# do-it-yourself long burntime LED light

#### Coherence

##### Newly Enlightened
Here is an easy to do project:

Take out the electronics from a Brinkman LED light, was (still is?) available from WalMart for about \$12. You get a 2 transistor circuit on a board about 1"x1" that functions as a voltage step up.

here is what I measure:

Volts....mA
0.0......0
1.0......10 <-LED begins glowing
1.5......23
3.0......110

Even at 1.5 volts, the light glows fairly brightly, try it by removing one of the batteries in the Brinkman and shorting + terminal to - terminal (in the battery compartment, not the battery itself!).

At 3 volts it is really bright (can't stare into it). Of course with the circuit exposed it does not have the tight lens focused beam, as a bonus.

Okay I'll do the battery math, got the battery data from data.energizer.com

usable lifetime(in hours) = battery capacity (in mAh) / discharge rate (in mA)

for alkaline batteries @ 1.5 volt
AAA= 1150/23 = 50 hours
AA = 2850/23 = 124 hours
C = 8350/23 = 363 hour
D = 18000/23 = 783 hours = 32 days

I know that these times are based on cells getting down to 0.7 volt, so they are overly optimistic as the circuit ceases to function below 1.0 volt. However, this does not seem to be a constant power type circuit, as the current draw reduces as voltage is reduced. In other words, it will get dimmer as the cell is used, but stretching out the life of the cell.

If you want to drive the circuit at 3 volts then you will use 2 cells about 5 times faster than the above run times, the price you pay for high brightness. For instance, a 123A lithium cell (3 volt) will last about 1300/110 = 12 hours.

Okay, here is what I want if anyone knows how to do this:

make a 2 AA light where you can switch between using the batteries in series (bright mode ~25 hours) and in parallel (long runtime, ~250 hours), with no chance of one cell discharging through another while in parallel.
Is there an easy way to do this?

also if I made any math/reasoning errors someone correct me...

#### Coherence

##### Newly Enlightened
Yep, that is it. I just started a runtime test on one Energizer AA alkaline. Right now I rate it as about 1/4 as bright as a new Photon 2, somewhat dimmer than the PalLight on low beam. Will keep posting as battery dies (should be a few days of runtime).

#### Coherence

##### Newly Enlightened
I don't have the Satcure circuit, but even if I did I would be very reluctant to connect the input of it to the output of the Brinkman driver. These are not made to work together. I can easily imagine thermal runaway in such a setup, given enough power supply current. This means heat/smoke/melting/fire.

As I see it, the Brinkman circuit is the same charge coil / discharge coil as the Satcure. The output voltage is inversely proportional to the load. If the load has high resistance, the output voltage is high.

#### Coherence

##### Newly Enlightened
Hi KenBar,

>I connected the OUTPUT of the SatCure to the power leads INPUT of the Brinkman.

Yes, I did misread your post, should not matter though. The circuits function similarly, voltage step-up. Imagine two SatCures wired in series, output of one to input of other. Output of first one will be high voltage when discharging through a relatively high impedance load (input of second SatCure), and to complicate matters the output varies over time (oscillator). It is like connecting the output of a 100 watt stereo into the input of another 100 watt stereo. You may expect 200 watt output, but this does not happen, if you are lucky you get 100 watt undistorted output, if unlucky you fry the input stage of second stereo.

>I only have one goal. I want as much light balanced with as long a run time as possible with a D cell within reasonable means.

I like this goal. If you mean a light with only one D cell, one problem is that the step up circuits I have seen have trouble when the input voltage fall below 1 volt. Best I have actually measured is the module in the Matrix headlamp, which still keeps the LED lit at 0.2 volts, see this thread

#### Coherence

##### Newly Enlightened
As expected the Brinkman module ran on one AA alkaline for about 4 days bright, dimming on the fourth day (yesterday), voltage was about 1.2 volts.

Today (5th day)it is dimmer still, voltage at 0.92 or so. Not bad for one cell.

#### Coherence

##### Newly Enlightened
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR> Coherence, I am doing the same thing except with one D cell. It will take forever to go down I think.
Do you have any idea which of the 3 resistors to play with to lower the brightness? I don't need that brightness all the time. I hate to blow it as I only have one module and the Wallmarts that I can go to are out of them. <HR></BLOCKQUOTE>

Well, if you are only using one cell = 1.5v you should not have to worry about the module... normally in a Brinkman light it is running at 3 volts.

Would be nice to be able to vary the brightness though. I just poked around with a multimeter measuring the voltage across the two resistors (one of those resistor looking things is a coil, is is fatter and on the board underneath it is printed an "L", see the bottom of your photo).

R1 = orange blue red gold = 3.6 x 10^2 ohms = 360 ohms
I measure 0.2v across it. The LED noticeably brightened while taking the measurement.

R2 = brown black orange = 10,000 ohms
I measure 0.6v across it. The LED dimmed while taking the measurement.

These voltages are probably meaningless since this is not a DC circuit, but at least it shows you can alter how it works.
I have a cheapie meter rated at 20,000 ohms per volt, which is why it changed the circuit enough to affect it.

If you want to just play with it, stick some (high value at first) resistor in parallel with R1 or R2 and see what happens. Use a 1.5V source with a current meter. If any of the following happen, quickly disconnect the power source:

current goes way up
transistors q1 or q2 gets hot to the touch
led gets really blue, then dimmer, then... (yes I've done it - not with this circuit though)

On another note, have you considered using two cells wired in series (3v)? Just connect the white LED straight to them and you have a super simple long runtime light.

Just measured the current draw - get this: 1.7 mA!

Okay for laughs here is the battery lifetime 2D alkaline

9000 hours = 375 days!