Involuntary (Pneumatic) Leak Test

On Thursday evening I had some visitors that wanted to see the plane. A 17-year old boy named Patrick who wants to study Aeronautical Engineering was the driving reason for this interest.
Among other things I also wanted to give him an idea of modern experimental avionic systems and so I fired up the Skyview. He was impressed and even more so as the SV showed a constant airspeed of 59 knots. Whatever I did did nothing to change the airspeed and I was getting seriously worried that I would have to replace the ADAHRS unit again.
We even (lightly) blew into the pitot tube to see if was open and yes, it did increase the airspeed but only to settle back at 59 knots when we stopped.

What was this?

Well, that night I was too occupied to show Patrick everything he wanted to see and I delayed the troubleshooting till the next day.

The solution occurred to me while I was at work. I remember that I had previously plumbed up the ADAHRS and I had even posted this a few weeks back on the blog. That by itself was not the problem though. What I had forgotten during the plumbing was the fact that I had still tape over the static ports as I am still worried about mud daubers clogging up this vital ports.
Now with the tape in place and the static system hooked up, I had created a closed system that changes pressure with the ambient temperature.
When we looked at the Skyview, it was night and bit less hot than on the day I sealed the system. This naturally causes a slight vacuum while the pitot was still measuring ambient pressure. Now the pitot pressure was greater than the static pressure measured by the SV and that must show as airspeed.

Perfectly normal!

I was anxious to get home and confirm my theory by removing the tape from one of the static ports and sure enough, the speed went down immediately from what was then 29 knots. It was much warmer than on the night when we saw the 59 but less hot than on the day I sealed the system.

This also means that this static system had gone through quite some pressure cycles and did obviously not leak enough to come to equilibrium with ambient pressure.

I think this is a great low leak rate and therefore I have involuntarily tested the static system. I will stil perform a formal test though but I am very confident that this will just verify what I have already determined.

The Good, The Bad, And The Ugly

There was one more cable swap to do and then I deemed myself ready for upgrading the Skyview from a 3.3.2 to the latest Version 6.0.0 of the firmware. That version jump looks more impressive than it actually is. It represents merely 12 months of firmware releases as my version was last updated in July 2012.
So what about that cable swap? Van's had released a notification on August 31st of 2012 that there was a wire swap in a conversion harness (for the ones that upgraded from a D180 system to a Skyview avionics system without swapping out the wiring). The wires of the secondary SV bus system connecting the two autopilot servos had to get "unswapped", so the backup bus system would function, in the unlikely event of the primary wiring failing.
I should add that although the backup wiring was always installed (and advertised by Dynon as a safety feature) it was actually never used in pre-4.0 versions of the firmware. The obvious problem of having the bus wires swapped (rendering the backup system compromised) was never detected before version 4.0 and before upgrading it was suggested to fix this issue.
That sounded easy enough.

All I had to do was to pull the Options harness plug and swap pins 3 and 4. It would have been quickly done but I had no means of testing that the swap was necessary or correctly done as my software version at the time of change was 3.3.2 and it did not use the backup wiring.
So I decided to first upgrade the SV display to the latest 6.0.0 firmware. And as all my previous experience with Dynon products, this was neither easy no very successful. No, it was rather frustrating again.

The first step was to grab the Dynon USB stick (4GB capacity) and copy the pure firmware update file on it (the one ending in .dub). The more modern update files with a .duc ending are not recognized by the old 3.3.2 firmware.
Now I did not delete the backup files and the 3.3.2 update files from the memory stick as I did not want to lose them.
Once I started the firmware update, I got to this screen:

This validation process appeared to take forever and no progress was seen. After more than 5 minutes, I ended up in the setup menu without any change. This update did not succeed. I tried a different memory stick (1GB with almost no free space left) and now saw an error message during validation, complaining that the update file might be corrupt.
I tried another, newer memory stick with 32GB capacity and practically empty and this time it worked. It appears that the SV system is using the memory stick for some temporary storage without saying so and not correctly complaining about too little free capacity. Great! That was the Bad - referring to the title.

Once all the external devices have been updated as well, SV reboots and comes up with these slightly modified screen.

As you can see, I still have not entered an activation key for the map yet. At least I started the process today and emailed Dynon my license key and my serial number.
Now, before I went to the Network Setup to see if the bus swap was necessary, I was eager to check if my display was affected by Dynon's Service Bulletin that indicated that a series of SVs had a board in them that was prone to failure due to a manufacturing process of a sub-contractor they had used. The setup showed that my system was in hardware group 1 which was not affected by this problem. That was the Good - in reference to the title of this article.

Now for the Network Setup and the check of the secondary bus system. Avionics and Autopilot switches were in the on-position.

So, it detected a connection fault (4/8 obviously indicates the secondary bus) but the system is operational. That seems to be consistent with a backup wiring failure. So, I turned the unit off, swapped the pins and powered the SV back up.

Huh? What is that?! Still two systems with a wiring fault???
A closer look revealed that now it showed a failure with the backup system of only one servo (which makes no sense as the servo connections are spliced and share the same bus harness and plugs) but now the ADAHRS has a problem with the backup system. How weird is that?
Now, to see if the swap of the pins really affected anything with the ADAHRS (I checked the wiring diagram and it should not do anything to anything but the two servos) I switched the servos off by turning off the Autopilot switch.

Just as I thought, the ADAHRS is fine and not affected by the wiring change. Once the servos are off, the problem with the ADAHRS disappears.
Well, if you are a long-time reader of this blog you will remember that I had a similar problem with the Dynon before. Actually, it was pretty much the very same problem, only that it affected the one and only primary bus system back then. The effect was that my ADAHRS was reporting just fine while the Autopilot was off but disappeared as soon as the servos had power. Initially both servos were working fine, a few hours into debugging the problem one servo stopped working, the other one was working fine and the ADAHRS was still dead as soon as the Autopilot switch was in the on-position.
After another dreadful experience with Dynon who replaced my brand-new servo with a refurbished one (but hey, it was free!) instead of another new one that would work, the problem was fixed. Or so it seemed until today.
My very strong guess is, that no-one ever checked the servos on their backup system and either the original servo or the refurbished one is now compromising the backup bus as soon as it comes online.
The one that is least accessible in the plane at this point is the roll servo, so I would guess that this is the one that is having the problem right now. The only sure way to find out is to get to the spade connectors and unhook one servo at the time and see which one makes the ADAHRS backup system disappear. That's the one that has the issue.
However, I am completely not in the right mindset to work on this POS again. Absolutely not in this heat and given that I have my experience with Dynon and know that their overwhelming helpfulness in this matter will possibly make me go postal.
For now I accept that I have a bug in the system that should get addressed but it is working right now and there is a safe way of getting the servos off the network and that is to cut their power supply with the autopilot switch. This is safe to get the bird flying. Once it is cooler, I can then crawl into the cone and see if my guess is right. So much for the Ugly in the title...

I continued to bring the SV system to the latest versions of the maps and data files.

I had bought a bunch of very nice antennas from Delta Pop Aviation. They are very well made and Don  manufactures them himself and really knows what he is doing. They are not TSO'd but very affordable.
Here are the transponder and UAT antennas:

The UAT antenna is the one for the ADS-B receiver. You could use a transponder antenna but that is not optimal as the ADS-B frequency for the receiver is quite a bit lower than the one for the transponder and the signal will be weaker if you do not use an UAT antenna.
The additional transponder antenna will be connected to the PCAS and it should be as far away from the active transponder antenna of the RV-12 as possible.

I also manufactured two backplates for the small antennas:

I marked a good spot for the ADS-B antenna about a foot behind the aft bulkhead and close to the left side of the tailcone to keep it away from the cables in the cone and to increase the distance to the PCAS antenna which I will likely put on the opposite side.

The #8 is not a drill bit size but a screw size. These holes actually need a #19 drill bit and the center has to get opened to 1/2".

I will need a helping hand to hold the antenna in place while I attach the nuts on the studs from the inside, so I should try to get the installation of both antennas done at the same time. Also, I am out of star washers which I will need to put under the nuts.

Plumbed Up

With another day off from work after Independence Day, I used the cooler (or better: not so hot) mornings to continue work on my RV12.

First I replaced the plastic lock nuts on the throttle cables with ones from metal - MS21042-3 to be precise. This is the better lock nut in hot environments and the engine compartment is certainly such an environment.

Then I had to reroute the flex hose that runs fresh air to the voltage regulator. As you can see the way I had routed the hose before, prevented the air filter from sliding on.

Routing the hose so it would run under the water hose in this critical area did the trick.

While working on that water hose, I noticed a possible chafing area between the hose and the forward motor mount as it touch it. I put a cushion in form of a leftover piece of that water hose over the motor mount and tied it down.

Then I safety-wired both air filters.

Considering how tight it is around those air filters, there is literally no way how they could ever fall off and away from the aircraft but it's easier to do now than during certification when they might require you to do it anyway.

So, finally I felt like the engine compartment was good enough to be left alone for a while. The next step was to open up the avionics cover and finish the plumbing of the Pitot line and some fixes to the wiring.

Thanks to the Stein quick connectors, the plumbing was quickly done - including the splits for the hook-up of the additional steam gauges that I will add after certification.

This is the T for the static line which will run to the ALT and ASI instruments

This T is for the Pitot line that runs back to the ADAHRS and the ASI

The open ends are plugged for now and both pneumatic systems should be ready for testing now.

In this VAF article I had learned that there was a wire missing between the Garmin radio and the AV-50000 black box that made it impossible for the SkyView system to read the currently selected frequency from the radio. The rest of the wiring between the black box and the SkyView system was in place however, so it appeared that adding the missing wire should fix this little "problem".
It is indeed a bit tricky to add that wire as space is extremely limited but others had tried to use a pair of needle nose pliers and succeeded without removing the radio tray.
So I followed those tracks to do the same.

With the help of a small mirror and my smallest needle nose pliers I was able to insert the female socket into the plug (pin 3, red wire is pin 1) without removing the tray.

The question now just is, if I used the correct size socket. I used one that is for male .040" D-sub pins which is on the other end of this wire and hooks up to the AV-50000's COM/NAV port (pin 6, red wire indicates pin 1). I put this wire together from two cut off pieces from harnesses that were not used anymore. So the wire that is yellow on the radio side turns white/orange on the AV-50000 side.

I was even able, after years of pulling my joints out of their sockets working in all the tight spots on the plane, to figure a way to get a couple of tie-wraps installed to secure the wire behind the radio tray.

By then it was approaching noon and the temps their usual uncomfortable level of 100 or more and I decided to seek shelter from this searing heat inside.

I think the additional antenna installations might be next. At least the small ones for the ADS-B receiver and the PCAS. I might leave the APRS one off for now as it is really large and very visible.

Avionics Tested

Before continuing on the Avionics installation and working off my to-do list from yesterday, I scrounged up a piece of aluminum I had bought quite a while ago. It's the POH holder that goes under the panel base for the D-180 dual display variation. I think it would perfectly work with the map box to keep it available for other things. That's why I bought it back then and fortunately Tony T. had provided me with some photos from the assembly instructions.
I cut out the little stop angle and deburred the part. I also verified that my panel base has all necessary holes already pre-punched, so the installation will be a quick riveting job only.

I held off on the actual installation as this one is easy and I want to make sure I am not creating obstacles now that I will later curse at when doing trickier jobs deeper down in the bowels.

Back to the Avionics. The first task was installing the fuel pump switch in preparation for powering up the Avionics system. I quickly determined that there is a lot more to real experimental aviation than just finding a nice position on the dashboard and drilling a hole for a switch. Maybe I should have kept in mind that a switch has a body attached to it that should also fit in the space allotted?

Fortunately, I was so far off, that I could reposition the hole and open it up without breaking into the previous hole. The way I intended to install it quickly wiped out any trace of my mishap.

This switch is on when the red protector is pushed down. You can turn the switch to off when you open the protector. So on my pre-flight list there will be an entry saying "Check fuel pump switch cover down".
I then routed the EMS wire from the switch to the EMS plug and decided to connect it to the General Purpose input next to the stall warning, which was pin 10.

I also did the full connection of the optional GPS mount harness, but I forgot to take photos. I connected the harness following the instructions on 42D-33 and just used the wires I had to make the connection to the plug.

And then.... I powered the thing up. Using a 30A switching power supply I had lying around, I connected a thick wire from the 12V terminal to the plug in the back of the Switch Block. This one would usually come from the relay on the front of the firewall and be switched by the Master switch. Without all this the system will come to live the moment you turn on the 12V.

The GPS power supply was obviously working. Zoom in and notice the plug symbol in the top status bar. In order to work, the Avionics switch must be on and you will have to add a fuse in the otherwise open slot for GPS/ADSB. I used a 3A fuse as this is how Garmin had fused it too.

What I found very odd is that Van's wants you to turn on the Autopilot switch to provide the ADAHRS and the servos with power. I wouldn't think of the autopilot if all I want is airspeed, pressure and temperature. However, with the SkyView Bus as the source of power for all attached devices that use that bus, I guess there has to be just one switch for all of this. I wouldn't call that one Autopilot though.

After the SV system had found all the hardware, I saw that the servos needed new firmware:

Fortunately, you don't have to download anything. The SkyView has the firmware already in it and can update the servos right away.

And after a short time of maybe 2 or 3 minutes, the system was ready to go.

Hey, if you ever felt that Van's instruction manuals are hard to read, get ready for an even better piece of literature! The Dynon Skyview Manuals easily outperform anything I have seen so far from the State of Oregon!

Debugging the radio problem I had posted earlier (PTT swap) took the rest of this day. I'm glad I can say that the system is ready for take off!
Now the rest of the plane has to follow :-).

I assume I cannot do very much over the next days while I am waiting for the 4" yoke and some other parts to complete the panel installation. I will probably go back to cutting the bulkhead cover and the floor plate. If there's more time to burn, I have some fiberglass tips for the rudder and the vertical stabilizer that need some trimming.
As soon as the panel is done, I'll clean the tunnel from debris, turn the plane on the belly again and roll it in the carport where the legs will get added. In the carport is the only flat area around where I can put the wings on and match-drill the flaperon brackets. The rigging of the rudder and stabilator will follow next.

WH-00036 Pins NOT Swapped!

This is an update to a previous post that did not contain the word "NOT" in the header.
When the electric diagram was released a guy on VAF found a discrepancy between the plans and reality and assumed that something was wrong with the wiring harness. After reading his post I checked the plans, compared reality and also found it at fault. However, when I fired up the Avionics today, I found that reality was obviously in order and the electric diagram must have a problem instead.
I refuse to dig deeper into the problem which is clearly a job Van's should do, but I traced it thus far: The Fuselage plug has two pins controlling the PTTs. Pin 1 should be connected to the Pilot PTT and Pin 3 should be the one hooked up to the Co-Pilot PTT. This is also what you can read from the electric diagram if you just look at the plug connection between the Fuselage pug and Van's Black Box.
If you follow beyond that point and route the connection to the actual switches, things get confusing quickly. It turns out that in the current plans the Pilot PTT from Pin 1 is effectively connected to the Co-Pilot PTT Switch. The same swap goes for Pin 3 which ends up at the Pilot Switch. However, the pins are indeed correct and what must be wrong is the naming in the electric diagram.
How I know? Well, I tested the radio and the Co-Pilot PTT triggered the pilot's voice to transmit and vice versa. I swapped them back and verified that now everything is working fine. The fact that the modulation on transmit is very distorted I contribute to the strong signal and the receiver being way too close to the transmitting antenna for now.

Avionics Almost Completed

With the current heat, I should have definitely started earlier. However, I was tired from the paying job and was working late to help with fixing a problem on the mountain where the telescope is located. So, I felt I needed my sleep and didn't get to work on the plane before 9am.
Last night I had tried to touch up the part of the center panel that looked a little funky after the previous paint application. I should know better never to touch-up flat paint as it just won't work. And it got confirmed again. The area that looks funky has now been increased in size but the initial problem has not really been improved.

Because of the location I decided against respraying the whole center panel. Instead I went ahead to use it and install.
I also know that this paint improves its sheen after two days of curing, so it might be less visible after that and in the end, I just really don't care that much about looks.

As you can see I decided to install the GPS mount before certification. Reasoning is that 42D explicitly gave me the option to cut out a section on the center panel to install my Garmin x96 GPS. Mine is a Garmin Aera 500 that has the very same interfacing that the x96 had. I will however, just put the mount on and not attach a GPS to the mount. Wiring of the mount can also be done exactly following the instructions in 42D, so I don't feel I am modifying anything here, just following the instructions and applying them to the case at hand.

For now I didn't connect the mount yet, as I will need some time to study the diagram and see where the best hook-up point is. I believe it is in the GPS plug but I want to make sure.
I went on to install the radio, or better, the tray for the radio. The tray is prewired with the Communication harness and only needs the antenna plug connected to it. However, this turned out to be more complicated than it read. The problem was that the routing of the antenna cable is very distinctive and after installing the tray there is hardly any room around the tray anymore. This thing really fills up the avionics compartment!
So, to pull out enough antenna cable to route it without bends, I had to cut a lot of the wire ties in the tunnel and make a medium mess down there.

Then I was able to locate the tray in its approximate position on the tray support angles.

Look how tight the wiring routes around the tray. This was taken before I collected some of the stray wires and tied them down to avoid chafing.

The screws went in ok, easier than I thought, actually - but be careful, read the instructions carefully and notice that the washers for the #6 screws at the center panel brackets got BETWEEN the tray and the bracket.
Sliding the radio in was easy and everything fit. However, it took me a while to understand how the locking cam mechanism works. Initially I had it tightened all the way and was wondering why turning the cam wouldn't lock the radio in the tray but bend the tray out a bit. You have to turn the cam counter-clockwise as you would unscrew a screw a bit before attempting to lock the radio in place. Then it will catch on the locking notch correctly and you can carefully tighten it a bit. Remember though, it doesn't take much to get 4 ft./lbs.!
When I wanted to install the left panel piece, I ran into some legacy problems. I must have accidentally picked the wrong nutplates when installing them in the left and right canopy ribs. I think the ones I installed where for AN3 bolts and it clearly says I needed K1000-06. Both nutplates are in one tray in my magazine, so I must have been tired when pulling them out of there. However, I drilled them out and installed the correct nutplates within a few minutes.

No big deal, everything was accessible.

And then, like magic, that left panel went in without any problems. It's amazing how easy things are when you just do it right...

I slid the SkyView in and added some blue tape to hold down the screen protective plastic and cover the buttons a bit from dust getting inside too easily. I used the deeper Fastenal screws that were mentioned on VAF a few days ago so to not round out the provided flimsy looking screws.
BTW: I should mentioned here, as you can see the front of the fuse panel, that I replaced the 2A starter fuse with a 3A because Bill H. had measured that the current for that relay is too close to the fuse value and it might burn it during normal operation.

I used the lower of the two SkyView Network plugs as the upper one is easier to get at for later extension if that would ever be needed or desired.

FlightCom 403, ELT remote control and the APDC button were installed, to complete the left panel installation. I ran the ground wire of the ELT to a corner screw of the Ignition module and installed the screw holding it down with Loctite blue. I appeared to be easier to do that running the short wire all across the other harnesses.

Looks really nice from the pilot seat too.

On to the right panel and the installation of the map box. I have the recommended 3" yoke for my squeezer and I had the desire to get a 4" one before but could resist it then. Now, I gave up. Despite my extra cutouts in the panel, I cannot reach 6 of the rivet holes and I HATE those CCR264 rivets. I do not want these being visible on my panel, so I ordered a 4" yoke now to squeeze all of the rivets.

I got enough rivets set to continue with the match-drilling steps for the two screws that connect the deep end of the map box with the panel base.

There were two holes already punched in the panel base that looked like #30. These are the ones you are supposed to open with a #19 bit in this match-drilling task. I used a wooden block to avoid drilling into my hand again and to safely hold down the map box during the drill process.

I deburred the holes and sat the right panel aside as I won't install it before finishing riveting the map box on.

This task concluded an 8.5 hour day working on the plane and the avionics are all installed. I need to hook up the two antennas on the belly to make it official but I can't do that before putting the plane on the gear legs.
I have yet to install the fuel pump switch and hook it up to the EMS plug, I also have to connect the power supply for the GPS mount and have to find a way how to temporarily provide current to the system so I can power it up and see if I basically get the correct responses. The servos and the ADAHRS should get detected, so does the GPS antenna. The radio could get tested with a temporary antenna connection and even the power supply for the backup GPS could get verified. The plug to connect the fuses to the battery seems to be part of the Powerplant kit because I haven't seen anything like that yet standing out.