For the LED experts, how do the 1-cell LED flashlights work?

they are very simple the led runs on 1.5 volts and that's exactly what a d cell is ok!

Alan

A white LED runs at 4-5V, so the flashlight would have to up the voltage using a simple boost circuit.  This circuit may not be visible because there are "no user serviceable parts" in business end of Maglight. 

rrman posted:

A white LED runs at 4-5V, so the flashlight would have to up the voltage using a simple boost circuit.  This circuit may not be visible because there are "no user serviceable parts" in business end of Maglight. 

That is possible in the case of the Mag-Lite, I have never seen one in person.

However, the Eveready definitely has no electronics, unless they are in the base of the bulb! I did a quick ohmmeter check and the resistance is polarity-sensitive but a standard diode range on a multimeter won't turn it on. I will put it on the bench tomorrow to see if I can learn anything; if it does contain a step-up converter I would expect some radiated or conducted noise at the oscillator frequency. There is sure not much space in that base, though, unless they are doing the step-up at a high frequency with a voltage multiplier or something like that with no inductor. And virtually no cost. Or else they have come up with a new technology LED that really works at 1.5 volts.

Lumens per Watt. 

Your MagLite single-AAA battery version provides 40 Lumens for 2 hours with a 1.5V Alkaline AAA.  An alkaline AAA is, say, 1 Amp-Hr.  So that's ~60 Lumens/Watt which is a credible for consumer LED technology.

I could not find specs on your Eveready single-D but found a Ray-O-Vac single-D flashlight providing (only) 9 Lumens but running for 160 hours.  An alkaline D is, say, 17 A-Hr.  So that's also about ~60 Lumens/Watt (imagine that!).

Are you really asking how does a white LED operate at 1.5V with NO electronics?   As RRMAN says, are you sure there are no electronics?  You can make a very compact circuit to step-up a 1.5V battery to the voltage that drives a typical White LED.  Or, like a blinking LED all-in-one package, you can embed more than the light-emitting part into an LED.

As I see it, for a flashlight which uses a generic alkaline battery which droops in voltage essentially as soon as you put it into service, you want some kind of electronic power management circuitry to regulate power into the LED as the battery voltage drains/droops.

I'd be amused to see the business model of a technology company that thinks it wise to develop a White LED that intrinsically operates between 1.5V down to, say, 1.0V to put in a flashlight driven by a single-cell (1.5V) battery.  Just my opinion, but I would not buy stock in that company.

I'll bet money and give you odds that it contains electronics.  I love the costs of repairs.  The whole thing lists for $20, but if you happen to need spares...

$12.50 for a switch repair kit?  I suspect we've found the electronics.

stan2004 posted:

Lumens per Watt. 

Your MagLite single-AAA battery version provides 40 Lumens for 2 hours with a 1.5V Alkaline AAA.  An alkaline AAA is, say, 1 Amp-Hr.  So that's ~60 Lumens/Watt which is a credible for consumer LED technology.

I could not find specs on your Eveready single-D but found a Ray-O-Vac single-D flashlight providing (only) 9 Lumens but running for 160 hours.  An alkaline D is, say, 17 A-Hr.  So that's also about ~60 Lumens/Watt (imagine that!).

Are you really asking how does a white LED operate at 1.5V with NO electronics?   As RRMAN says, are you sure there are no electronics?  You can make a very compact circuit to step-up a 1.5V battery to the voltage that drives a typical White LED.  Or, like a blinking LED all-in-one package, you can embed more than the light-emitting part into an LED.

As I see it, for a flashlight which uses a generic alkaline battery which droops in voltage essentially as soon as you put it into service, you want some kind of electronic power management circuitry to regulate power into the LED as the battery voltage drains/droops.

I'd be amused to see the business model of a technology company that thinks it wise to develop a White LED that intrinsically operates between 1.5V down to, say, 1.0V to put in a flashlight driven by a single-cell (1.5V) battery.  Just my opinion, but I would not buy stock in that company.

Hi Stan-

Well, my original question was not rhetorical, but technical. As I see it, in the Eveready case, at least, the step-up circuitry must be built into the lamp base. I note that Eveready no longer lists ANY incandescent flashlights on their website!!! So, it is either a step-up in the lamp, or a true 1.5V LED - I vote for the step-up.

http://www.eveready.com/lighti...handheld-lights.aspx

Was thinking how handy this could be for GRJ and others adding directional headlights to locos. Almost nobody would miss 1.5 V in a motor circuit. Unfortunately, I have not seen any of these 1.5 volt lamps available for separate sale. The lamps in my Eveready flashlights are marked "KT-LB1.5V," but a Google search on that number comes up empty. 

I will try to put these on the bench tomorrow and get some electrical parameters, or maybe some insight into what is going on inside the lamp. Best guess would be a free-running oscillator in the high kHz range driving a Cockroft-Walton voltage multiplier; that could be built with no inductors, and chip capacitors at that voltage can be tiny indeed. 

Stay tuned.

When I first started messing around with LEDs for my garden railway, I was buying sets of Christmas lights on sale after the holiday.  I simply cut them up into single LEDs and powered them with batteries.  No resistors, just two AA or AAA cells.  Surprising how long those batteries lasted.  

I've been wondering about this myself. Does it require a boost circuit to get a higher voltage, or has there been some advance in LED technology that allows them to run directly from 1.5 volts? Do some LED's have a micro-booster-circuit built into them? I have a mini-flashlight that runs off one AA battery but it doesn't readily disassemble for an autopsy.

I have to say that I think they're simply using an voltage conversion circuit.

White LED Inverter

Here's the only low voltage white LED I found, it's 2V: Lumex SSL-DSP5093UWC

PLCProf posted:

Was thinking how handy this could be for GRJ and others adding directional headlights to locos. Almost nobody would miss 1.5 V in a motor circuit.  

If tied to the motor circuit output, wouldn't the headlight turn OFF when stopped/neutral?

stan2004 posted:

PLCProf posted:

Was thinking how handy this could be for GRJ and others adding directional headlights to locos. Almost nobody would miss 1.5 V in a motor circuit.  

If tied to the motor circuit output, wouldn't the headlight turn OFF when stopped/neutral?

Picky picky!!! 

I am basically a Toy Train guy who is working hard to better himself. Don't discourage me!!  

stan2004 posted:

PLCProf posted:

Was thinking how handy this could be for GRJ and others adding directional headlights to locos. Almost nobody would miss 1.5 V in a motor circuit.  

If tied to the motor circuit output, wouldn't the headlight turn OFF when stopped/neutral?

Yes

Well, I would never have believed it, but it looks to be true.

I put the 1.5 volt LED bulb on the bench and made a few tests. I don't know exactly what is going on inside, but there is definitely an oscillator that runs a little over 300kHz. Attached screen shots show the input current for various voltage levels.

I don't have a way to measure lumens, but I can measure footcandles, which should be OK as long as the distribution pattern doesn't change with voltage.

1.5 volts - 108 mA -  100%

1.25 volts - 108 mA -  55%

1.0 volts - 56 mA - 20%

I may hook up a photodiode later to see if the light output is modulated.

Hard to believe they can put a step-up converter in the base of a plain old flashlight bulb, let alone sell the whole flashlight for $2, battery included..

Lumens per Watt.

Your Eveready datasheet link claims 9 Lumens.   You're putting in 1.5V x 108mA = 0.16 Watts.   That's 56 Lumens/Watt.  That matches the Maglite single-AAA and is credible in terms of present white LED technology  It also makes sense in terms of power since that's about 10 times better luminous efficacy than a typical incandescent flashlight and Eveready claims this flashlight lasts 8 times longer than an incandescent.

I'm not up on the latest and greatest in single-cell flashlight technology but a quick web search found this tiny 3-pin SOT-23 package that only needs an external inductor to step-up a single-cell 1.5V at the power levels you're measuring.

http://www.diodes.com/_files/datasheets/ZXLD381.pdf

With this design, the LED current is indeed pulsing and in this case about 350 kHz.

Very interesting.  Thanks to PLCPROF and STAN2004 for the info and references.

The ZXLD381 is a single cell LED driver designed for applications where step-up voltage conversion from a very low input voltage is required. These applications mainly operate from 1.5V or 1.2V cells. The IC generates constant current pulses that are ideal for driving single or multiple LEDs over a wide range of operating voltages. The ZXLD381 uses a PFM control technique to drive an internal switching transistor which a low saturation resistance. This ensures high efficiency, even for input voltages as low as 1V.

stan2004 posted:

Lumens per Watt.

Your Eveready datasheet link claims 9 Lumens.   You're putting in 1.5V x 108mA = 0.16 Watts.   That's 56 Lumens/Watt.  That matches the Maglite single-AAA and is credible in terms of present white LED technology  It also makes sense in terms of power since that's about 10 times better luminous efficacy than a typical incandescent flashlight and Eveready claims this flashlight lasts 8 times longer than an incandescent.

I'm not up on the latest and greatest in single-cell flashlight technology but a quick web search found this tiny 3-pin SOT-23 package that only needs an external inductor to step-up a single-cell 1.5V at the power levels you're measuring.

http://www.diodes.com/_files/datasheets/ZXLD381.pdf

 

Yes, I think you nailed it. I scratched around a bit and I see the price of the ZXLD381 in the single pennies range. I will need to sacrifice one of these bulbs to see what they are using for an inductor. 

I need to study up on this. I wonder if anybody is doing the opposite, putting a step-down converter in a 12 volt LED package to get rid of the heat dissipated by the customary dropping resistor.

There are blinking and flickering LED bulbs, no reason to believe they couldn't embed the boost circuit.

i just bought some bulbs for my car. My car is a couple years old and has an amber bulb (single wire, single filament) that is used for the directional and the DRL (daytime running lamp). The replacement bulb is the newer style white LED for the DRL but is amber for the directional. The bulb contains a circuit that knows which color to use.

I was googling for KT-LB1.5v and found this board...  To answer the $20,000 question -- YES.  It has electronic bits inside the base. 

On a 8mm x 12mm PCB, it has:

• pale green resistor with orange, white, blue, silver bands (in that order)

• small transistor or IC

• very tiny rectangle diode

The "pale green resistor" might be an inductor.  Hard to tell from the photo but might that blue stripe be black?  In which case that would be a 39 uH inductor.  The 3-pin IC discussed earlier in the thread has a suggested circuit for which 39 uH would fit.  The datasheet shows a hookup which additionally uses a diode and capacitor (labeled D1 and C1 on your board).

Thanks to MIDAS1X and stan2004 for getting to the bottom of this conundrum. For a guy like me who started out with octal tube sockets the extent to which passive components have been miniaturized is astounding. Back in the day a 39 uH inductor or a 1 uF capacitor was large, regardless of construction. Now you can do these in surface mount. Hard to believe. 

Yeah, I would say that is a 39 uH inductor. Something I have noticed more than a few times is that many commercial products are exact copies of the examples in the manufacturers data sheets. Just recently I had to repair a switching power supply in a guitar amplifier, sure enough, the power supply matched the application note for the control IC in every detail.

PLCProf posted:

Something I have noticed more than a few times is that many commercial products are exact copies of the examples in the manufacturers data sheets. Just recently I had to repair a switching power supply in a guitar amplifier, sure enough, the power supply matched the application note for the control IC in every detail.

 

Why reinvent the wheel?

I believe it is a black stripe, yes.  So, in essence, the LED core has a 4-5v range and uses this micro PCB to ramp up the voltage from a 1.5v D cell battery?  That is pretty amazing.

Midas1x posted:

I was googling for KT-LB1.5v and found this board...  To answer the $20,000 question -- YES.  It has electronic bits inside the base. 

On a 8mm x 12mm PCB, it has:

• pale green resistor with orange, white, blue, silver bands (in that order)

• small transistor or IC

• very tiny rectangle diode

 

 

Does this mean I can sacrifice a bulb to get the electronics without all that messy parts acquisition and board soldering?

The size of our electronic parts today is amazing to me as well. That small pcb in the LED is really something. 

The work very poorly in our experience.  They last about a week or two before developing a switch problem so they blink a lot or just wink out.  Probably why they cost less than $5 each.

rex desilets posted:

 

Does this mean I can sacrifice a bulb to get the electronics without all that messy parts acquisition and board soldering?

It depends what you're trying to do.  The workings of the IC chip mentioned above (if indeed that is what the flashlight board uses) gets into some nerdy details, but the chip does not actually generate a regulated voltage output.  It has a pulsing current output (see PLCProf's waveform pics) that is/was specifically designed to drive an LED's particular voltage-current characteristics.

If you're looking for a general purpose widget that converts a single-cell (1.5V) battery to a regulated 5V voltage output, consider a $1 eBay modules.  They're not as tiny but the price is right and, to your point, no messy parts acquisition and board soldering.