have this gear leg cover, so my laser altimeter will get installed inside the hell hole near the gear.  Also, note in the pic below that Marc mounted the 12V-5V converter near the laser altimeter, whereas my converter will be located in the AEM box above.

Finally, in a discussion I had with Marco I send him this pic of my GRT Mini-X Magnetometer.  Since I had it on hand I thought I would post it here.

It’s been a bit of effort to get Marc Z’s new gear system implemented, especially with redesigning the 1.2A backup battery emergency extend circuit back into Marc’s scheme. But in the end I think all this extra effort will definitely result in a much more refined, simple, optimized and user-friendly nose gear system.

 

Chapter 22 – Nose Gear AEM Box

Here’s the latest status on the Nose Gear Auto Extension Module (AEM) that will replace Jack Wilhelmson’s original AEX feature.  First off, I sent this pic to Marco to help show him why I had a requirement that the AEM box be 1.8″ or less: since the AEM box will be mounted in the old AEX box spot, where I have a notch in the top aft side of my NG30 cover that is just a hair wider/deeper than 1.8″.

After a number of discussions back & forth with Marco on the particular specs of the AEM box he was able to generate these fantastic renderings of the box.

Here’s the initial rendering.

Then one with the lid and raised letter labeling.

One of the aft, left and top side.

Finally, one showing the internal standoffs for the two airspeed switches and mounting screws.  As you can see, he has included the 15-pin D-Sub connector in the renderings as well.

In the next few days Marco will 3D print this AEM box when he gets a chance.  I just really have to say that everything for the new nose gear wiring & AEX system is going exceedingly well!

 

Chapter 22 – Final Brackets, this time!

After the BID glass cured on the Throttle Handle electronics cable P4 connector bracket, I pulled the peel ply and razor trimmed the glass.  As I was redrilling the connector mounting holes I set this connector body in place to ensure the spacing was good.

So, here’s the final product for Throttle Handle electronics cable P4 connector bracket.

I then did pretty much the same thing for the Dynon Intercom bracket: pulled peel plied, razor trimmed, redrilled holes and sanded it all to clean it up.  I then set the intercom in place to see how it fit.  I’m definitely happy with how this intercom mounting is turning out so far.

What I’m not happy with is the forward right bracket nutplate.  It’s giving me fits and I’m going to have to drill it out and remount just a hair forward and inboard for it to align correctly with the intercom mounting hole.

Of course the other 3 nutplates went in without any issues, unlike the last one!

Besides redoing the 4th nutplate, I’ll also assess whether or not I need an angled strut at one or either end of the bracket to help support the cantilevered intercom.  I must say that this bracket is amazingly strong, but with vibrations abounding in flight I think I’ll throw on one small corner bracket to bolster it a bit.  It will weigh next to nothing but will add a lot of structural support.