Longest life from a micropuck, series or parallel?

shankus

Flashlight Enthusiast
Joined
Feb 16, 2003
Messages
1,472
Location
Mojave, CA
I am thinking of putting something together with a micropuck, and two AA cells. Would the longest battery life be with the cells in parallel, with their capacity at it's highest, or in series, with their voltage at it's highest?
See, I'm confused about how it operates, I guess. It will regulate down to 0.8 V, so 2.4 V would make that range longer, right? But and unregulated circuit would get more life from parallel configuration. /ubbthreads/images/graemlins/confused.gif
 
If it's a constant-current output, my guess is you would get better results from the higher voltage. As it is a "boost-type" switching regulator, the closer Vin is to Vout, the higher the efficiency. So therefore, series would be longer-lasting than parallel. This is all assuming Vout is going to be at least 3V, or else you'll blow something up.
 
Hard to say without trying, but my guess is parallel will give substancially more life, but lower light out.

The clue is the light output (LED) current drops as the batteries age.

Or so it seems to me.

Cheers.

Doug Owen
 
Well, the data sheet on the micropuck says it's a constant current device, 375 mA constant.

"This driver has been designed to provide maximum illumination to the LED while still mimicking the light dropoff of an incandescent bulb, which dims as the batteries are used up."

Is it maintaining the current, at the cost of the voltage? i.e. current remains regulated, but V out drops as V in drops?

And now that I think of it, why would you want to mimic an incandescent light's output? The purpose of regulation is to get constant light output, right? (along with using as much of the battery capacity as possible.)
So in this situation, would it be better to have a constant voltage regulator? (If I want constant light output until cells are drained?)

I was thinking of driving 9 white LEDs in parallel from one.
 
Because of the way LEDs are designed, If is going to increase with Vf, so the two are almost directly related. However, according to the Luxeon data sheets, Vf at a given current has a negative temperature coefficient which means constant voltage supplies can be dangerous. This is known as the thermal runaway effect. "Direct Drive" setups offer some protection, as you then rely on the battery's internal resistance to limit current. This is why you would not want to direct drive a setup with 3 D cells when it is designed for 3 AA cells; the D size cells have a lower internal resistance and can provide more current while maintaining voltage. A non-regulating step-up converter will simply reduce overall output power as its input voltage decreases. This design offers longer overall runtime but with no regulation, of course. A regulated design will draw more and more current from the battery supply (up to a certain point) which actually reduces overall runtime, but maintains a constant brightness throughout the discharge.

It all depends on what type of power supply you are using (regulated or non-regulated) and if it's non regulated, what kind of light output you want. My first post was on the understanding that the MicroPuck is regulated, which I hear now it is not...if you are interested purely in runtime, a parallel setup would work best with a non-regulated boost, but with less light output. A series setup would provide more initial light, with less runtime. A series setup is also the best way to drive a regulated power supply. So...it all boils down to, do you want more initial light, or is less light but longer runtime desired? Or, reasonable runtime with constant light output?

On the flipside, operating a buck-type regulator is far more efficient than a boost-type regulator. This is because the buck regulator merely switches current on and off through an inductor, it being used only to smooth out the pulses...boost regulators must "charge" the inductor by connecting it directly across Vin and then switching it in series with Vin so the voltage pulse from the inductor + Vin = Vout.

As for driving in paralell, be careful...variations in Vf from LED to LED can cause some to "hog" current, resulting in overdriven LEDs and reduced lifetime.
 
Cell impediances

While most of what's been said it true, the statement "the D size cells have a lower internal resistance and can provide more current while maintaining voltage" is not exactly true. Check the specs. Say <http://www.duracell.com/oem/Primary/Alkaline/alkaline_manganese_data.asp> .

The AA (MX 1500) is listed at 81 miliohms, the D (MX 1300) as 126. A 50% *increase*.

Or so the maker sez.....

Doug Owen
 
Re: Cell impediances

Interesting. That really doesn't make sense, given the larger plate structure in a D-size cell.

Does this mean it would be better to replace a single D-size cell with 3 AA cells in parallel? /ubbthreads/images/graemlins/confused.gif
 
Re: Cell impediances

MicroPucks do not appear to be true constant current drivers. As Vin goes down, Iout seems to decrease faster then Vout. (??) I hooked one with the Q1K Star it came with to a lousy 1.5V N cell (not as bright as w/3Vin) and it put a decent spot on the wall for a couple hours. The following many nights, I continued trying to kill that N cell. It was dim, but still boosting enough to light the LS after over 40 hours! I goofed and left it on, and it finally died over the next 18. Disconnected it, waited a minute, and guess what- lit again! They do suck a cell down to under .8V. Doesn't answer shankus's question, but I'm going to eventually build an N cell LS keychain light out of one.

Larry
 
Re: Cell impediances

If this stuff wasn't counterintuitive from time to time, it wouldn't be much fun.

I suspect that the key to the castle is that while the AA has less area, it also has much shorter path? The 114 miliohms for the AAA tends to support this, IMO.

Not that internal resistance is the only part of it, either.

None the less, them's the facts......

Cheers.

Doug Owen
 

Latest posts

Back
Top