1.5V LED boost circuit board project

[datiLED]

If anyone is interested, I have shared a fun little circuit board on OSH Park for driving a 5mm white LED with a 1.5V alkaline battery. The description in the link below includes a bill of materials.

https://oshpark.com/shared_projects/FGbApUVi

I have built several of these circuits, and they work very well. The through hole components are inexpensive, easy to solder and would be a fun project for flashlight enthusiasts or curious experimenters.


Top - Component side



Bottom - Solder side


Schematic of circuit


If anyone builds a few of these, I would be interested to see how you put them to use!

[Bimmerboy]

Very cool. Subscribing so I don't forget to ask questions later tonight... haha.

[Katherine Alicia]

I`v got 9 of these boards on order, it should go well with that bag of really nice Warm White 5mm leds I have

[gunga]

What host are you using?

[Katherine Alicia]

I have absolutely no idea yet! LOL I`ll probably just make one, or several different ones, that`s all part of the fun!

worst comes to the worst, there are some filthy cheap AA lights on ebay (as much as you can carry for a tenner type cheap!) and have a mess about with a few of those.

[datiLED]

I`v got 9 of these boards on order, it should go well with that bag of really nice Warm White 5mm leds I have

That should keep you busy for a while!

[Katherine Alicia]

yeah, it`ll probably take an afternoon to make them all up, then the fun starts putting them to use! I already have plans for a couple of Night Lights that use D cells for a good runtime.

[Lynx_Arc]

Reminds me of my early days of 5mm LED tinkering. I once got upset that there wasn't any AA LED headlamps at all on the market so I took a cell phone charger circuit I got for a dollar and swapped the inductor for a different one and got a board (perfboard w/lots of holes) and mounted 24 5mm LEDs on it wiring them and cutting the reflector of a cheap $2 2AA incan halloween clearance headlamp. It worked well but the LEDs were no very bright at the time and splotchy looking and over time they dimmed in use. I also modded a 2AA flashlight putting in a bulb I made with 3x5mm LEDs soldered to it and used a small boost circuit I got from somewhere. All of this was fun but then Dorcy came out with a 1AAA LED flashlight and people would just buy one of them for the circuit board and put it in other lights.

[datiLED]

Lynx_Arc, I have also built a lot of 5mm based lights. I remember the Dorcy AAA lights, and the mods to hotrod them. (I saved all of the schematics and tutorials for the different levels of mods, including the "high power" XR-E version.) Fun stuff.

[Lynx_Arc]

Lynx_Arc, I have also built a lot of 5mm based lights. I remember the Dorcy AAA lights, and the mods to hotrod them. (I saved all of the schematics and tutorials for the different levels of mods, including the "high power" XR-E version.) Fun stuff.

Yup..... I've been reading your posts for over 14 years now.... my tinkering back then was using whatever I could manage parts from junk and clearance items even stuff I pulled from dumpsters. Back in the early days 1.5v boost circuitry was nearly unobtanium with the exception of arc AAA lights there just wasn't any out there. A 1.5v boost circuit brings back early memories and what amazes me is off the alkaleak batteries back then that you got 15 lumens you can get 5 times that much output now. I'm guessing the latest 5mm LEDs are several times more efficient too.

[olo]

what amazes me is off the alkaleak batteries back then that you got 15 lumens you can get 5 times that much output now. I'm guessing the latest 5mm LEDs are several times more efficient too.

Keychain flashlights routinely get 100 lm running on a single AAA alkaline cell. They only stay in regulation for a few minutes, but it’s only because dc-dc converters are so inefficient. If only we had at least 85% efficiency instead of typical 65%...

[Katherine Alicia]

the QX5252 claims an efficiency up to 84% if that`s any help?

[Lynx_Arc]

Keychain flashlights routinely get 100 lm running on a single AAA alkaline cell. They only stay in regulation for a few minutes, but it’s only because dc-dc converters are so inefficient. If only we had at least 85% efficiency instead of typical 65%...

The problem is less about efficiency more about alkaleaks as the high current draw required to power an LED at 100 lumens is unsustainable by your average alkaleak cell. Put a nimh cell in it and it does 100 lumens just fine. Even at 100% efficiency and a 150 lumens/watt LED at 3V needs about 0.67 watts from the battery and that means about 220ma at 3v. If you have to boost 1.5v to 3v you need about 440ma from the battery and the internal resistance of the battery itself will drop the voltage more so that you may need close to 500ma from an alkaleak 1.5v battery and they just cannot deliver that amount of current for very long. If you toss a nimh cell in with a lot lower internal resistance that can handle several amps of current you find 120 lumen output possible.
Regulation can be a curse in a light that the battery struggles to keep up with the load on it as it caves in voltage due to its own internal resistance load the circuit wants to keep the output the same so the current draw is increased which at some point the battery cannot sustain the load on it and caves such that either the light starts flashing/strobing or totally drops out of regulation.

[olo]

the QX5252 claims an efficiency up to 84% if that`s any help?

It very well might be 84% for a specific V-in, V-out, and I-out combination. It's hard to tell how much it falls once you leave the sweet spot it was designed for. Just take a look at the charts on page 4 here. And that's one of the best converters on the market for driving med-power LEDs from a single-cell alkaline.

[olo]

The problem is less about efficiency more about alkaleaks

Duracell Coppertop AAA respectfully disagrees. It can sustain 500 mW output for well over an hour and 1 W output for close to 30 minutes.

Even at 100% efficiency and a 150 lumens/watt LED at 3V needs about 0.67 watts

Are you talking about high CRI LEDs? Because if you are willing to consider R70 you can get 220+ lm/W. Look at Nichia NF2W757H-F1. P13 rank gives you 42.8 lm at 65 mA and 2.72 V. If you drive it at 155 mA and 2.8 V you should get close to 100 lm at 435 mW. There you have it, regulated 100 lm for an hour from a single AAA alkaline cell. If only we could get 87% driver efficiency.

[Lynx_Arc]

From what I read the LED cannot be driven at 155ma 120ma is the max and likely the lumens/watt takes a hit at that output. Also the battery voltages drop pretty hard at a 500mw load and at 87% efficiency it is very likely that the efficiency of any driver drops considerably the larger the voltage in/out range is even if you got 87% at 1.4v it could drop to 75% at 1.2v and 60% at 1.0v and the driver would end up drawing increasingly more power in higher currents from the battery that the batteries internal resistance would count more in the equation.
I think that in the future when you see LEDs in the 350 lumens/watt range and drivers that are more efficient over wider range you may be able to do what you desire. IMO a nimh battery can probably do what you want to now as its lower internal resistance which also means more sustainable voltage under heavier loads keeps the boost driver in a more efficient range.

[olo]

From what I read the LED cannot be driven at 155ma 120ma is the max and likely the lumens/watt takes a hit at that output.

I checked and 180 mA is the max for Nichia NF2W757H-F1 (240 mA in pulse) and it can dissipate 540 mW. Both 155 mA and 435 mW are well within spec. There is a drop in efficiency at higher current, you are right, but it's barely noticeable.

the efficiency of any driver drops considerably the larger the voltage in/out range is

You hit the nail on the head. I wish I had the knowledge and the skills to build prototypes with the newer converters, like TI TPS61029-Q1.

[Lynx_Arc]

I checked and 180 mA is the max for Nichia NF2W757H-F1 (240 mA in pulse) and it can dissipate 540 mW. Both 155 mA and 435 mW are well within spec. There is a drop in efficiency at higher current, you are right, but it's barely noticeable.You hit the nail on the head. I wish I had the knowledge and the skills to build prototypes with the newer converters, like TI TPS61029-Q1.

I don't have the steadiness of hand to solder smd components myself so I typically just buy premade stuff. I've adapted guts out of cheap lights back in the day for boost circuits. One of the biggest problems I've seen with 1.5v boost circuits is the cut off voltage. If you used nimh you want it to cut off around 1v while alkaleaks you may want it to go down to 0.5v depending on the lumen output. I think my main "wish" these days is less about LEDs and drivers but rather battery breakthroughs something that is superior to lithium ion technology and perhaps something to eliminate alkaleaks totally with better price/performance.

[DIWdiver]

olo, I think you misread the graph in the coppertop DS. The 1W discharge curve hits 0.9V at around 16 minutes.

The TPS61029-Q1 DS doesn't have perfect graphs for seeing what it would do in this situation, but my interpretation is that it would start out with an efficiency in the low to mid 80's, drop rapidly to 80%, and end up down at 70% (that point is on the graph), at which point you'd be pulling 621 mW from the cell to get 435 to the LED. I would estimate that you should theoretically be able to get your 435 mW for a good part of an hour, say 45-50 minutes, but definitely not for an hour.

For comparison, the QX5252 DS has no mention of efficiency except to say "maximum drive efficiency can exceed 84%". Likely in this application it would be lower than the TPS, but that's not certain. I would say it's very unlikely to be better than the TPS part overall, and probably worse.

I checked out what NiMH should do. Discounting the BatteryGuy brand that claims 1800 mA-H, and PKCELL, with whom I am not familiar, the highest rated capacity was a cell from FDK (division of Fujitsu) at 930 mA-H. It's DS shows average discharge voltage (at 1A) is very close to the coppertop (at 500 mA) at around 1.2V. But the 1000 mA curve hits 1.0V (compared to 0.9 for coppertop) at 900 mA-H, or 1080 mW-H. That's double the coppertop's performance.

Unfortunately, it's really hard to acheive the datasheet specs when working within the constraints of a tiny flashlight, especially if the budget is low. The fact that it's bleeding edge tech combines with that to hamper you from obtaining the performance you find theoretically possible. Still, it looks like Linx Arc is correct that you should be able to get 100 lm for >1 hour from a NiMH AAA cell. It would only take >50% efficiency in the driver. That should be achievable with either of the chips mentioned in this thread. And if 1800 mA-H cells ever become real, WOW!

And FYI, we will NEVER see 350 lm/W on ANY device creating white light. The exact number depends on the color temp and other spectral characteristics, but it's generally agreed that the maximum possible for white light is around 270 lm/W. This is because 270 lm of any light that we would consider white (with no ultraviolet or infrared) contains around 1W of radiant power. Thus it would take 100% radiometric efficiency to reach 270 lm/W. It is physically impossible to exceed that no matter the technology.

It's true that you can exceed 270 lm/W, but only by adding extra green and/or nearby colors. Then it's no longer what most of us would accept as 'white'.

[olo]

olo, I think you misread the graph in the coppertop DS. The 1W discharge curve hits 0.9V at around 16 minutes.

You are absolutely right, I mistook the 750 mW curve for the 1 W one. Shame on me

The TPS61029-Q1 DS doesn't have perfect graphs for seeing what it would do in this situation, but my interpretation is that it would start out with an efficiency in the low to mid 80's, drop rapidly to 80%, and end up down at 70% (that point is on the graph), at which point you'd be pulling 621 mW from the cell to get 435 to the LED. I would estimate that you should theoretically be able to get your 435 mW for a good part of an hour, say 45-50 minutes, but definitely not for an hour.

I'll take 45 minutes. And maybe there are more suitable converters, who knows, I'm not an engineer

the highest rated capacity was a cell from FDK (division of Fujitsu) at 930 mA-H. It's DS shows average discharge voltage (at 1A) is very close to the coppertop (at 500 mA) at around 1.2V. But the 1000 mA curve hits 1.0V (compared to 0.9 for coppertop) at 900 mA-H, or 1080 mW-H. That's double the coppertop's performance.

I spent some time looking for a constant power discharge graph for any AAA NiMH - couldn't find anything. But based on AA graphs I remember seeing I wouldn't be surprised if NiMH has double Wh of alkalines at 500 mW. And I don't see any lights producing 100 lm for 1,5 hours on a single NiMH AAA, even though that should be easily possible, since NiMH spends more time at higher voltage, where converters are more efficient.

[Lynx_Arc]

One other forgotten thing is lights are typically rated at OTF lumens (out the front) vs Emitter lumens. I've read that from about 25% to 35% of the emitter lumens are lost in reflectors, lenses, and optics in LED lights. If you assume 30% loss to get 100 lumens out you would need about 140 emitter lumens. This is very substantial as most lights these days are ANSI rated and that is OTF lumens.

[olo]

One other forgotten thing is lights are typically rated at OTF lumens (out the front) vs Emitter lumens. I've read that from about 25% to 35% of the emitter lumens are lost in reflectors, lenses, and optics in LED lights. If you assume 30% loss to get 100 lumens out you would need about 140 emitter lumens. This is very substantial as most lights these days are ANSI rated and that is OTF lumens.

Good point. And while I doubt that batteries will improve by 20% in my lifetime, reducing 25% light loss to 5% is trivial - smartphone camera modules are like $25 and have glass optics with AR coating in addition to the sensor. So a good lens is like what, $10?

[Lynx_Arc]

Good point. And while I doubt that batteries will improve by 20% in my lifetime, reducing 25% light loss to 5% is trivial - smartphone camera modules are like $25 and have glass optics with AR coating in addition to the sensor. So a good lens is like what, $10?

I agree that it is unlikely we will see any breakthroughs anytime soon that will make it to market at a cost effective price as I've seen a lot of reports about battery tech in the last dozen years or so that never manifested itself.
As for the loss I was saying total not each item that was that much loss but the combined set of items with each contributing a portion of the loss and upgrading the items to better quality items may still end up with a 15% or so loss in lumens at best the only way around it is to do away with any focusing additions and no lens at all. Perhaps a lens of high quality AR coated etc only loses 5-10% I'm not really sure what the number is but I recall several people long ago replacing lenses and reflectors in Mag lights to get a few extra lumens but also to deal with heat from high output incan bulbs.

[DIWdiver]

I spent some time looking for a constant power discharge graph for any AAA NiMH - couldn't find anything. But based on AA graphs I remember seeing I wouldn't be surprised if NiMH has double Wh of alkalines at 500 mW. And I don't see any lights producing 100 lm for 1,5 hours on a single NiMH AAA, even though that should be easily possible, since NiMH spends more time at higher voltage, where converters are more efficient.

I didn't look beyond that one DS for a constant power curve, because the one I wanted had to lie somewhere between the 1000 and 200 mA constant current curves, and those aren't that far apart for NiMH. For alkaline they are dramatically different.

At fairly low currents, the alkaline actually has higher capacity. At high currents, considerably lower.