17 October 2015

Avionics/Panel: Panel design.

Update 19-Apr-17:  Click the following to see a number of images of the installed panel in different lighting conditions.  Here is what the final panel looks like installed.

Update 5-Nov-15:   Designing your own panel?  Don't forget to include a "defrost" switch for the fans you install on page 38-32.  I went back and added the switch after noticing this image (lower rightmost "defog" switch), which I took when I visited Van's last year and saw the prototype N214VA and the factoryUpdate 7-Dec-15:  Also add a "canopy open" indicator light for the switch on page 38-19 if that indication is not managed by your EFIS.

I have completed the design of my panel.  Here is what it will look like (later changes included replacing the GMA-200 with a GMA-245 and replacing the TruTrak ADI2 with a Konardia Horis 57).

An explanation of my design is warranted.  Forgive the forthcoming uncharacteristic logorrhea, as I had intended my blog to be image-rich and coupled to brevity in language.
Here is what each panel section looks like in the CAD program.  Red lines indicate physical instrument borders and "no-go" areas, not cut lines (the curved lines aren't accurately placed, but they worked well enough for the purpose).  The remaining lines are cut lines and are colored by layers which I defined.  My guess is I'll need two panels cut to get things right.  The first will be for finding the spacing and fitment errors I made.  The second will be the final piece.

I printed each section out to scale and held it up to the panel frame, with the sun in front of me, to verify that there would be no steric conflicts or instruments overlapping with the frame.  I got it right the first time!  Here is about half of the left section from a previous layout revision.  You can see that the instruments do not overlap the panel boundaries.  Update 5-Dec-15:  The TruTrak ADI2 (the empty space above the 6 switches) had to be moved down so it wouldn't impinge on the Canopy Close-Out in the canopy frame, causing a subsequent reorientation of the switches. 

For the curious, I'll describe how and what tools I used to design my panel.  I didn't want to pay for additional software (there are panel design software tools available) and I knew that everything I needed was already available in open source options.  
  • To get a feel for the physical layup and visual layout I used PowerPoint.  If you don't have access to that program, you can also use the free OpenOffice (a spectacular office suite I've been using since it was StarOffice in the late '90s).  Using a CAD program prior to designing your layout is an exercise in futility.  I would advise against it until you have your layout resolved. 
    • For the instrument panel outline, I downloaded the DXF files from Van's and used an online DXF viewer to screen capture an image of each panel section.  In PowerPoint, you can easily copy-paste then flip the right section to make a left section.  
      • Each image was properly scaled to actual size then lined up correctly in PowerPoint after making the page size 45x12".
    • For each instrument I wanted, I found high resolution images of them online and scaled them to actual size. 
      • This also required getting the installation manuals for nearly every instrument, so I can find out what the actual dimensions were.  I'll need them anyway, so it was time well-spent.
      • Once I settled on using the Otto rocker switches, I just made a properly sized graphic of the switches using basic layered shapes in PowerPoint.  Same for the indicator lights.
    • Now I could just drag and drop my layout, trying different permutations to see what I liked, what I could do, what wouldn't work and what compromises were necessary (the latter mostly due to having a limited available real-estate). 
    • One thing to consider is that not all of the panel is available for use.  
  • With my layout completed, I could move to getting the CAD drawing ready.  
    • There are a number of free CAD programs available.  CAD is complex and requires a significant investment in time to learn how to do.  I wanted to minimize that.  I knew I didn't need a 3D CAD program, so I could limit my software search for 2D only, thus reducing the associated learning overhead. 
      • I landed on LibreCAD.  Though its documentation is severely lacking, many users have created online tutorial videos.  I watched about 15 minutes total from several by "TheReimber" and I was ready to go.  That was 1/4 of the time it took me to widdle through all the other lower quality tutorials.  And that was all the learning I needed to use the program.  It's that simple.
    • Using the necessary cutout dimensions I found from collecting the install manuals, I could design each cutout in LibreCAD and place them where they were needed.  
    • Each instrument was designed as a "block" so I could easily copy-paste as needed rather than, for example, redraw the switch cutout and border each time.
    • I used multiple "layers" so I could easily focus on which part of the panel I was interested in.  
    • I also created a red "borders" layer which defined where the external outline of the instruments would sit so I could avoid having them overlap each other, even though their cutouts might fit.  This border layer also included a 0.25" ring around the panel.

10 October 2015

Fuselage: Rudder and brake systems. Completed.

The rudder pedal assemblies took some time to get right.  The fitting of the correct washers and bolt tightness is a never ending back-and-forth, remove-and-redo game.  Eventually, it comes out right.

As an aside, the spacers from page 33-03 are difficult to cut straight.  I tried multiple times and just couldn't do it on my band saw.  Whilst I'm sure I'm not the first amateur machinist to devise this approach, here's what I came up with:  I cut the spacers intentionally too large, then chucked them into my drill press and brought them down on a file on both sides of the spacer until each was perfectly flat and just the right size.  I tried to get a picture of this, but the camera refused to focus on the subject, instead opting for the uninteresting background.  My apologies for the poor quality.

Installation of the assembly into the fuselage is quite easy, however you really have to pay attention to avoid over-torquing the nuts.  The balance between friction and free-play is made quite clear in the plans.  

The rudder cable guides took several hours to install.  It's nearly impossible to get access to the nuts and bolts and frustration was abound.  Here's how I ultimately secured the nuts:  First I positioned my wrench to work against the opposing bolt (left), then taped it down so it wouldn't fall when I was torquing the other side (right).

Eventually, each side gets done.  Left side shown below.

I think it's worth routing the rudder cables to the front of the aircraft prior to installing the baggage floors in Section 32.  It's still doable in Section 33, but it's much easier without the floors.  Anyway, I decided to prime and paint the cable CS-00015 cable links black to give them some contrast against the rest of the interior (cotter pins were not bent in this image).

The lines themselves are easy enough to cut, prepare and install.  The boiling water technique explained in Section 5-29 is very helpful to push the brass insert entirely into the tubing.  Boiling water?  Yep.  I love the varied building techniques used in this project!

Completed (I didn't finish the brake line routing until after I had installed parts from Section 35 as I was waiting for a #29 drill bit, hence the reason this image wouldn't match the end of Section 33).

08 October 2015

Powerplant: Engine delivered.

The engine was delivered today.  Twice.

I had no idea it had even been shipped (a week afterwards, I received a letter saying it was shipped the week before).  Luckily, I was home.  The first UPS truck showed up about 1:30 P.M.  When I answered the door, the driver said "I'm here to tell you that your engine is not being delivered today."  Apparently when it was shipped, the address was not identified as a residential one so the truck was not equipped with a lift gate.  The driver said they would try to deliver it tomorrow and would call ahead to arrange a time.

About an hour and fifteen minutes later, another truck arrived for delivery attempt #2.  The original driver had given his cargo to his second, lift gate-equipped colleague.  Again, no communication from the driver.  He just showed up assuming I was home.

He was not willing to place the pallet on my purpose-acquired dolly, so I had him place it behind the plane where it would reside in an unimposing manner.

Dimensions: 38"x25"x39" (WxHxD).  Weight:  379 pounds.  For the the civilized world, that's 96.5x63.5x99 cm and 172 kg.  The box rests on three 2x4s and that is then rested on a tired, old, ratted-out pallet.

The "Tip N Tell" was new to me.  Very ingenious (demo video here).

So the engine shall rest in that spot for...oh...probably most of next year.

03 October 2015

Fun Stuff: Number of rivets in wing, empennage, fuselage and finish plans.

I counted the rivets called out for in the finish plans and updated the chart on the lower right of the blog's main page.  Here is just the finish rivet count.  The break in the consecutive section numbering is probably due to the associated sections belonging to the FWF kit.

Number of Rivets in an RV-14A 
Finish kit only (see complete chart here)

38. Canopy and Window 650
39. Seatbacks 162
40. Tricycle Landing Gear and Engine Mnt 0
41. Wing Attachment 88
42. Miscellenea 60
45. Cowling 166
46A. Tricycle Gear Leg and Wheel Fairings 139

Empennage 4,446
Wings 8,276
Fuselage 6,250
Finish 1,265

Total: 20,237
The tail dragger would be very, very close to this number (in fact, same for wings).

I will update this list as my build progresses.  As always, it will live on the blog in the right column at the very bottom underneath the parts origin and skin thickness lists.  Direct link will be http://rv-14a.blogspot.com/#rivet_number

Want to wrap your brain around the mindlessness of the tasks involved in building a kit plane?  Consider that each of those rivets represents at least one hole (sometimes many more when parts are stacked) and each of those holes gets at least two deburring tasks and sometimes match/final drilling and/or countersinking/dimpling and/or sealant application too.  You can then calculate an upper-bound on the number of tasks per hole using some worst-case assumptions:  Just 6x the number of rivets to get a grasp on the number of tasks involved with the holes.

02 October 2015

Fuselage: Fuel system. Complete.

The fuel system is completed.  The fuel selector flange requires countersinking so that the screws sit flush.  It's important to clamp the part down solidly and with some backing for the countersink's pilot to avoid chattering the bit and making goofy holes.  I got the two side holes done perfectly, but the aft center hole had some chatter probably because I didn't clamp things down hard enough.  It's not too noticeable when the screw is inserted.

The fuel lines take some patience to get bent properly.  I had my friend, who build a -9A, complete the two lines from the wing tanks to the selector.  His approach was to use one section of tube as a sacrificial prototype to get the lengths, bend radii and locations correct.  He did well.  I used a 13/16" crow's foot to torque the outboard bolts.

The line that goes from the selector to the fuel filter was relatively easy for me to do.  I got it right on the second try.  The first one, per plans, was too long, though the bend is properly located.

The line that extends from the fuel pump to the firewall was exceptionally difficult for me to fabricate, probably because it requires manipulating the line precisely in all three dimensions with bends and rotations.  I tried three times and gave up, purchasing instead a prefabricated line from Aircraft Specialty  for $56.17 shipped.  One individual, posting on VAF, indicated his preference for a flexible fuel line into the firewall in case of an incident.  I have no expertise to comment.

Not shown in the pictures is that, following proper torquing of the lines, they are marked with torque seal to support future inspections.

01 October 2015

Finish: Seatbacks. Completed.

This is a very rapid and easy section, making for quick satisfaction in its completion.  Having said that, I actually finished it over several months in an effort to chase efficiency by fitting in the various steps to match the needs of other work (e.g., prepping for priming so it could be added to the priming of a bolus of other parts, painting with other parts, etc.).  

Nearly every rivet can be squeezed.  Those that can't are best done with help as it's easy to dent the hinge when the bucking bar slips off the manufactured head.  In this, I have some experience, hence my caution.  Here was how we positioned the seat backs to rivet across the hinge in the center.

Painted and assembled.  Complete with hinge pins.  The seat backs are remarkably sturdy, which is comforting should I decide to do a 6 g pull in the plane.  I  have yet to add the safety wire.

Remember, when you're building these that you have to build two (unless you want to fly solo all the time).  So every step gets done twice.