How many AA/AAA Batteries required to power SST90 to 2200LM?

[AusKipper]

Hi,

I'm plotting something crazy, the question is easy enough though, for someone who knows how many Watts a AA and AAA can put out, and how many the SST90 requires.

So the first question is, how many AA batteries, wired in series, are required to run the SST90 LED up to 2200 lm

The second is the same but with AAA batteries.

I tried searching the internet for information on the max amount of watts a AA or AAA battery can put out but could not find it... :S

 

[VegasF6]

You are not quite on the right track searching for watts. Take a look at discharge test done, many right here on AA size batteries. Amps are what you are looking for, cross reference needed amps with voltage drop and you can calculate watts from there.

I could look at the datasheet myself, but it would be easier if you told me. Approx how much current at and what voltage do you need? I am guessing something like 9 amps 3.9V? That could be done for a short period of time with 4 quality nimh aa in series, but 4x2 series parallel would be better.

Take a look for instance at some of the discharge tests done in this thread. If the readily available eneloops won't do what you want, you can look at ultrahigh discharge cells such as elite from cheap battery packs.com

I really wouldn't even bother with AAA cells, but then, if you must, you will need quite a few.

After doing the math if you have any other questions I will try and help you.

 

[AusKipper]

OK i was being an idiot and searching for the wrong thing.

I did manage to find a website suggesting that the AA battery can supply 2.4A at 1.5V, thus meaning it outputs 3.6w?

The SST-90 I believe uses 25w to produce 2200lm, so we need 7 AA batteries.

So looking at something TK40 sized.

I found another site suggesting AAA can put out 1 amp, again at 1.5 volts, so 1.5 w.

17 AAA batteries would be required.

Neither of the above is pocket able so I guess i'm back to plotting things with/learning about capacitors.

 

[AusKipper]

You are not quite on the right track searching for watts. Take a look at discharge test done, many right here on AA size batteries. Amps are what you are looking for, cross reference needed amps with voltage drop and you can calculate watts from there.

I could look at the datasheet myself, but it would be easier if you told me. Approx how much current at and what voltage do you need? I am guessing something like 9 amps 3.9V? That could be done for a short period of time with 4 quality nimh aa in series, but 4x2 series parallel would be better.

Take a look for instance at some of the discharge tests done in this thread. If the readily available eneloops won't do what you want, you can look at ultrahigh discharge cells such as elite from cheap battery packs.com

I really wouldn't even bother with AAA cells, but then, if you must, you will need quite a few.

After doing the math if you have any other questions I will try and help you.

Thanks, i started writing my post before i saw your reply. I'll edit this post again once I check the links and stuff in your post..

Edit:

So it looks like the 'average' cell can put out around 7.5w for about 13 minutes, so if we had 4xAA we could get to the required wattage for an SST90 at 2200lm for around... 15 mins.

a 4xAA light could be made pocketable.

 

[VegasF6]

Looking at the discharge tests done here (probably should have linked that to begin with) the eneloop cell for instance at 10 amps discharge will give you right around 1V for ~9 minutes. Hard to tell since there are no times on the graph. I am sure near the end voltage drop significantly.

It appears the SST90 will need approx 3.85-3.95V to achieve 9 amps. This of course does depend on temperature. It may be fine for short bursts, certainly not for extended run times, near 36 watts A smaller flashlight of course will have less surface area to disperse heat. While maybe not the most sensible light, it would make a neat "shock and awe" design.

Are you going to attempt to build one?

 

[AusKipper]

Are you going to attempt to build one?

Me build? lol no. I couldn't build a flashlight if i was given one disassembled (slight exaggeration but still..)

A while ago i started a thread about superblast not existing in flashlights:

http://www.candlepowerforums.com/vb/showthread.php?t=272752

Basically i got a bunch of responses calling me crazy and what not, but I just cant get the idea out of my head lol.

It just seems to be so plasible to me to have a light that is pocket able run at a very high output for a few seconds. And I would find such a light quite usefull.

Hopefully I can convince Fenix or 4Sevens of its plausibility and they can build it, then i can just buy it

 

[VegasF6]

Gotcha, more of just a mental exercise. I didn't read your whole other thread, but I see that macs EDC SST-50 was mentioned. What a beautiful light, huh?

I am going to guess that the topic of possible dangers of using lithium batteries to fire a light at such a high rate of current came up. Hence, why you are wanting to do it with AA instead? Even IMR cells are a SAFE-ER chemistry, not 100% safe though. Any battery for that matter can vent, including nickel based chemistry lights.

You may be able to convince a maker like Fenix there is a market for this light, but I dunno, it is a niche market, and not cheap by any means. However I am sure one of the fine custom builders here could be convinced for a price.

I can remember very few 4xAA lights. A few Princeton Tec and some early Energizer lights. Really cool actually, they are both a rectangular body mated to a round head which conveys certain challenges in itself.

Then of course there are the batteries. While we have determined there are AA cells capable of such power, it isn't just any old of the shelf AA battery that can.

Heat we already addressed. It really isn't such a challenge to use a microprocessor controlled circuit to limit ontime to say, 10 seconds. Then there are actual thermal probes as well. A simple thermistor (not so accurate) or better still, a highly accurate sensor such as the LM35 could bias the driver.

Speaking of the driver, there are I think a handful or so created by users here, but no large scale solution. Really, it isn't difficult, simply takes a high current FET. Problem is though, much more heat generated there too. And voltage drop. Look at the heatsink on this driver! Going to take a rather large flashlight to fit that in

 

[AusKipper]

I can remember very few 4xAA lights. A few Princeton Tec and some early Energizer lights. Really cool actually, they are both a rectangular body mated to a round head which conveys certain challenges in itself.

Streamlight and Fenix both make them too now (the Fenix is kinda new)

All of the 4xAA lights have a plastic body though, which wont help dissapate the heat.

I think the current should be able to be supplied with 3xCR123 cells, and would still be pocketable, and eliminates machining problems...

 

[VegasF6]

Those are cool, haven't seen the Fenix until you pointed them out.
Alternatively you could still have a round body with 4xAA in series. Could even bore out a mag D, though its tight. I guess with the early mags it can even be done without boring. Commision one! :0

 

[Starlight]

The thing that you are missing from your mental exercise is that you will POOF an SST-90 with either 7AA in series or 3 CR123 in series. You can not put that much voltage on a LED.

 

[pepko]

4xAA eneloop is ok ...

http://www.candlepowerforums.com/vb/showthread.php?t=280018

 

[VegasF6]

4xAA eneloop is ok ...

http://www.candlepowerforums.com/vb/showthread.php?t=280018

Pepko, you beat me to it, I was looking at that host last night after this conversation. Cool. That one, the Ultrafire FC-MCU-C88 and the SUNWAYMAN M40A.

Starlight: No one said direct drive.

 

[Starlight]

VegasF6: Nowhere in this thread, does the OP use the word regulator.

 

[Chodes]

I think your idea of a superblaster is sound.
I don't understand the "crazy" responses. As long as you clearly stated the intended limited run time , what's the problem?

I have plans to make very close (or exactly) what you are talking about.

I tested an SST90 with AA Eneloops. 4 resulted in about 5.9Amps , 5 about 7.5Amps.
If you were to use for example Elite 1700mAh AAs , I'd guess you'd get 6-7Amps from 4 cells due to the lower voltage drop.
4 x Sub C NiMh should also exceed the Eneloop performance.

An IMR16340 will put out over 5 Amps into an SST50 for at least several seconds. I think that would be an easier solution. (I measured 5.6A at turn on , took around 15 seconds to drop to 4A.)

edit - in that pic I have 3 primary 123 cells - approx 280mA! LOLA option! with 2 IMR16340s and a dummy cell , about 100mA. ELOLA! Was not expecting such convenient low output options.

So you could (have someone else ) assemble a 3P clone with an SST50 and IMR16340 direct drive.


This is my triple SS50 / triple IMR16340 direct drive that I measured the 5.6A at turn on:

So far longest run has been 30 seconds at 5.2A using C cells.
Using IMR16340s the "superblaster" function is automatic. It's the cells I am pushing to the limit , not the LEDs.

A whole lot easier to make a single SST50 version (with standard surefire head))

 

[radellaf]

IMR18650 probably smallest cell that, with some boost circuit, could make a reasonable 9A pocket light for 10-15 min. Where to put the 30+W of heat would then, more than anything, be what limits your runtime to less than a minute, I'd think.

Other than that, sure, I wouldn't call it the least bit crazy to want an "upgraded" TK40 with a SST90.
To boost the Sunway M40 would... take some really special AA cells.