Disclaimer

This blog is for entertainment purposes only, and is not meant to teach you how to build anything. The author is not responsible for any accident, injury, or loss that occurs as a result of reading this blog. Read this blog at your own risk.

Sunday, July 30, 2017

Ch 22 - Electrical/Avionics - Part 6

General wiring and the junction box/plate

In the last episode I discussed how the plan is (for now) to bring all the individual wire bundles to a junction box/plate, then do the actual system interconnections on its back side. 

Today I will finish the first version of this wiring, knowing full well that it will change soon for a few reasons, one… I’m sill missing the 429 ARINC module, two… I might not get all the functionality I need from this configuration, and three… I’ve never done this before, and I might just blow up the shop.

The first thing on my plate was providing a power supply to all devices used, and since this is commonly done through the use of a “bus bar”, I ordered one from SteinAir and started prepping it for panel installation. This was pretty straight forward since I had made the Circuit Breakers spacing to be 3/4” (19 mm), so all I needed to do was drill screw holes every 3/4” on the bus bar.


Drilling my new power distribution ba

I bent the top end in order to clear the big 5A breaker/switch for the servo motors, and provide a way to connect to the existing Essential bus.


The circuit breakers will connect to the holes in the copper bar

After shortening it a bit, I mounted it in place by screwing all Circuit breakers to it. 


New bus bar mounted

When the existing Essential bus gets connected to the copper strip, all the CBs will receive power, and trip as necessary if a short or overload is detected within their respective domains.

The second thing I needed to make was a ground return path. I chose to tie all returning grounds together at the back of the junction box, in order to have only one wire exit the panel, and keep it more easily removable.


Ground return path being worked out

All grounds will obviously be tied together here

A ground bar is born

Most of the wiring done (Yep! One pin at a time.)


Now, all I’ve done so far is bringing in power and devices to the junction box, not actually connecting anything to anything else. The hard part will be doing the “smart wiring” inside the junction box (aka… behind the junction plate).

This obviously requires a lot of time spent on each device’s installation manual to understand what connects to what, why and how. I will spare you the learning curve and show you this intermediate diagram I drew up as my guide at this stage of the wiring…


Not the final version yet, but a good start.

If you notice some resemblances to Wade Parton’s diagrams, it is because I’m trying to keep as much commonality as possible with his design in order to simplify possible diagnostic and troubleshooting. Since we both have at least one Mini EFIS, a GNS480, and a TT-22 transponder in common, this makes it easier for him to understand my design choices, and vice versa.

So, with the junction box/plate (JB) removed from the panel and flipped over, let’s start stringing wires based on my schematics.


Junction box/plate removed for initial wiring

This is the actual brain of the operation

I never said it would look orderly. Perhaps I should go back to the enclosed box concept.

After finishing the main wiring of the plywood instrument panel, I did an initial power up test. The "magic smoke" remained inside the radios 😁, so all is well at this point. 

Because GRT Avionics hadn't shipped the ARINC module yet, the radio was not able to communicate with the EFIS. Also, to keep things simple at this stage, Pitch and Roll servos were not plugged in yet, though the homemade junction box is already pre-wired for them. 





Turning on the switch for the first time




The panel is now a huge step closer to the actual airplane installation, and a vindication of my junction box idea, which so far has performed flawlessly. 

Of course none of this would have been possible without some "electrical mentoring" at a distance from Wade Parton, a lot of time digging through the many manuals, and asking “stupid questions”.


Wednesday, July 19, 2017

Ch 22 - Electrical/Avionics - Part 5

Back plate (aka Junction Box)

With the business end of the instrument panel somewhat functional, it is now time to focus on the logistics of this upgrade, how to tie this panel into the existing electrical system, and how to connect all parts to each other.

Did I mention this is still a work in progress? Any or all of this might change in the near future, and at this time I am considering what I am doing as a proof of concept, knowing it will surely need some refining later on.

“Boy, do I smell another rabbit hole!”

So, let me first introduce you to the last component in the upgrade… the servos.


GRT servo

Ok, technically there's another item missing, the ARINC module, but since I haven’t received it yet I will kick that can down the road a little further.

So, what do the servos do, and how many do I need? 

Well, in the simplest of terms the servos are meant to replace the pilot's right arm (usually) so that he no longer has to hand-fly the plane. 

Personally, I love to hand-fly airplanes, and I do it as much as possible, as often as I can, but there are good reasons for letting “George” (autopilot’s stereotypical name) handle it sometimes... for example... very long flights.

The Long EZ has a standard fuel capacity of 52 gallons. With a reasonable 6.5 gal/hr fuel consumption, we are talking about an 8 hour endurance. I think I’d let George get his fair share of flying on a day like that.

Another perhaps less obvious reason is task saturation. Hand-flying an airplane can be very demanding and use up a lot of one’s attention, that leaves less brain power to plan ahead, especially in a single pilot scenario. 

While planning ahead is crucial when flying in general, it is of vital importance flying single pilot IFR (i.e. in the clouds), which I might end up doing from time to time in order to get back home. In such situations I would consider an autopilot essential, if not mandatory.

Back to the servos, what they do is take orders from the autopilot controller inside the Mini-AP (the AP stands for AutoPilot), which in turn takes orders from the pilot, and move one or more of the control surfaces to achieve the pilot’s directional objective. A little like the CNC mini-mill when you think about it.

The plane that Terry built has only one servo controlling lateral movements, allowing it to fly a course-line to a destination automatically, but still requiring the pilot to manually handle the vertical profile of the flight. 

While this is not a perfect solution, it does spare enough brain power to more easily stay mentally ahead of the airplane, and handle menial tasks like checking the weather, managing power plant and fuel quantity, looking for other airplanes, talking to Air Traffic Control or the passenger, etc. etc.

I chose to upgrade the system to two servos, one for roll (as per current setup) and an additional one for pitch. This will allow me to set, seek, and maintain any chosen altitude, and fly approaches to destination hands-off in crummy weather.

All of this electronic wizardry has to be installed somewhere, and be able to communicate with each other in order to work, and today I’ll be taking the first few steps into putting together the real backbone of the upgrade. 

Let’s talk about communication first, then I’ll show you the physical details of the upgrade. 

One very popular electrical connection is the D-subminiatures type, aka “dsub” connectors. If you were to look at the back of an older computer you might find a few  there (the VGA monitor's plug comes to mind). As you shall see in the photos below these plugs are indeed widespread.


D-subs on the back of the Minis

D-sub on the GRT servo

Yep! D-sub on the magnetometer.

Oh my! D-sub bonanza on the Garmin GNS480.

Even the still missing ARINC module has two such connectors on it.

Buying already made harnesses might be possible for some fairly standard installations, but not in my situation. Besides, I alway thought a better way would be to figure out what connects to what, and learn how to make the harnesses myself.

Believe it or not, this is not too difficult a task, and there is a lot of help out there for those wanting to venture into this arena. I’d like to mention SteinAir's YouTube channel on wiring, as they are very informative, plus one can buy all that is needed from them as well, or EAA's own videos on this subject.


D-sub connector making supplies

Since the connector-type decision was pretty much made up for me by Garmin and GRT, let's look at the D-subs a little more closely.

There are two types of D-subs, the ones you solder the wires to, and the ones you have to insert the pins into after having crimped the wire onto the pin. I chose the latter, and purchased the crimping tool with standard and high density pin dies. Once you get used to crimping, this method is so much faster, and your fingertips are safe from burns.


Life's too short for cheap wire strippers. This tool make stripping wires a pleasure.


22 AWG wire stripped and ready for crimping

Wire needs to show through the witness hole before crimping

Ratcheting crimping tool in action

Pin #9 getting inserted into the D-sub shell

"CLICK!"   The pins easily snap in place.

A fully configured EFIS D-sub connector

Back-shells are optional, but such a good idea for longevity.

Working on the opposite D-sub

In a previous post I might have mentioned that the electrical schematics of Terry's plane have been lost by one of the former owners, a real shame if you ask me. So, how is one supposed to tie this new panel into the existing electrical system?

Hmmmm... Because I didn’t want to have to undo any of the current wire bundles, I needed to design a completely self-contained system that connected only to power and ground. I also wanted to make it easily removable in order to take it home to diagnose, rewire, or reconfigure as necessary.

That's how I came up with the "junction box" idea. I figured I could bring the wires from all devices to the box, then do the "smart wiring" inside it. This means that all the harnesses outside the box would basically be straight-through harnesses (for the most part), so that a 37 pin D-sub from the radio would be brought to an identical 37 pins D-sub at the junction box. 

What this setup gives up in efficiency, it more than makes up  for in ease of installation and maintenance, since any harness could be easily removed and tested, and any remaining issue confined to within the removable box.

As I started putting pen to paper, the junction box morphed into a plate, but I still refer to it as the junction box. 

Wanna see how I made it?


The junction box/plate is held by 5" (12.7 cm) standoffs

A front view of the removable instrument panel

Reverse engineering the D-sub geometry. Many tries were needed to dial in the CAM G-code.

Successful 37, 15 and 9 pin D-sub G-code at last.

Making one of the six standoffs

Junction box/plate

Actually making the plate

Testing the D-sub fit

Front and rear structures

A view from the nose backward

Checking for fit of the Mini-AP

D-sub connectors mounted to the junction box/plate

Trying on the male 37 pin D-sub

Here are both Minis running on internal battery mounted in the new structure

Starting to string the MFD wires

As always, there were a few issues that needed to be solved along the way, but overall the junction box idea proved to be a great concept for centralizing all connections.


PFD harness completed

PFD, MFD and GNS P7 harnesses finished and installed.

Two of the four GNS 480 connectors completed and installed.

Next time I'll try to finish the rest of the wiring, then flip the proverbial switch, and hope that the "magic smoke" remains inside the expensive electronic equipment.


Monday, July 03, 2017

Ch 22 - Electrical/Avionics - Part 4

Face plate

Looking at all the holes in the current instrument panel, it became obvious that some additional glass work would be needed no matter what else I chose to do. Some holes needed to be closed and new ones opened. I decided move forward by cutting a big square hole into the existing panel and replace it with an aluminum plate. This would also allow me to do all of the work comfortably at home.

Here’s my initial CAD recreation of the plate/panel I am going for…


The bottom screen will be used as a map, of course.

These Mini-APs are very shallow, and that includes the battery backup.

I will have to replicate this design today (the top right will slope down a little)

I decided to use 0.090” (2.3 mm) 2024 aluminum plate, but I only had a 0.063” (1.6 mm) on hand, so I started version #1 with that. I figured I’d have enough configuration changes or mistakes to warrant at least a redo or two, so I got on with it, while I put in an order for the thicker stuff.


Test fitting the plate after cutting the top by hand

As usual my biggest problem was how to cut this big plate with my little mill.


Actual dimensions of the panel


I chose to cut it in three separate setups, knowing full well that every setup takes forever, and reduces precision inversely to the setup time.


First cutting operation

Op #2

Plate flipped on its back to final cutting

This required making a fixture, so I reused the one I made for the brake heat shields, except I flipped it over for a fresh surface on which to drill a bunch of holes.


If this looks vaguely familiar, it's the brake heat shield fixture.

Fixture flipped. Back in business!

Op #1 went down with a hitch.


Looking pretty good, but a lot more work to be done.

Ops #2 and #3 required careful, time consuming alignment using the just-cut machined surfaces and a dial test indicator. With that done, they also presented no real cutting challenge.


Op #2, circuit braker holes being cut.

End of operation #2

With the plate flipped on its back and carefully aligned, op #3 could begin.

The last operation only has to finish cutting the two EFIS holes

Finished plate on top of the original drawing

Finally, I got a chance to temporarily install the EFIS screens, switches, a couple of CBs, and the Korry-318 indicator lights, and check for issues.


Test fitting a few items

All is well except for the bottom row of indicator lights

Unfortunately my PFD hole was cut about 0.060” (1.5 mm) too high (my bad), and the second row of Korry-318 lights ended up too close to the PFD. I pushed them in half way anyway in order to take the above photo, but as I predicted I’ll have to cut this plate again.

Regardless, I used it to cut the wooden instrument panel, then I installed it, and ran some battery power to all the components for a real glamour shot.



All items powered, though not connected together yet.