DIY Fiber Optic link

[mike50]

I've been working on converting an RCM PBY-2 that I built some 25 years ago to electric. On the PBY, the twin engines are in nacelles in the wing, which is mounted on a pylon above the fuselage. The battery won't fit in the wing, so it has to go in the fuselage, somewhat forward. This means I need a fairly long run of wire from the battery to the speed controls. I've done what I can to reduce problems associated with long wires between battery and ESC. I'm using 12-gauge wire and have mounted fairly large electrolytic capacitors at the ESCs.

I'm also concerned that the signal line from the receiver to the ESCs must be strung along the same route through the pylon as the power wires. It seems there could be a good chance of interference from the 40+ amps in the power leads.

So, I decided to make a fiber-optic link between receiver and ESCs. I have used the commercially available IF-E96 and IF-D95 fiber-optic transmitter/receiver pair before, but I thought I could build something a little smaller and lighter.

I know this is probably not really necessary, but it is fun to do and I wanted to try my hand at making some circuit boards using photosensitive boards and to try using SMT components.

I bought some infrared LEDs and infrared phototransitors from Jameco and came up with the following circuits:

[mike50]

I was able to produce some nice circuit boards, easier (though more expensive) than using the toner transfer method.

[mike50]

The trickiest part was coming up with a way to hold the fiber securely, and squarely against the end of the LED and phototransistor. What I finally came up with was a set of nested aluminum tubes (K & S Metals tubing from the LHS). The outermost tube's ID was just the right size to fit the "flange" on the LED and phototransistor's T-1 case. The middle tube's ID was just right to fit the 3mm T-1 body and the innermost tube's ID just fit the fiber's casing. I added a small piece of fuel tubing on the end, which holds the fiber nice and snugly.

[mike50]

Here are the final results:

The transmitter and receiver circuits each weigh about 0.1 ounces. The fiber is about 0.04 ounce per foot.

[bderksen]

Looks prettly slick. Have you had a chance to try it out, yet? Where do you source the fibre? Would be interesting to see if the electronics/optics affect the signal pulse-width at all.

Cheers
Bill

[mike50]

I haven't tried it in the air yet, but it seems to work fine with the testing I have done. I've run it with up to ten feet of fiber. Testing it with a servo, I haven't noticed any difference in centering or end points. I did notice that if I pull the fiber a little bit away from the LED or transistor, the servo center point moves, indicating that the pulse width is changing.

I got the fiber from Jameco, although this is very common, jacketed 1mm plastic fiber.

Mike