Sunday, May 22, 2011

The last you'll reed on the subject of reed switches

I can understand the perceived need for some kind of electrical interlock system for these wing pins. What I can't understand is why Van's chose this inopportune moment to stray from the KISS Principle. If it was good enough for Kelly Johnson, it's good enough for me. Magnetism is fine for most of the things it's used for, but I think this particular application was too clever by half. This really should have been done using the same type of mechanical switch used to trigger the stall warning.

But as I noted before, one must press on in the face of magnetic adversity; I had to get these things working. I diddled around with the positioning of the little reed switch, I fiddled around with the positioning of the wing pin. I eventually let the wing pin plunger snap back too quickly and the epoxy that was holding the magnet inside the plunger gave way and the magnet came out. This was actually a good thing - the indications that I was seeing on the multimeter made it look as if it was possible that the magnet wasn't all the way down at the bottom of the plunger.

I decided that rather than just gluing the magnet back in, I'd put the pin in the plane and move the magnet around inside the plunger to see if I could find a positioning that worked. I cleaned up the parts (remember that I had wiped a little grease on the plunger to help it move more freely in the wing pin) and carried the parts over to the plane. I put the wing pin in. Nothing on the multimeter. I pulled it back out of the hole to do the simple always-works test to make sure the wires were all still correctly attached to the multimeter.



I checked the wires and they were all fine. I checked with the other wing pin and it worked too.


I decided to pull the magnet back out of the plunger and try using it directly against the switch.

No magnet.

Uh-oh! I'd lost the magnet!

It couldn't have gotten too lost though; I'd only gone four or five feet from the work bench to the airplane - that should be a pretty manageable area.

Twenty minutes. It took twenty minutes to find that magnet. I was at the end of my rope, convinced that I had looked everywhere that the magnet could have gone.

Then I turned over my drill.

With that problem solved, I returned my attention to the bigger issue: the switch still wouldn't work. I had been shaving material off of the wing pin plunger tube in tiny little increments up to this point, but this was the last straw. I decided to grind away the full 1/8" allowed to me by law.

That fixed it. I did the same to the left side, and it worked too. I'm not sure why I didn't just do that in the first place.

The next page was the installation of the autopilot disconnect switch. While I may not have held the wing pin reed switches in very high regard, I have the utmost respect for the AP disconnect switch. Should the autopilot ever decide to play the fool and run away with the control of one of the flight controls that I feel it is my right as pilot to always have the final say on, I want this switch to put such an affair right. This switch is so important that the autopilot won't even engage if it isn't working correctly. It is this switch that cost me a few trips to Lancaster to try to debug an autopilot problem on a friend's RV-12. The first few hours of that job were spent trying to prove the switch was actually, well, switching. After that experience, I thought it prudent to test the switch before installing it.

It's good!

There's a tiny pilot hole in the panel that has to be reamed out to half an inch. That's just the job for the Uni-bit. With the morning that I had already had, it came as no surprise when the battery on the drill died on the very last step of the Uni-bit.

Once I got just enough charge on the drill's battery to finish the hole, I installed the switch. The wiring of the switch is a little tricky. It has four wires coming out of it; two of them are pretty short. Those get crimped into a splice with a couple of longer wires. Their limited length is such that they barely make it though the avionics shelf bushing. It's a good thing I left room in the aft bushing!

The other ends of the splices get long white wires crimped in, and those wires get fed all the way back to the wiring duct. I don't need those wires - they're for the optional lighting kit that I won't be buying. They provide power to the cockpit lighting that I won't have.  At least not for $1,300 anyway.

"Say," you're asking yourself, "what does the autopilot have to do with cockpit lights?"

A fair question. As I understand it, the Dynon screen that will provide most of the flight and engine management displays has an auto-dim feature that causes it to reduce its brightness in the dark so as to not ruin the pilot's night vision. The autopilot disconnect switch has a little red light in it that will also dim when the Dynon dims. It knows to do this, apparently, from the two colored wires that are spliced in. The white cockpit light wires are also merged into this circuit so that when the Dynon dims, so do the cockpit lights.

Clever, that, and I'd be enormously impressed, if I was buying the lighting kit. I may change my mind someday, so it makes sense to run the wires now. I'll just coil them up somewhere out of the way until such time as they're needed.

The next step of the autopilot disconnect switch is to plug the other, much longer two wires into the big DB-25 connector that will itself plug into the avionics control box. These wires are what were ultimately found to be the problem on that other RV-12 - they were plugged into the wrong holes. That was unfortunate because it required the use of a pin extraction tool to get them out of there. Not knowing how to use a pin extraction tool, I made myself scarce once the need for said tool was discovered.

I swore that I'd get these wires in the correct holes so as not to ever need to figure out how to use a pin extraction tool. The black and white wires on the outside of the black shrink wrap are the wires in question. I'm nearly 100% certain that I got them in the right holes.

"Wait, did you say outside of the shrink wrap??"

Well, yes. Yes I did.

So, here's what the pin extraction tool looks like in use:

And here are the wires correctly installed and inside the shrink wrap:

And here is the shrink wrap, shrunk:

Shrink wrap is pretty neat stuff. You just apply some heat (I have an electric heat gun, but a blow dryer would work) and it tightens up into a nice protective tube.

Which then gets installed into what's called a clamshell or backshell:

Next comes the installation of the Outside Air Temperature (OAT) probe. This probe will tell the flight computer what the temperature is outside of the airplane. That will be used by the computer to calculate true airspeed. In the world of GPS, where we know to a tenth of a mile per hour how fast we're moving across the ground, true airspeed is actually just shy of useless when it comes to any kind of utility in flying, but knowing the temperature outside of the airplane is useful in other ways.

What's notable about this installation is that this is the first time that I've ever had to drill a hole in the plane with nothing more to guide me than "put it right around here somewhere" directions. There has always been at least a pilot hole before.

Oh, also of note is that it's on the left side of the plane, which is currently another way of saying, "Oh, crap! That's gonna be a royal pain to get at."

It was.

With that done, I needed to take the rest of the day off to rest, by which I mean "mow the yard."

1 comment:

Leon said...

once again a most excellent blog posting. i have been putting together a blog for my upcoming travels. so far i have put up stuff mostly to make sure i understand how things work. it maybe just me but i have found it much harder and more time consuming than i would have thought. if it is not just me thanks for the time and work. you and a select few others(sippican cottage when he chooses to write is greatness) have really shown me the potential a blog can have.

Post a Comment