Years ago my flashlight collecting consisted of whatever walmart and costco had to offer. This is an antique led lenser I picked at costco. It has now been resurrected, it’s the light source for my illuminated magnifier (taped to the light is a 10X eye loupe) . I use this thing at work at least 10 times a day.
:+1:
Very creative.
that looks like my fluke except it's clean and about 30 years newer .
Yep, that’s a new 87V, which is the handheld I use at work, and it is one heck of DMM. I have the original 87 I use at home, probably as old as yours. The only maintanence the old 87 required were new bananas and a new rotary switch wiper.
You just can’t kill these things.
Cheers
Doesn’t necessarily have to be an “old” light, but crappy 9-LED 3×AAA lights, I’ll pull the LED board out and toss the rest. They make decent 2-wire modules for throwing ugly blue light in dark dank dungeons and the like, where quality isn’t an issue.
A 10Ω–12Ω resistor is enough to limit current from 5V (usb) to 150mA–180mA or so. Higher ohmage and lower current if you don’t need them at full brightness (eg, lighting up a “no parking” or “keep off the grass” sign, etc.).
Thanks. I’m a pack rat. I can never bring myself to get rid of anything 
Thanks for the info.
The problem with current limiting resistors is that they’re inefficient. I’ve been using LT boost and buck converters. Working with LT parts is painless when the design is run through LT spice.
I appreciate your comments. You guys really know your stuff and being the noob I am I have learned a great deal here, so thanks!
+1…. I could not say it any better. 
For craplights like that, it’s overkill.
Each LED should draw no more’n 15mA or so, 20mA max, and they’re all in parallel, so who cares if you’re powering it from a USB charger and “waste” 1/3W or so? A single resistor is way easier to wire-in than going through the headache of designing and building a driver, and hoping it works.
Even 50¢ for a buck-board would drop the voltage, but not limit current. At least, not without a resistor! And chances are with current so low (160mA-180mA for the 9 LEDs), the power used by the circuitry would barely offset the current/wattage savings vs a simple stoopit resistor.
Think of it like shopping around for the best gas prices in the neighborhood. Go a few miles away, and you save a few cents per gallon. But if your tank only takes, say, 10gal on a top-off, how much did you really save? Sure, maybe a buk in terms of cash, but how much did you waste getting to the cheap station across town and then back? And the time/effort wasted planning it all out?
Not worth it, not to me. Bad enough I got too many projects stalled for one reason or another, vowing to roll my own as far as a fancy driver, vs throwing in a resistor whose value I worked out in my head? No contest…
I agree with everything you’ve said. The thing is, I do this stuff for a living, and I take it home, where I’m a hobbyist and and electronics enthusiast. I’ve probably powered an LED every possible way. At work I needed a blinking panel light indicator. I could have simply bought one and that would have been that. Way to easy. I wanted the blink of the LED to look like the blink of aircraft warning lamps on transmission towers. I ended up programming an ATmega 283 processor to output PWM to get this panel indicator blinking old-school. It was fun doing it, the panel indicator looks great, and technically it was productive work.
I take that mindset home where I play with electronics for fun. I’m thrilled when I can have a 3 watt low drop-out regulator in a SOT-23 package. Tiny, efficient and big-time fun. So yes, a dropping resistor is the easy way out, but it’s not fun to do.
There’s another problem with dropping resistors. What happens when the battery starts crapping out (even with the flat discharge curve of LiIon)? When the battery voltage drops a dropping resistor is a hindrance. A low dropout regulator would squeeze every ounce of useful energy from a battery. This becomes meaningful for anyone (myself included) who backs off the tailcap to avoid parasitic drain, regardless of what the parasitic drain actually is. I back off the heads of my Thrunites, even though the parasitic drain of my TC20 and Catapult V6 are around 40uA. It shouldn’t bug me, but it does.
My wall clock runs a year on a single AA cell. 15 years ago I bought a G-Shock watch. Moving hands and digital display. The darn thing is still running. I have a lab timer that I’ve had since I worked at Bio-Tek, 22 years ago. There are a pair of AA’s in there, and the darn thing is still working with 22 year old batteries. The LCD is barely visible, but it is still there ,I’ve been blinking an LED (LM3909. obsolete, but an awesome chip) once a second for over four years, running off a depleted C alkaline I rescued from the recycle bin. . And I should worry about .000045 A of parasitic loss? I shouldn’t, but I do. It’s not rational, but it is what it is
OOps… Sorry… Steps off soapbox now…
Cheers
Wellp, I did mention “from a usb charger”…
If you want best efficiency, just use a LFP cell, 3.2V, pretty much direct-drive. Flat discharge curve from out-of-the-charger to crapping-out-now, and a great match for most LEDs.
Yeah, the Pink Floyd album “Pulse” had a 3909 and small cell powering it, to blink an LED right in the ceedee cover!
Hang on… gotta be some mention of it…
Nyup, right here:
https://en.wikipedia.org/wiki/Pulse\_(Pink_Floyd_album)
“Early CD versions came with a flashing red LED on the side of the case. This was designed by EMI contractor Jon Kempner, who was awarded the platinum disc, using the now discontinued LM3909 LED flasher IC. The circuit was powered by a single AA battery; the battery life was stated to be over six months. Some versions were also made with 2 AA batteries and later editions of the CD set did not feature the blinking LED.”
I had a few dozen or more of those, as well as the usual 555s, 4046es, etc.
O my goodness…
The 3909’s can be had NOS but the last time I grabbed a few was on ebay where I paid like $4 each. Ouch.