Is it possible to run a MC-E off two AA?

Badbeams3

Flashlight Enthusiast
Joined
Sep 28, 2000
Messages
4,389
I don`t see any manufacture doing this. Is it just not practical?
 

Marduke

Flashaholic
Joined
Jun 19, 2007
Messages
10,110
Location
Huntsville, AL
Sure, but nowhere near full power. You would only get a VERY small boost in brightness and/or efficiency vs a regular XR-E since you would be running each die at a lower current.
 

Gunner12

Flashaholic
Joined
Dec 18, 2006
Messages
10,063
Location
Bay Area, CA
At max power or in general?

You can direct drive a MC-E wired in parallel off of 2 AA batteries, it won't be bright but it will run for a good amount of time.

At max output, you'll need a big light for the heatsinking, a driver able to get that much current out, and 2 very capable AA batteries (not practical).
 

roadie

Enlightened
Joined
Sep 17, 2006
Messages
825
Location
Singapore
if using 3 AA , might be possible as the nimh voltage is 1.2v each, therefore 3 will be 3.6 V, which is the ideal voltage to drive the LED.

:)
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
Wired in 4P, for example, an MC-E would get 250mA per core if you drove it with a boost driver rated at 1000mA output. The trick is if the driver can actually reach regulation when powered by two AA cells.

If the cells are AA Eneloops, then I believe a good quality boost driver can run in regulation. If the cells are AA alkalines, then the driver might run in regulation for the initial turn-on, depending on the boost driver. But subsequent turn-ons may very well not make it to regulation, once the cells are no longer fresh. Thus, instead of using a boost driver with 1000mA output and having problems reaching regulation, you might go with a driver with 750mA output instead.

It all depends on the boost regulator IC. Often these ICs have a switch current limit of around 1.5A or 2A. If the limit is 1.5A, the driver might reach regulation with Eneloops because #1, the Eneloops can hold a higher voltage under load, resulting in #2 a lower current demand from the batteries, and #3 barely making it under the 1.5A switch current limit. In contrast, AA alkalines will probably suffer greater voltage sag (especially when the cells aren't fresh anymore), requiring a higher current from the batteries to deliver the power necessary for a 1000mA driver output, and potentially exceed the 1.5A switch current limit.

If you can reach regulation and drive the 4P MC-E at 250mA per core, an M-bin MC-E could deliver over 300 emitter lumens, with a very low forward voltage, and thus a very low power consumption. For the comparable Seoul P7, I measured Vf of 3.07V at 250mA per core, or just about 3W power delivered to the LED. If your have a 90% efficient boost driver and are using NiMH cells that can hold 1.3V per cell for a few minutes under say a 1A-2A load, then

0.9*2.6V*Ibatt = 3.07W (using the Vf data from the P7 I measured)

and

Ibatt = 1.3A.

Here is the difference when underdriving a quad-core LED like the MC-E. For a typical single die emitter like a Cree XR-E or Seoul P4, you might get a forward voltage of about 3.5V at 1000mA forward current (which might give about 240 emitter lumens). With this higher forward voltage, the same calculation as above gives Ibatt = 1.5A. Thus, the lower Vf for the underdriven MC-E could make the difference between running in regulation or not.

Now if you change the driver output current to 750mA (188mA per core, or about 240 emitter lumens), then

0.9*2.6V*Ibatt = 3.0V*0.75A

and

Ibatt = 0.96A (this assumes that Vf=3.0V at 188mA per core)

Even if you use AA alkalines that might hold 2.2V, Ibatt~1.1A. So, most likely a boost driver rated to deliver 750mA output current can reach regulation with just about any type of AA cell chemistry.

My guess as to why you don't see lights like this is because buyers want and expect huge lumens from quad-core LEDs. They are not interested in underdriven MC-Es and P7s, despite IMO the clear advantages vs single-die emitters. Also, in some sense, there is probably the perception that the LED is being wasted on such a low output, even though the emitter lumens is probably 30% greater than that for a single-die LED while running probably at least 10% more efficiently. That 30%/10% gain in output/efficiency is probably just not compelling enough to a flashlight maker.
 
Last edited:

monkeyboy

Flashlight Enthusiast
Joined
Mar 7, 2006
Messages
2,327
Location
UK
This is one way to do it:

Wire the MC-E in parallel then you could use 6 x 500mA micropucks in parallel to drive at 3A. You would need high current NiMh AA capable of around 5A continuous draw. Eneloops or Elite would do.

So it is possible but not practical.
 

MWClint

Enlightened
Joined
May 27, 2008
Messages
849
Location
Albany, NY
This is one way to do it:

Wire the MC-E in parallel then you could use 6 x 500mA micropucks in parallel to drive at 3A. You would need high current NiMh AA capable of around 5A continuous draw. Eneloops or Elite would do.

So it is possible but not practical.

:laughing: i'd love to see a mad boost setup like that.

2 gd1000's would take a nice bite out of the mc-e.
(i think u can get them up to 1.2amps each with a r1 swap to .05.)
 
Last edited:

Badbeams3

Flashlight Enthusiast
Joined
Sep 28, 2000
Messages
4,389
Thank you all for answering. Maybe I`m wierd, but I would be happy with a little brighter light...and a little more run time for a bit more money. But yea, I understand that most folks would want more...even if it mean`t a larger light.

Again, thanks!
 
Last edited:

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
Darkzero has made 3xGD and 3xSOB sandwiches to output ~3A. As stated above, running such drivers is possible but not very practical in terms of price.

IIRC, the GD uses the same sense resistor formula as the SOB. Thus, one way (of several) to get a GD1200 is to use .05 and .22 sense resistors (you get a GD1227).
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
I also forgot that you could use 2x14500 Li-ions, which are AA sized. Then you need a straightforward buck driver. If you wired the MC-E in 2S2P, you probably can drive the LED at 500mA per core or even 600mA per core. The cells would have to be able to deliver around 1A or so to feed the driver. You'd get at least 500-600 emitter lumens (at the start anyway). The catch is that run time is shorter than the heavy underdriving case and heat generation is greater.

Of course, you could go with a 4P MC-E with a buck driver feeding the emitter at 250mA per core. There is no question about running in regulation since the input voltage is well above Vf. Current draw from the batteries here should be only about 1/2 amp.

If you went with 1x14500, you could use a buck-boost driver on the 4P MC-E at 250mA per core.
 

Alan B

Flashlight Enthusiast
Joined
Nov 19, 2007
Messages
1,963
Location
San Francisco Bay Area
2S2P MCE driven by two li-ion AA's has been done. Saw it at the San Jose CPF meeting two days ago. Measured about 1000 lumens in the integrating sphere. Heat and runtime are an issue. Hard on switches, too. There's a thread around here on it somewhere.
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
Can you provide more details? Presumably, this is a homebuilt/custom job?

Not sure how you get 1000 out the front lumens (I assume that's what you mean by 1000 lumens measured in an IS) from an MC-E unless you are running it way beyond 700mA per core. At the usual 20% reflector and window losses, the emitter lumens must be around 1200 lumens. Even an M-bin MC-E is rated only at 430-490 lumens at 350mA per core. At the max forward current of 700mA per core, the MC-E might deliver about 750-850 emitter lumens based on an estimated relative luminous flux factor of 1.75.

Also, even if we assume 2S2P at 700mA per core, you are probably looking at Ibatt~1.5A, which is probably greater than the recommended 2C discharge rate for standard 14500 Li-Co chemistry Li-ions (I doubt, for example, that the AW14500 cells really have 750mAh capacity at 1A-2A load -- the AW16340 cells certainly don't). In your presumably overdriven case, you are probably asking the 14500 Li-ions to deliver closer to 2A. The cells won't be happy with that kind of treatment. And in a small AA-sized light, the inevitable junction temperature heatup is going to drop that initially high output anyway.
 
Last edited:

Alan B

Flashlight Enthusiast
Joined
Nov 19, 2007
Messages
1,963
Location
San Francisco Bay Area
The 1000 lumen mini-mag is not a 'practical' light in the sense that it cannot be run very long, both from the standpoint of heat dissipation and the standpoint of battery and LED overcurrent. Rather like the 300 plus watt hotwires. Though with a regulator we can run them down to lower levels and get long runtimes.

In the minimag AA there is not much room for controllers, but if one was added to the MCE setup it might be more practical and still retain the amazing output, albeit for short TURBO bursts. It would be a tight squeeze.
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
Wayne's 14mm drivers such as the SOB are the same diameter as a AA cell. You could source an SOB1500 and drive the 2S2P MC-E at 750mA per core if you wanted big lumens. Better apply good thermal management to the driver or it will cook itself. Unfortunately, sinking both the driver and the LED to the same heat sink is a recipe for failure in a AA Mini Mag if you are going for high power. I'd probably go with an SOB750 at most.
 
Last edited:

bigchelis

Flashlight Enthusiast
Joined
Oct 30, 2008
Messages
3,604
Location
Prunedale, CA
The 1000 lumen mini-mag is not a 'practical' light in the sense that it cannot be run very long, both from the standpoint of heat dissipation and the standpoint of battery and LED overcurrent. Rather like the 300 plus watt hotwires. Though with a regulator we can run them down to lower levels and get long runtimes.

In the minimag AA there is not much room for controllers, but if one was added to the MCE setup it might be more practical and still retain the amazing output, albeit for short TURBO bursts. It would be a tight squeeze.

The mini Mag sure did close to 1000 out the front lumens as tested by MrGman, but it was with no bezel on. Additionally; it had the biggest dip in lumens after turn-on:laughing:

Still; the member who made it desserves :twothumbs. This proves the MC-E AA option is possible and if want to wow the crowds, this light certainly does it. Now how could go for more than 5 minutes:candle::candle::candle:

I fully agree this is not a practical light, but I want it...............Of course it would be carried in addition to my EDC.
bigchelis
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
Let's call a spade a spade. This is a direct quote from MrGman on this MC-E direct drive AA Mini Mag:

'Please note that this light put out those very high lumens numbers for less than 1 second!!!!!. I used the "peak" capture feature of the meter to get these. Just using the normal reading mode which takes 1 second to update it was down in the 400 lumen range and give it 3 seconds to warm up and it was down into the 300 lumen range. neither the batteries nor the LED and its Phosphors can sustain this output for anything other than a turn on pulse.'

Impractical IMO doesn't come close to describing this mod. And if there was no bezel used for the test, the actual lumens are probably even less (say by 20%) and heat buildup probably even worse.
 

jcvjcvjcvjcv

Enlightened
Joined
May 1, 2008
Messages
273
Location
The Netherlands
What about running the MC-E in 4P off a AW 14670? The 14670 is sold as being able to supply 4A if I remember right. Also; they are 1100 mAh

3.6 Volts x 1,1 Ah = 3,96 Whr
3,96 Whr / (3.6 Volts x 2.8A) = 24 minutes runtime

in theory..
 
Top