The power connector for my Toshiba R15 had begun to fray, so I decided that instead of just tossing it into a land fill, I would fix it up. Since I was going to have to repair it anyway, wouldn't a magnetic connector be better? I think that if I were to do it again, I would not make the dongle quite as long, but as it is, it breaks away with a sharp tug just like a macbook.
In the end, I have a very cheap repaired power cord with a lot more functionality!
Step 1: The Problem...
Below is the before and after. It's not especially pretty, but I am happy with the results, and if I were to do it again, I think it would turn out better.
Step 2: Decisions...
Later, I decided that an ambidextrous connector would be easier and more appropriate, and, in order to have the connector self correct for orientation, I switched to 1/4" square by 1/2" long magnets. by using two of these, the connectors want to connect in the right way, saving a lot of fiddling.
Step 3: Creating the Molds
The second picture is not quite accurate, as the brass anode terminals were added prior to pouring in the epoxy. The molds were made by using drill bits to hog out the main design of the connectors, the using an XActo knife to refine the design. As it turns out, I forgot to "refine" the outlet side connector mold, so it ended up just being a 1/2" hole flanked by 3/8" holes, but it turned out alright.
The magnets were soldered to a piece of 18 ga wire before being fitted into the molds. I left about a 1/32" around the magnets as possible to insulate them.
At this time I also decided I wanted cool green LEDs to indicate that power was connected (another prod from my wife!), so I fashioned up some LEDs from some SMD components I had lying around (eventually when I learn to program AVRs, I'll be using them for firefly projects). As you can see, they are easy to fit in, being so small. Initially I only soldered them the cathode ends to the magnets and left the anode lead sticking out slightly above the level of the epoxy. This was done so that I wouldn't have to disturb the brass terminals which were only tentatively held in place by being stuck into the wax. After the epoxy cured, I soldered a small jumper from the LED lead to the brass terminals. Had I thought it out, I would have made the leads longer and just bent them over to the brass terminal at this point.
Step 4: Soldering and Heatshrink!
I used a small bit of sculpting clay to mold the flat end of the connector to a more streamlined shape. This was then covered with heat shrink, layered in smaller diameters until it fit the wires snugly.
The same clay technique was used again on the outlet side connector, though not pictured. The clay was added in a cone just to the left of the soldered wires in the last picture, covering the already heat shrunk twisted wires. (Last picture. this soldering step was taken after the second epoxy molding as the positive terminals had to be added after the first epoxy mold had cured)
I used several layers of heat shrink to provide additional strain relief and ensure that the energized wires were adequately insulated. In the end it creates a nice finished cord look.
Step 5: Demolding and Remolding
I apologize for being remiss, but I seem to have forgotten to take pictures of the next step...
Once the epoxy had cured and the connectors were demolded, I soldered positive wires to the brass terminals and two two finishing nails that I cut to be short enough to fit entirely within the outlet side connector. I then drilled two holes on either side of the magnet on the outlet connector (the one without the LEDs) almost all the way through. I then drilled a small diameter hole just large enough to fit the nails, through the rest of the connector. The nails were inserted into these holes and end up about 5-7 mm from the end of the connector. This depth prevents the energized nails from accidentally contacting and energizing anything else. I felt this was important given the propensity of the magnets to stick to anything ferrous. While it would probably create a short and blow a fuse before starting a fire, I didn't want to take that chance.
Now the connectors are pretty much done, but all the positive soldered terminals (and the loose nails) are still exposed on the wire sides of the connectors. To cover these, I drilled a 1/2" through hole in the wax block and shaped the bottom side to fit the wire side (where I just soldered all the positive connections) of the connectors. I also did a little extra shaping of the hole to make the epoxy taper up to the 1/2" hole . The bottom of the wax block, with the two connectors sticking out (almost all the way) was then sealed with melted wax (or so I thought!), and epoxy was poured into the 1/2" hole on the top.
As it turned out, the holes for the two nails leaked like a seive and all the epoxy ran out the bottom. This also filled the holes I had created for the nail terminals :( It was a bummer. In the end, I just waited until the epoxy firmed up and then added it back in. When I demolded the connectors the second time I had to redrill the holes for the positive terminals, but in the end it worked out fine. I would suggest somehow sealing the nail heads with very thick epoxy prior to putting them in the second mold.
In the second picture you can see the 1/2" round of epoxy (just to the left of the note box) that covers the positive terminals on the back side of the connector. The computer side connector is similar, but does not have a clay cone to shape it down to the wires (clay cone not show for this connector, but is almost identical to the previous one). I'll live, though, and it provides more exposed epoxy for the LEDs to shine through.
Step 6: The Results
Again, the dongle is a bit long, but as you'll see in the brief video of the connector in use, it is still effective. The magnets are pretty strong, each exerting about 5 pounds, so you could still drag the laptop off the table if you pulled slowly, but then, there is no worry that the connection is just going to fall out. And, in reality, most of the accidents happen from sudden tugs rather than sustained pulls. And as you can see, sudden tugs barely affect the computer at all.
As a bonus, the LEDs work very well as they are super bright and are an effective indicator that the connector is supplying power to the plug (I did get caught on my initial tests in that I pulled the connector straight away from the computer and the plug disengaged enough that when I connected the magnetic connector, no power went to the computer. Not the fault of the connector, though).
In the videos you can see the LEDs extinguish when the connector is apart. You will also see how the connector attempts to align itself, and how, if it is not aligned, no current is flowing to exposed terminals. Safety first!
Thanks for taking a look.
Approximate cost for this project:
Casting Epoxy - $13
Candle Wax - $4
SMD LEDs - $0.26
SMD resistors - $0.18
Heat shrink - I have a bunch of this, but radio shack also sells some for about $3, might be cheaper to get it at Lowe's, though
18 ga wire - I already had it, but you could get a spool for $3 at radio shack
brass terminals and nails I already had these too, and most should be able to find something conductive that would work for free
Total cost (for me) : $17.44 and I still have more than 75% of both the epoxy and wax left!
This instructable was inspired by the Thinksafe instructable by "breath". Thanks for the idea!