Most lamp bases are very, very similar. My first instinct would be to just replace the lamp base itself. It is hard to tell from the picture if it mounts like a standard base, but it likely does.
Do you have other overhead lights in your garage? Maybe putting a motion sensor or other switching mechanism might be an option if you can’t just replace the base.
The fixture is a sealed unit, and the lamp base is molded into the enclosure. Clearly not designed with replacement in mind. 
I do have a couple of fluorescent shop light fixtures I swapped in in place of the smaller incandescent fixtues the garage came with. Maybe just put a motion sensor in place of the current wall switch?
That’s what I was thinking. If the switch is in an appropriate place, then this is an easy solution that actually serves multiple purposes. It would come on automatically if you come in the walk-in door also (if you have one).
Yep. The light switch is right there by the door from the garage into the house and has clear line-of sight to both garage doors. So this may be an ideal solution. Probably a whole lot cheaper than replacing the fixtures with identical parts (even if I could find them), and definitely lower risk.
While I really don’t think the repairs I made are particularly risky, and I’m pretty comfortable working with AC, having installed or replaced many outlets, switches, and fixtures over the years, I’m nonetheless happier with a solution that reduces risk while providing me an even better outcome, namely more reliable lighting.
Once again, the TinyCLR community comes through with great advice. 
If you want it driven by the opener, and the opener uses IR to drive it, then a simple IR phototransistor + comparator + transistor + relay setup ought to be able to replicate it.
Charging $100 for that is highway robbery.
The opener is probably encrypted RF. Would probably be easier to detect current draw than the signal. Personally, I’d go with the $15 motion sensor switch replacement.
@ ianlee74
So, next question…
Does this seem like it’d be the right choice:
The concern I have is that it looks like the way this one detects motion is by a heat signature against the background, which might not allow it to work when coming home in the car…though now that I think about it, it’d probably pick up the heat signal from the engine, so perhaps I’m overthinking this.
Eh…it’s only $18, so worst case, I’m out a few bucks. Worth a shot.
Yep. That’s what I had in mind. You can pick them up at Lowes or Home Depot for about the same price. You might get lucky and it’ll also detect the temp difference in the concrete as soon as the door starts opening. You can always take it back if it doesn’t do what you want.
Ordered it from Amazon…and they conveniently offered the faceplate as an add-on item, so free shipping on the whole thing.
Will post a follow-up with how it works out…arrives Wednesday, and shouldn’t take long to install, if all goes well.
Why not a simple reed switch + magnet mounted on the rails and the door?
That might work, too, but would require a good deal more wiring, for one, and would require one switch for each door.
It also would not solve the use case of someone entering the garage from the door to the house.
The motion sensor switch, assuming it works as I hope it will, covers both scenarios, and is a pretty simple wiring job. Doesn’t hurt that it’s comparatively cheap, and I already know how to install it.
Following up on this. Epic ball of FAIL on the motion sensor switch.
Hoping maybe my esteemed colleagues might have some ideas on troubleshooting before I send this puppy back to Amazon. If you take a look at the image here:
the switch I got is the one on the right.
The switch has four wires coming out:
[ol]Black (hot)
White (neutral)
Red (load)
Green (ground)[/ol]
There are 3 wires in the box where I’m installing the switch, black, white, and bare copper.
I’m fairly certain that I’ve wired the switch correctly per the diagram, with black to black, green to ground, and white to neutral and load. I also used an outlet tester to make sure that the wires in the box were actually what their color indicated as well, and they tested fine.
But when I put the switch on “Auto” (the setting where the light should be turned on and off by the sensor), it simply clicks on and off repeatedly until I either switch it to the on or off position.
This switch is supposed to be able to handle both fluorescent and incandescent lights, and I’m using it on a circuit that has two 3’ dual-tube fluorescent shop lights, nothing fancy.
Has anyone messed with these switches before and seen something similar? I noticed that there were a couple of other folks who’d reviewed the switch who seem to have run into this, but I’m wondering if that means that every so often there’s a bad switch, or there’s something in common that we did wrong, or some other commonality that might help me troubleshoot.
Thanks in advance for any suggestions!
More info…looking like I didn’t do enough research on the switch I needed. Apparently there are both 3-wire and 4-wire variations on the occupancy sensor switches, and I ended up with the 4 wire, which requires a separate neutral, which unfortunately doesn’t exist for this switch.
Sigh.
So, seems like this one won’t work. And there’s some question whether or not the switches that don’t require a neutral will work with my lights. It’s very clear from the reading I’m doing that the 3-wire switches won’t work for CFL, due to the electronic ballast. What’s less clear is whether they should work with shop lights.
The saga continues…
Closing out the thread (hopefully)…picked up one of these today:
2A was just right for running two 0.8A work lights, wiring could not have been simpler. Works great out of the box.
May tweak it a bit to prevent it coming on in daylight, but otherwise it seems great.
Now to remove the troublesome wireless fixtures, and I’ll be good to go, and will be able to sleep soundly.
Thanks, all, for the comments and suggestions.