High-powered LED Mag-lite Conversion

Step 9 Comparison to Standard Mag-lite

After some feedback, I decided to show how this light compares to the Mag-lite I started with. I've done some calculations from what I could find on the internet and here's the results:

A standard D-cell Alkaline battery contains about 12000 mAh of charge.
Battery Capacity

A 3-cell Mag has 4.5V...this means the whole flashlight has...36000mAh
A 4-cell Mag has 6V--------------------------------------------------------48000mAh
A 6-cell Mag has 9V--------------------------------------------------------72000mAh total

This means that theoretically, with a 1-amp load, a 3-cell should last 45 hours, a 4-cell 48, and a 6-cell 72 hours.

---

I looked long and hard, and found a site that sells Mag-lite replacement bulbs.
MagLite Replacement Bulbs

Here's an excerpt: (Keep in mind my flashlight is a standard Mag about 8 years old. I'm pretty sure it doesn't have a Krypton bulb.)

MagLite Bulb Type---Base Type---Gas Fill---Batteries---Max Peak Beam---Average---Max Batt
.....................................................................................................Candlepower----Lumens---Life Hours
White Star 3-cell------PR Flange---Krypton---3-C / 3-D---20,000 / 22,000------76.8------4-5 / 9-10
White Star 4-cell------PR Flange---Krypton---4-C / 4-D---24,800 / 23,000-----122.1-----4-5 / 9-10
White Star 6-cell------PR Flange---Krypton---6-C / 6-D---30,100 / 30,000-----162.6-----4-5 / 9-10

---

So using these Battery life ratings, we can find that:

Flashlight----Volts----Amps----Watts----Assumed Life
3-cell----------4.5V-------3A------13.5W------12 hours
4-cell-----------6V--------4A--------24W-------12 hours
6-cell-----------9V--------6A--------54W-------12 hours

Again, I'm not sure how accurate these numbers are, but from what I could find, these seem reasonable enough.

---

Using a really cool website I found, the power consumption of these LED arrays can be clearly found. LED series parallel array wizard

For the 3 cell:

4.5-----Source voltage
3.56---diode forward voltage
20-----diode forward current (mA)
12-----number of LEDs in your array

This yields the following information:

• each 47 ohm resistor dissipates 18.8 mW
• the wizard thinks 1/4W resistors are fine for your application
• together, all resistors dissipate 225.6 mW
• together, the diodes dissipate 854.4 mW
• total power dissipated by the array is 1080 mW
• the array draws current of 240 mA from the source.

This means that the whole array puts out 1.08W of power and draws .24A. This is significantly less than any of the bulbs.

---

As for brightness, see for yourself. These LEDs do not require a reflector because they have a parabolic cross-section which internally focuses each LED individually.

The LEDs I used claim a 12 degree focusing pattern. Not all LEDs do. Many have a 25 degree, while others are as much as 55 degrees. I did a subjective test, and I believe that the pattern is somewhere around 12 degrees. This does mean that across a football field, much of the light would dissipate into a circle 64 feet wide, but at any 'reasonable' distance, the light works great.

As for intensity of light, candlepower is a hard thing to objectively compare, since the overall spread is figured into the equation. First of all, candlepower is no longer a recognized standard unit. It is approximately equal to 1cd or 1 candela. These LEDs advertise a brightness of 130cd apiece x 12 = 1560cd. I'm not sure where the discrepancy is, but after being blinded many times, I know that my roommates can attest to the intensity of this light!

---

Also, when constructed correctly, the brightness of this light should be the same whether you are using a 3, 4, or 6-cell Mag-lite. The only difference will be overall battery life, which I project to be:

Flashlight----Capacity----Current Draw----Projected Life---Life Compared to Standard 12 Hours
--3-cell-------36000mAh------240mA----------150 hours----------------12.5 times greater
--4-cell-------48000mAh------240mA----------200 hours----------------16.7 times greater
--6-cell-------72000mAh------240mA----------300 hours-----------------25 times greater

Comments

[ac-dc]

Batteries in series don't have additive mAH capacity. In any combination there is still only 12K mAH.

Unfortunately from the mag LED spec you list, your LED conversion does not put out as much light as the original bulb did. The reason it looks brighter is because it is blue light in a room with more yellowish ambient light, plus the original maglight was more focused. Regardless, increasing the efficiency, I mean runtime, is still reason enough to do the mod.

Again, the light output is not really 12'. Most of the usable light will be within about 36'.

I roughly estimate your light puts out about 100 lumens, so I was wrong in a previous comment that it's not as bright, it is as bright as some Mag xenon bulbs, but this is only if my estimate does not count light-reducing heat, battery drain which redues current to the LEDs, or LED aging. We can probably ignore the LED aging unless you had a very regular use of the light to put many hours on it, and even then the savings in batteries more than makes up for cost to replace the LEDs at some point.

Check out the http://www.dealextreme.com website, they have many nice high powered 1-3W LEDs separate or including a reflector set that would drop in to a flashlight and produce more light, more efficiently.

[MustangChris429]

Whoa...lots of comments. I realize now the mAh capacity doesn't add. Thanks. Also, the light fits within 12 degrees (mostly), but the brightest part is really around 5-10 degrees. I didn't use PCB because this is supposed to be a cheap, easy thing that anyone can do. By my own experimentation, the LEDs throw off WAY less heat than the original bulb did, so I see the heatsinking as a little excessive. Again, this is not supposed to be the ultimate flashlight, but a cheap, easy way to improve the regular old Mag-lite that many people have lying around. Thanks for the tips on where to find more powerful LEDs--if this instructable proves to be as popular as it has been, I might have to check out your lead! Also, the assembly is as waterproof as it originally was, so there is no need to epoxy anything, and the array is held in place simply by pressure...it doesn't shake around at all.

[ac-dc]

THe LEDS really do not shine mostly within 12 degrees, see the Besthongkong chart, it is at least +-12' which is 24' minimum but you see on the chart that there is a lot of light left out of that 24' pattern. Yes the LEDs produce less heat but the important thing is they are susceptible to heat when the bulb is not (Much at all). They will get hot and that reduces light. Don't get me wrong that light is nice, and I'm sure if you're like the rest of us you will be building many more lights in the future so I'm just throwing out some ideas.

[MustangChris429]

Thanks, it seems that you really do know your stuff. What I said comes from my personal [very un-scientific] testing. I just took an estimate of the spread size at a known distance and noticed that it seemed the majority of the light seemed to be at about a 7 or 8 degree spread. Also, it stands to reason that if the 12 degree is the 3dB threshold, there would be significantly more light at 7 degrees than at 12 degrees. Just a thought.

[spannercrab]

A D-Cell battery does indeed have around 12Ah of charge - depending on type - i.e. alkaline, NiMh etc ... however battery capacity does not increase when cells are strung together in series. 6 x 12,000mAh cells in series will give you 9V @ 12,000mAh - if you strung the same cells together in parallel, this will give you 12,000mAh @ 1.5v (or whatever the single cell voltage is). Also, if you increase the pack voltage, your current drawn through the LED's will also increase and likely destroy your LED's - the only way you can get around this is to either modify the resistors based on your torch voltage or to use a constant current power supply to ensure an even current supply to the LED's.