0

Flashligh Battery Selection
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

Re: Flashligh Battery Selection

by Lord Cupcake on Thu Apr 15, 2010 11:19 pm

They may be expensive, but they sure can get the job done. Note that it can last ~8 hours if the flashlight is at the highest setting possible. And that is if it is on constantly. I'm sure that I could get way more, as I usually don’t use a flashlight for more than 10 minutes straight at a time, and I usually don't use the brightest setting possible.

I did consider using NiMH and Li-ion, but NiMH batteries don't like being used in a high drain application, and don't have as high a charge capacity (up to 2500mAh). I wouldn't call my project high drain, but I wouldn't call it low drain either. As for Li-ion, they are very volatile if used incorrectly, and require more components for charging. I'm on a tight component budget, so that was thrown out to. Also, you can't really bring a charging cable to use in the wilderness, where I would most likely be using the flashlight, so you can't recharge the battery when you need it the most. Since AA is a very common size, and since I have a lot of Duracell Ultra AA's lying around, I figured that that would be the best way to go. You can also bring lots of spares into the wilderness!

Thanks for the information about battery configurations. That is defiantly something that I would need to know in the future.
The power of tyrants is limited by the endurance of those whom they oppress.
User avatar
Lord Cupcake
 
Posts: 16
Joined: Mon Apr 12, 2010 6:21 pm

Re: Flashligh Battery Selection

by Entropy on Fri Apr 16, 2010 10:33 am

Lord Cupcake wrote:I did consider using NiMH and Li-ion, but NiMH batteries don't like being used in a high drain application, and don't have as high a charge capacity (up to 2500mAh).

Um, NO. NiMHs are MUCH better suited to high-drain situations than alkalines due to significantly lower internal resistance. This is why running photographic flashes off of alkalines is considered a "last resort" in emergencies - recycle times and capacities are abysmal due to the high internal resistance of alkalines.

NiCDs are even better than NiMHs in high-discharge situations but their total capacity is very low, as a result they're only used in extreme high current situations (such as R/C cars), and in most of those situations are being displaced rapidly by Li-Ion.

I think <2 hours for 15 5mm LEDs seems a bit overconservative from 2AAs... Although pretty much the way you're doing it (massively parallel with individual dropping resistors) is highly inefficient which hurts your runtime.

As an example, the Coleman MAX 2AA is driving a high power LED (one of the Cree XLamp series) at a power level of at least 1W - it lasts 4-5 hours on 2 NiMH AAs. Traditionally, 1W Luxeons were considered to put out as much light as 12-15 "traditional" 5MM LEDs.

Also, you can find high-power 5mm LEDs now. (Such as the "5-chip" units sold by eBay vendor Topbright) - these are driven at 100 mA. You could probably make a very efficient and compact light source using a constant-current step-up driver with two of these in series.

candlepowerforums.com has a LOT of info on DIY flashlights. A warning though: The technical expertise on those forums is VERY high so I would strongly advise doing a LOT of reading and research prior to posting there.
Entropy
 
Posts: 472
Joined: Tue Jan 08, 2008 12:43 am
Location: Owego, NY USA

Re: Flashligh Battery Selection

by Lord Cupcake on Sat Apr 17, 2010 11:06 am

Entropy wrote:Um, NO. NiMHs are MUCH better suited to high-drain situations than alkalines due to significantly lower internal resistance.

Strange. It says on the Adafruit battery page that they
don't like to be used in a high-drain system.
That is where I have been getting some of my information about batteries. Did Ladyada make a typo talking about the "Con's" of Ni-MH? If there is indeed a typo, would Ni-MH batteries last longer than Alkalines, even though they have less mAh's of charge capacity?

Entropy wrote:I think <2 hours for 15 5mm LEDs seems a bit overconservative from 2AAs... Although pretty much the way you're doing it (massively parallel with individual dropping resistors) is highly inefficient which hurts your runtime.

Actually, I am using 3AA's, and I am arranging the LED's parallel in groups of 5, with each group having two .5W resistors. The user can select through which resistor current is fed with a switch, which is how I am switching between 20mA and 10mA. Is using this kind of setup still as inefficient? Also, what exactly do you mean by "I think <2 hours for 15 5mm LEDs seems a bit overconservative from 2AAs". Are you saying that less than 2 hours from 2AA's is low, or did you mean to say greater than 2hours is high?
The power of tyrants is limited by the endurance of those whom they oppress.
User avatar
Lord Cupcake
 
Posts: 16
Joined: Mon Apr 12, 2010 6:21 pm

Re: Flashligh Battery Selection

by pm6041141 on Sat Apr 17, 2010 4:51 pm

Actually, I am using 3AA's, and I am arranging the LED's parallel in groups of 5, with each group having two .5W resistors. The user can select through which resistor current is fed with a switch, which is how I am switching between 20mA and 10mA. Is using this kind of setup still as inefficient?


From what I can tell battery life is a pretty big concern in your design. Have you considered PWM-ing the led's instead of the straight resistive load and your battery life should improve.
pm6041141
 
Posts: 79
Joined: Thu May 29, 2008 5:26 pm

Re: Flashligh Battery Selection

by Lord Cupcake on Sun Apr 18, 2010 1:09 am

I'm sorry, but I have absolutely no clue what PWM-ing is. Can you explain it in layman's terms, and how much could my battery life improve?
The power of tyrants is limited by the endurance of those whom they oppress.
User avatar
Lord Cupcake
 
Posts: 16
Joined: Mon Apr 12, 2010 6:21 pm

Re: Flashligh Battery Selection

by pm6041141 on Sun Apr 18, 2010 7:50 am

Sorry, I was speaking about PWM or Pulse Width Modulation. http://en.wikipedia.org/wiki/Pulse-width_modulation
pm6041141
 
Posts: 79
Joined: Thu May 29, 2008 5:26 pm

Re: Flashligh Battery Selection

by Lord Cupcake on Sun Apr 18, 2010 9:37 am

Part of the flashlight is a 555 being used as an astable multivibrator. Using a switch, I can switch between, constant on and flashing. Would pulse width modulation affect the timing of the flashes?
The power of tyrants is limited by the endurance of those whom they oppress.
User avatar
Lord Cupcake
 
Posts: 16
Joined: Mon Apr 12, 2010 6:21 pm

Re: Flashligh Battery Selection

by pm6041141 on Sun Apr 18, 2010 10:08 am

Would pulse width modulation affect the timing of the flashes?


Possibly not. It depends on how you design the circuit. I think it may be a good idea to prototype the circuit you are making then take some measurements to see if the battery life works for your design requirements. Maybe even ignore the flashing part for now and test the LED's full on to see how long they stay on.
pm6041141
 
Posts: 79
Joined: Thu May 29, 2008 5:26 pm

Re: Flashligh Battery Selection

by Entropy on Mon Apr 19, 2010 10:14 am

Lord Cupcake wrote:
Entropy wrote:Um, NO. NiMHs are MUCH better suited to high-drain situations than alkalines due to significantly lower internal resistance.

Strange. It says on the Adafruit battery page that they
don't like to be used in a high-drain system.
That is where I have been getting some of my information about batteries. Did Ladyada make a typo talking about the "Con's" of Ni-MH? If there is indeed a typo, would Ni-MH batteries last longer than Alkalines, even though they have less mAh's of charge capacity?

Entropy wrote:I think <2 hours for 15 5mm LEDs seems a bit overconservative from 2AAs... Although pretty much the way you're doing it (massively parallel with individual dropping resistors) is highly inefficient which hurts your runtime.

Actually, I am using 3AA's, and I am arranging the LED's parallel in groups of 5, with each group having two .5W resistors. The user can select through which resistor current is fed with a switch, which is how I am switching between 20mA and 10mA. Is using this kind of setup still as inefficient? Also, what exactly do you mean by "I think <2 hours for 15 5mm LEDs seems a bit overconservative from 2AAs". Are you saying that less than 2 hours from 2AA's is low, or did you mean to say greater than 2hours is high?

That's definately a typo, unless the page is comparing it to the other battery chemistries listed (other than alkaline).

NiCds are better than NiMHs in extremely high-discharge apps, but their reduced capacity makes them worse for nearly ever app except RC cars and power tools.

Lithium nonrechargeables and Lithium-Ion rechargeables are the kings of high-discharge.

Lead-acids are good at high-discharge mainly because they tend to be large. Lead-acid is the king of the energy-to-volume ratio, but the lead kills their energy-to-weight ratio. Lead-acids can actually be optimized for high-discharge operation vs. deep-discharge operation by their physical structure. (This is why you have "starter" batteries for cars and "deep cycle" batteries for RVs)

Alkalines, however, are just plain awful at high or even moderate discharge rates. This is why it's strongly reccommended not to ever power digital cameras or photographic flashes with alkalines. I recall one person who put alkalines in a high-end flash saw 30-40 second recycle times and a battery life <40-50 pops - performance was 150-200+ pops and 5-8 second recycles with NiMHs.

As to "conservative" I mean that <2 hours seems awfully low for 15 5mm LEDs driven from 2xAAs, given that there are flashlights that push more power with longer runtimes. However, your driving method is pretty inefficient. (However, dropping resistors ARE simple, constant-current switching regulators are much tougher to work/design with.)

Also, driving multiple LEDs in parallel off of a single dropping resistor is bad.
1) LED Vfs are not all the same due to manufacturing variability. This means that some LEDs will get more current than others.
2) If an LED fails, the other LEDs will take up its current allotment, which will usually lead to a cascading failure as the remaining LEDs become overdriven, fail, and cause the remaining LEDs to become even more overdriven.

Edit: A few other things wrong with that page:
Good NiMHs are up to around 2800-2900 mAh now
There are now "ultra low self discharge" NiMHs that don't have the "high self-discharge" con, however at the cost of reduced capacity (2000-2100 mAh)
The self-discharge issue isn't listed as a con for NiCd when it should be...
Entropy
 
Posts: 472
Joined: Tue Jan 08, 2008 12:43 am
Location: Owego, NY USA

Re: Flashligh Battery Selection

by Lord Cupcake on Mon Apr 19, 2010 6:30 pm

How high is the high self-discharge of NiMH, and would getting the ultra low self-discharge be worth the reduced current capacity?
The power of tyrants is limited by the endurance of those whom they oppress.
User avatar
Lord Cupcake
 
Posts: 16
Joined: Mon Apr 12, 2010 6:21 pm

Re: Flashligh Battery Selection

by pm6041141 on Mon Apr 19, 2010 8:12 pm

Honestly, you need to prototype and test or read some datasheets and do the math.
pm6041141
 
Posts: 79
Joined: Thu May 29, 2008 5:26 pm

Re: Flashligh Battery Selection

by Entropy on Tue Apr 20, 2010 9:55 am

Lord Cupcake wrote:How high is the high self-discharge of NiMH, and would getting the ultra low self-discharge be worth the reduced current capacity?

It's a couple of percent a week. I forget exactly. It depends on how often you're going to be throwing them on a charger.

For stuff you don't use often - you want the ULSDs so you don't find the batteries are dead when you need them.
For stuff you use often and charge once or twice a week anyway - the self discharge isn't a problem.
Entropy
 
Posts: 472
Joined: Tue Jan 08, 2008 12:43 am
Location: Owego, NY USA

Re: Flashligh Battery Selection

by neutron spin on Thu May 06, 2010 4:49 pm

CR123
neutron spin
 
Posts: 163
Joined: Sat Apr 03, 2010 6:11 pm

Re: Flashligh Battery Selection

by Lord Cupcake on Mon May 10, 2010 10:48 pm

Does anyone have any suggestions as how to rapidly flash the LED's in order, giving them the appearance of all being on at the same time, while only having one LED on at a time? This would probably drastically reduce the current draw, thus increasing battery life. I would want to have three strips of five flasing separately, with a maximum of 3 LED's on at a time. This could solve my battery problem. I also want to be able to do it without having to program a microcontroller.
The power of tyrants is limited by the endurance of those whom they oppress.
User avatar
Lord Cupcake
 
Posts: 16
Joined: Mon Apr 12, 2010 6:21 pm

Re: Flashligh Battery Selection

by oPossum on Tue May 11, 2010 3:18 am

CD4017 or CD4022 + oscillator (4049UB, 74C04, etc..) + transistors to drive LEDs.

http://www.national.com/ds/CD/CD4017BC.pdf
http://www.fairchildsemi.com/an/AN/AN-118.pdf
I am the Possum, and I approve of this message. Sent from MacBook Wheel Sorry for my bad German.
oPossum
 
Posts: 636
Joined: Fri Oct 26, 2007 12:42 am
Location: Michigan, USA

Please be positive and constructive with your questions and comments.