6v LED brighter with 3v battery

[foggs]

I have a cheap headlamp that came with 2 CR2016 (3v) batteries in series (6v).

If I replace them with a single CR2032 (3v), the LED is significantly brighter (3000 lux vs over 9000 lux!)

The CR2016's are a little older but i measured all batteries to be very close to 3v.

Why is it doing this?

 

[Externet]

Measure cells voltage when the led is on.
CR2032 has more current capability; lower internal resistance.

 

[foggs]

Measure cells voltage when the led is on.
CR2032 has more current capability; lower internal resistance.

1 x CR2032: 2.98v (off) -> 2.82v (on)
2 x CR2016: 5.98v (off) -> 2.63v (on)

Interesting! Why does it do this?

 

[BobK]

Because they are not new, perhaps?

Bob

 

[dave9]

1 x CR2032: 2.98v (off) -> 2.82v (on)
2 x CR2016: 5.98v (off) -> 2.63v (on)

Interesting! Why does it do this?

Because when the circuit is trying to pull more current than the battery internal resistance allows, you get voltage droop.

The real question is what the LED specs allow for max current, which may also depend on how well it is heatsunk (or in this case, heatsinking an encapsulated bi-lead LED can depend upon a short LED cathode lead soldered to a good amount (thickness) of copper wire or PCB layer.

It could be that the designers knew the 2 x series CR2016 power supply was inherently current limiting so they didn't (need to) design for higher LED power, so driving the LED at higher current could overheat it.

To clarify it is not a "6V LED", it's (presumably white) 3.(n)V LED but the tiny coin cells are acting as a current limit, plus as already stated, older cells already at a lower voltage.

If you power this for long from the single CR2032 cell, you might find that the LED more quickly gets dim and a significant % of the CR2032's capacity is wasted because the partially discharged voltage plus voltage droop is below the forward voltage of the LED. CR2032 cells aren't "good" for much more than 20mA if that, driving a white LED.

 

[hevans1944]

You should compare this datasheet for CR2016 with this datasheet for the CR2032 to gain some insight into what is going on. BTW, neither cell was designed to power the relatively high current demands (20 mA and up) of LEDs. The fact that they will work in that application simply lines the pockets of the battery producing companies since you will be replacing the cells quite often. Take the SWAP (Size, Weight, And Power) hit and use a cheap alkaline cell, or better yet, a rechargeable battery. Size matters, but coin cells for flashlights? Well, maybe to find a keyhole in the dark, but you might as well throw yer cash at hearing-aid batteries!

 

[foggs]

On a somewhat related note, I replaced a CMOS battery (CR2032) on a motherboard, that was flat (the BIOS no longer retained settings).

I then found my old analog battery tester which measures button cells with a load, and it read pretty close to 0, despite measuring 3V on the multimeter.

I then tested the CR2016's from my headlamp, with a similar result.

Moral of the story: don't rely on volts to give an indication of remaining capacity on lithium button cells.

So I'm going to get 2 new CR2016's and see how the brightness compares to the single CR2032.

The fact that they will work in that application simply lines the pockets of the battery producing companies since you will be replacing the cells quite often. Take the SWAP (Size, Weight, And Power) hit and use a cheap alkaline cell, or better yet, a rechargeable battery. Size matters, but coin cells for flashlights? Well, maybe to find a keyhole in the dark, but you might as well throw yer cash at hearing-aid batteries!

It's a headlamp that I use around camp when hiking, so doesn't get used for very long. I'm more concerned about the weight. I have another headlamp that uses 3 x NiMH AAA batteries but it's much heavier.

 

[dave9]

^ I like headlamps with the battery pack in the back (of the strap on your head), more balanced and don't stick out as far. For shorter term use where high brightness isn't needed, I have an ~ 80 lumen, Olight I3S, 1 x AAA (NiMH rechargeable) with the pocket clip on it reversed so it clips onto a baseball cap.

The coin cells can be much less expensive to run if you buy a few dozen at a time from a seller on ebay. Local store prices for them in the US are highway robbery.

 

[Audioguru]

How many people buy 2016 coin cell batteries? Stores probably keep them for at least a few years then they are hardly ever new.
Sellers on ebay probably also keep very cheap no-name-brand ones for years.

 

[majoco]

Moral of the story: don't rely on volts to give an indication of remaining capacity on lithium button cells.

Story. A few (no, many!) years ago, we had an aircraft come in to the hangar with a pilot written-up defect "Port engine fuel pump fail". So the airframe engineers drained the fuel collector tank, removed the pump, replaced with new one from stores, test and still didn't run. Measure voltage at fuel pump plug with Fluke 77 DMM, 24 volts no problem. Engineers change pump again. Operations up on tarmac getting tetchy "where's our aircraft....". New fuel pump still doesn't run. Airframe engineers come into avionics workshop with sad story. Avionic engineer goes out to aircraft with magic test equipment - a 32Watt light bulb on two pins to fit fuel pump plug. Switch on fuel pump from cockpit - no light. Remove light - measure volts - 24 volts. Change cockpit switch and all good - fuel pump runs.
Moral of the story - always measure voltages ON LOAD!

 

[Audioguru]

I have a small flashlight given to me at a politician's park BBQ. It had one white LED and was fairly bright to see a lock for a key to enter. The flashlight got dimmer then quit. Inside was two tiny (coin) CR1220 3V Lithium cells in series and no current-limiting resistor.
So it used 3V to light the LED and the other 3V was used to try to push enough current.

The datasheet of an Energizer CR1220 3V lithium battery shows that it can provide twelve 2 seconds pulses per day at 2.8mA, for a few days. Then the two rare battery cells must be found then replaced.