SuperCap 9V battery

posted Dec 6th 2005 11:00am by Eliot Phillips
filed under: misc hacks

supercap battery

9V batteries are often found in devices that aren’t used very often. If you use a NiCd rechargeable battery you may find it completely discharged by the time you need it. Capacitors on the other hand can maintain their charge for years. This circuit uses a 10F cap with a switching voltage regulator to increase the voltage from 2.3V to 9V. With a light load the cap will last up to 3 hours and once discharged it can be recharged in less than 20 seconds. Warning: PDF link.

[thanks nullset]

Read

Comments [80]

Reader Comments

interesting… i might try this!

Posted at 11:50 am on Dec 6th, 2005 by Joel

10F! Whoa! I’m impressed with that alone… this is pretty sweet… now I’ve just gotta find something of mine that uses 9V batteries to try this out on!

Posted at 11:56 am on Dec 6th, 2005 by Archon

I wonder fi the gauss pistol could benefit from these gold-cap capacitators….

Posted at 12:09 pm on Dec 6th, 2005 by John

What’s depicted is a 10F cap, not a 1F one. 10 times the zap!

Posted at 12:12 pm on Dec 6th, 2005 by Wayne

does anybody knows any way to turn this into 4.5V and if it is a good powersource for a cheap digital camera?

Posted at 12:14 pm on Dec 6th, 2005 by rui maldonado

“The SuperCap can deliver a peak current of around 6 A, this equates to a lot of power (heat) dissipated in low resistance loads.”
I guess you wouldn’t want to use your toungue to test one of these, like you would a normal 9V…

Posted at 12:14 pm on Dec 6th, 2005 by Bob

I wonder why nothing like this is on the market? That seems ultra useful, especially for cameras, etc, where you may have access to electricity a lot but don’t want to wait hours for a charge.

Posted at 12:21 pm on Dec 6th, 2005 by shbazjinkens

I so want to stick my tongue to the end of that!

Posted at 12:24 pm on Dec 6th, 2005 by The Tech Department

Cor, Elektor Electronics to the rescue! I remember reading them when I was a little nipper and they used to have lots of cool stuff like this in their magazines.
Hmm, the article explains that one of the transisters is used as a zener to create an 8v value (the circuit produces around 8.5v).
I’m tempted to adjust the circuit and create a dummy battery to make a super-cap based 3v cell to stick into my wireless optical mouse. That way, I don’t even have to worry about memory effects (quirk of NiCads) and even if it gets flat, it’ll charge in an instant!

Posted at 12:43 pm on Dec 6th, 2005 by Sheldon

so, er , how about…

take that basic idea of using it in a wireless mouse at 3v, and adding a kinetic charger, that that moving your mouse charges your mouse.

might turn out to be a massive sized mouse…

or convert the wheel , or better i guess, on a two wheeled mouse, make one wheel a charger, …

Posted at 12:53 pm on Dec 6th, 2005 by altpersona

Great Article – more like this project please. I miss the days of Popular Electronics Magazine.

Posted at 12:55 pm on Dec 6th, 2005 by Vonskippy

This isn’t sold on the market because a AAA battery has 1000x the amp hours of this 10F supercapacitor. While a $5 10F supercapacitor is great for a capacitor, it is orders of magnitude too small for something like a camera. These caps are useful for a high peak power / low average power solar charged item as they have longer lifetime with less charge loss. See: http://electronics.howstuffworks.com/capacitor2.htm

Posted at 1:00 pm on Dec 6th, 2005 by Mike

The article doesn’t talk about charging the cap. From the circuit it looks like you have to charge the cap leads. The circuit doesn’t support charging from the 9v terminals.

bob: The voltage regulator limits the current available. The cap may be able to generate 6A, but the regulator will only do 1.5A (google the datasheet)

Posted at 1:12 pm on Dec 6th, 2005 by tech guy

If you put one of those tiny solar cells on the outside of a device with one of these inside, it could always be ready to go – as long as you didn’t keep it in a drawer.

This is ultra-neat.

Posted at 1:24 pm on Dec 6th, 2005 by Crosius

i’m pretty sure you can’t trickle charge a cap from a solar cell…

Posted at 1:50 pm on Dec 6th, 2005 by alex

THIS KICKS ASS!!! i am going to have to go this!

Posted at 1:56 pm on Dec 6th, 2005 by magnavoid

How well would this work in a Cmoy amplifier?

Posted at 1:57 pm on Dec 6th, 2005 by Theblunderbuss

Nice application of a supercap in such a way. But i dont like how it is done. A inductive switcher transistor pair with a zener clap is not energy efficient. (not to mention the quicent current usage).

I would like something DC-DC up converter to upconvert the low voltage. It would increase the efficiency a lot, i have a setup here with a solar cell and a MAX756 to power a small microcontroller on 3.3Volts. The setup can power a white led for half a hour on a 2.5V 3.3Farad goldcap.

Posted at 2:11 pm on Dec 6th, 2005 by BoomBox

why alex (#13), of COURSE you can trickle charge a capacitor. oh the joy of BEAM robotics:

http://www.beam-online.com/Robots/Circuits/1381.html

http://www.solarbotics.net/library/circuits/se.html

makes me miss building those little buggers. tildens spyder walker anyone?

Posted at 2:27 pm on Dec 6th, 2005 by steve

rui maldonado: Yep, replace T3 with a 4-V zener and get a 4.5V supercap battery. I guess how appropriate it would be for your cheap digital camera would depend upon its battery compartment (determines the size of the caps you could use) and how much current it will draw.
Crosius, alex, you could indeed charge it from a solar cell, but keep in mind that the total cell voltage static would have to be less than 2.3V at the cap terminals, or would have to at least 4V at the input of the charge regulator.

Posted at 2:30 pm on Dec 6th, 2005 by Pocketbrain

#13 alex – google for “beam solarengine” for different ways of charging caps with solar cells … my understanding of supercaps had been that they weren’t good at discharging quickly, has this changed recently?

Posted at 2:32 pm on Dec 6th, 2005 by brian

alex> couldn’t you use an accumulator like a BEAM robot does to charge the cap in short spurts?

Posted at 2:32 pm on Dec 6th, 2005 by Crosius

Posted at 2:33 pm on Dec 6th, 2005 by Pocketbrain

This could be made even better with one of the 10farad capacitros of the same size :D

Posted at 2:55 pm on Dec 6th, 2005 by martin

hey steve! nice to see another beam/tilden fan around! I’m only just getting the time to really play with beam circuits, but I really like the mostly-analog nature of tilden’s designs.

Posted at 2:56 pm on Dec 6th, 2005 by brian

This would be great for firealarms :)

Posted at 3:43 pm on Dec 6th, 2005 by t0ny

(#21)
Ir already uses a 10farad cap.

Posted at 4:03 pm on Dec 6th, 2005 by Alex

Fixed

Posted at 4:54 pm on Dec 6th, 2005 by Eliot Phillips

Ummmm… are you sure THAT cap is rated at 10 farad? Looks more like a 10 microfarad to me. If I am remembering right a 10 farad capacitor would be huge.

Posted at 5:18 pm on Dec 6th, 2005 by Mattrix

Just for the record i beleive they are micro-farad capasitors. a 10 farad capasitor would be the size of a waste basket.

Posted at 5:19 pm on Dec 6th, 2005 by jesse

jesse: Normally I would agree with you, but if you look at the cap in the picture, it clearly says 10F right on it. Then, if you had actually rtfa you would have noticed this part:

“The circuit shown here uses a small outline supercap with a capacitance of ‘just’ 10F (yes, that’s 10,000,000uF!)”

Posted at 6:02 pm on Dec 6th, 2005 by TNS

You can see the 10F rating in the picture – sure, a 10 farad capacitor would be huge, but these are 10 farad super-capacitors. I don’t know exactly how they are composed differently from normal capacitors but yes, this is a 10 Farad package.

Posted at 6:05 pm on Dec 6th, 2005 by smilr

Maxwell offers some ultracaps in a “D” cell format:
http://www.maxwell.com/news/release.asp?PRID=183

They have a lot of good information on their site.

Posted at 6:35 pm on Dec 6th, 2005 by Gray cube

this would be good for smokedetectors. cuz they are run on house circuts but the 9v is there for when power lines fail

Posted at 7:02 pm on Dec 6th, 2005 by phycon

Yay for BEAM. It’s been a while since I’ve followed anything in the BEAM area, anything interesting happen since, oh say… ‘99?

Just to re-inforce others, that is definitely 10 farads, as stated in the article, I don’t think a 10 microfarad capacitor would even drive a single cycle of the oscillator (seeing as the other two caps are 100 uF each). I’m more excited about puting this idea into devices, than using it to replace batteries, personally. If you need to keep an MP3 player going for the 30 minute commute to school/work, and you forgot to charge it the night before, it’s much easier to let it charge for under a minute when you’re rushing to get out the door. Oh the possibilities.

Posted at 7:20 pm on Dec 6th, 2005 by -Pat

http://www.mutr.co.uk/prodDetail.aspx?prodID=337

i guess it could be 10f then

Posted at 7:28 pm on Dec 6th, 2005 by captsnuffy

Guys, this 10F supercapacitor would run a mp3 player like my little muvo tx for less then 2 minutes on a single charge.

AAA battery of 1.5V and 2800mAh = 4.2 Wh. A 10F supercap at 2.5V would be 0.00694 Wh. My Muvo runs for 14 hours on the AAA battery. 14 hours * 0.00694 Wh /4.2 Wh = 0.02315 hours or 1.4 minutes.

Even a $25 50F supercap which is one of the most powerful ones would only give you 7 minutes of listening time. A camera would eat one of these up just as quick. Supercaps are great for beam, solar, leds etc but are a factors off to replace batteries in most applications. Supercaps are great for instant power (so fast charging/discharging) but not great for energy storage.

The reason these supercaps are so small and light is that they use aerogel for the large amount of surface area.

http://www.eepn.com/Locator/Products/ArticleID/28446/28446.html

Posted at 8:20 pm on Dec 6th, 2005 by Mike

Yay, my second mainpage post! (The other was the georgia tech student’s robotic guitar)

Posted at 8:47 pm on Dec 6th, 2005 by nullset

Unless your waste basket is the size of an AA battery, then yes… I’m staring at a 1 farad 5.5v super cap from allelectronics.com .78″ in diameter and .30″ thick. Not too huge at all.

Posted at 8:55 pm on Dec 6th, 2005 by Pedro

I Think that this is a wonderful thing, I mean you could have a flashlight that needed to charge for like no time, and could run for a couple hours

Mike, I think your numbers for the amount of power from a AAA are off by a factor of 2. According to Energizer, the bottom level Energizer AAA is only 1250 mah. That is roughly one half, but my point being, these capacitors hold a lot more juice than youd think….

Posted at 10:51 pm on Dec 6th, 2005 by Jethro

“I wonder fi the gauss pistol could benefit from these gold-cap capacitators….”

Actually, not really because the discharge time is in the seconds according to the article. Now if they made these with a much lower ESR…

Posted at 12:00 am on Dec 7th, 2005 by bakagamer

My number for AAA could be wrong so lets use 900mA instead of 2800mA. Point is thats 4 minutes of mp3 player time. The difference between a conventional battery and this 10F supercapacitor is on the order of 1000 times the capacity. Lets say I wasn’t half off but was 10 times off, that still is 100 times less capacity. I have supercapacitors here. For microcontrollers, leds, and pager motors they are great.

According to this school physics site, a 10F supercapacitor will do over 30 seconds with a standard flashlight bulb. A couple hours is 7200 seconds by my calculation. Off by 240 times.

http://www.djb.co.uk/SidePages/PhysicsPages/electricity.html
(bottom of page)

Posted at 12:14 am on Dec 7th, 2005 by Mike

He says it will run a light load for 3 hours without a recharge… but that’s a pretty big exaggeration.
He states the quiescent draw of the converter is 2ma at 2.3V, meaning a 4.6mW draw with no load, or 0.0046 joules per second.
Energy stored in a capacitor = 0.5*C*Vsquared
So stored energy = 0.5*10Farads*2.3*2.3 = 26.45 Joules.
26.45 / 0.0046 = 5750 seconds of runtime assuming that the 100% of the capacitor’s energy is used (not likely).
Meaning that after about 1.6 hours of this circuit powering absolutely NOTHING, the cap will have fully discharged. It’s a cute hack, but he should have used a proper low power dc-dc converter circuit.

Posted at 1:32 am on Dec 7th, 2005 by Afroman

“I Think that this is a wonderful thing, I mean you could have a flashlight that needed to charge for like no time, and could run for a couple hours”

You guys are greatly overestimating the power of these caps. A 10f 2.3v super cap stores about 25 joules; it could raise 1g of water by ~50F. That is also about the amount of energy used in the flash of a disposable camera.

That 10f cap would power only a single LED for a few minutes, almost useless for a flashlight.

Posted at 1:41 am on Dec 7th, 2005 by ...

Pocketbrain: Thanks! the compartiment is 3AAA cels sized. i wonder if 2 capacitors of lower size in series would have the same efect.
I also wonder if it could take, let’s say, 4 ours of photos. normal bateries drain in 7days if not used to feed the internal memory, wich i dont use.

Posted at 6:20 am on Dec 7th, 2005 by rui maldonado

oh gosh… i need autoupdate for firefox… again…
well, as far as i have seen, it is almost useless unless used for emergency power for a mobile phone, for example.
but really, has anyone experienced yet the autonomy of these cells?
I would love to experiment, even in a flash light.

Posted at 7:05 am on Dec 7th, 2005 by rui maldonado

Ugh, I just spent 15 minutes trying over an over to post a comment, is there a size limit that it is not telling me? It kept telling me rejecting my comment field.

Posted at 11:42 am on Dec 7th, 2005 by Hans

#26 and #27, I thought the same thing. 1 farad car audio caps are the size of soda cans or spraypaint cans… I did some googling and these supercapacitors are apparently different kind of thing, with different pluses/minuses.
wiki:
http://en.wikipedia.org/wiki/Supercapacitor
http://en.wikipedia.org/wiki/Capacitor

Also here’s some company page that mentions both:
http://www.cap-xx.com/resources/reviews/strge_cmprsn.htm

Unfortunately I couldn’t glean whether these supercapacitors make regular caps obsolete, or if they are just used for different things. I’m sure car audio buffs would prefer to use a tiny $5 unit that has ten times the storage as their $100 1farad caps if it were directly applicable… can anyone explain the difference?

Posted at 2:54 pm on Dec 7th, 2005 by M56

(not the same hans as in #44)

afroman post #40: “he states the quiescent draw of the converter is 2ma at 2.3v, meaning a 4.6mw draw with no load, or 0.0046 joules per second.”

Your mistake is that you take this number as a constant in the next calculation. As the voltage of the supercap decreases, so will the current (most probably), so the power (=U*I) will decrease even faster.

With this underestimation you get 1.6 hours, so I think that if you take the diminishing power draw into account, 3h seems very plausible.

Posted at 3:09 pm on Dec 7th, 2005 by Hans

Correct me if i’m wrong, but in beam, it merely uses the circuit for voltage detection and amplification to send the power to the motor. Basically, the solarcell charges the cap directly, and the rest of the circuitry is the trigger.

Posted at 6:30 pm on Dec 7th, 2005 by kmangwing

to m56:

Super caps are not much use (yet) for audiophiles (except for backing up your presets) as they have a very large Equivalent Series Resistance (ESR).

The aluminum-electrolytic caps (those monsters used for car audio) have an ESR of ~.01 to .0001 ohms which allows them to dump their energy quickly to boost the current available form the battery to make those huge bass notes possible.

a super cap on the other hand has an esr many times higher than that; this cap can only provide 9a, almost nothing compared to the thousands possible with electrolytic caps.

btw, elliot can we got rid of the Weblogs, Inc. Customer Service from the post activation e-mails?

Posted at 8:01 pm on Dec 7th, 2005 by ...

Where do I find the ESR on the spec sheets – is that “Internal Resistance”?

I was reading Maxwell’s UltraCap FAQ and they claim to have a much lower ESR than other super or ultra caps, but I’m not sure were that is specified on their spec sheets…

Here is their FAQ:
http://www.maxwell.com/support/faqs/type.asp?iType=15

Here is their 10f small ultracap:
http://www.maxwell.com/ultracapacitors/products/PC10.html

Posted at 6:58 am on Dec 9th, 2005 by Gray cube

Well at least projects like this can serve as learning project for those who are just learning electronics. Lets see more of the same in the future to make learning exiciting.

Posted at 1:24 am on Dec 16th, 2005 by LUKE

yes, more of these kind of projects, and of course great article

Posted at 6:32 am on Dec 17th, 2005 by livejasmin

I like the wireless mouse comments above. My thought would be to add the old school ball& rollers to an optical mouse and use the rollers as a charging circuit then even light movement would provide some kind of charge.

Posted at 5:05 pm on Dec 27th, 2005 by bjheu

Good hack idea for powering the suitable devices.

It stores little energy compared to batteries, it won’t power an LED very long, and isn’t practical for a digital camera. E = 0.5 * C * (V ** 2) = 0.5 * 10 F * (2.5V ** 2) = 31.25 Joules Compare this energy to a typical 9V battery: 9V * 0.1 AH = 9V * 0.1 AH * 3600 Seconds/Hour = 3240 Joules, greater than 100 times the energy!

Posted at 9:50 pm on Jan 1st, 2006 by Lawrence miller

a while back i came apon another hackaday episode, a usb charger. could you possibiy solarly charge the capacitor, to run the usb charger? http://ipod.hackaday.com/entry/1234000270029372/

Posted at 2:47 am on Jan 7th, 2006 by the0ne

Now that’s what I call a capacitor!

Posted at 12:43 pm on Jan 25th, 2006 by Unchain

If only there was one for a PSP… time to get the soldering iron heated.

Posted at 6:22 pm on Jan 28th, 2006 by jake Biggs

I wonder how well this will work with that diy usb charger at http://ipod.hackaday.com/entry/1234000270029372/

if it would work than for fun put on one of those tiny calculator solar panels!

Posted at 7:08 pm on Feb 2nd, 2006 by dellpickel

I know this is a really old post, but I’m a n00b to electronics and have a project that I have been working on for a few weeks that involves this circuit. What I’m attempting to do is build this little sucker and use it to power a single apple pro speaker, the 10W harmon kardon globes. Basically, I’m trying to run a contact mic directly to the speaker and use this circuit, attached to some solar cells, to power it. I’m totally unfamiliar with all of the equations that are posted above, however it seems like this may not be the option for me. I was wondering if someone would be gracious enough to help me out with this. I can provide more details if needs be.

Thanks!

Posted at 3:44 pm on Mar 15th, 2006 by Chris

#52: ESR = effective series resistance

This is the same as “internal resistance”, because a supercapacitor is modeled electrically as a capacitor in series with a resistor. Keep in mind that the ESR does vary with temperature, and directly affects charging voltage.

For instance: the Maxwell PC-10 is rated at 2.4V, but with the ESR figured in, you would have to charge it (measured across the terminals) to 2.55V.

Posted at 1:14 pm on Jun 7th, 2006 by direavenger

Does anyone have this article? The link is no longer valid.

Posted at 9:10 pm on Jun 9th, 2006 by drwho

You’d need about 10,000f to hold the juice of a single AA battery. There are cool applications for this 9V “milli-battery”, but battery replacement in general isn’t it. Now if those brainiacs at MIT are able to produce their carbon-filiment capacitor as hoped, we’ll have 10,000f caps in a AA form factor. THAT will be the cat’s pajamas!

Posted at 11:01 am on Jun 21st, 2006 by Chickenman

tui muon hack per fect

Posted at 1:37 am on Sep 18th, 2007 by duongga

Hey! where the heck did the .pdf go?!

Posted at 10:18 pm on Oct 21st, 2007 by john

Just in case someone’s looking for the PDF – I’ve found it here: http://www.digitalequilibrium.com/docs/9v_capacitor_battery.pdf

Posted at 6:36 pm on Dec 13th, 2007 by yaleman

Great circuit. Remember if you use typical 2N2222’s for the transistors, put a 200 volt or greater rated diode across the collector-emitter of the transistors connected to the inductor or you will blow the transistors. Also, you can just use a 9 volt zener in place of the one transistor with no connection to it’s collector.

Posted at 11:24 am on Aug 4th, 2008 by john mcc

The link to the PDF article is a broken link…

Posted at 4:22 am on Sep 22nd, 2008 by Mark

i want to try this but i don’t know where to buy

Posted at 8:38 pm on Nov 20th, 2008 by emmanuel

maganda

Posted at 6:49 pm on Nov 21st, 2008 by james

I can’t find the schematic for the “Supercap 9v Battery” by Eliot Phillips anywhere. Has it been removed. Thank you. Ed

Posted at 7:23 am on Mar 10th, 2009 by Ed Ianni

As mike, “…” and others have already pointed out: the energy a supercap holds is nowhere near that of a battery. You can use it however as backup power for things that use almost no energy (memory, some microchips on standby) or for something that runs intermittent. If parts are matched properly, a cap needs very little (read none) in charging circuits:
A small solar cell (6V for a 5V cap) can charge a cap directly. The cap will pull down the voltage and the voltage rises as the cap charges. Once it reaches 5V, you have to start using the cap, so the voltage doesn’t rise higher, or have a bypass zener.

Posted at 8:54 am on May 16th, 2009 by tantris

The PDF file link is giving an error.

Posted at 4:49 pm on Aug 24th, 2009 by spuffler

BTW, with SMT PWM and some finagling, you should be able to get 2 of those caps in there, for 2x the run time, which means very close to the same as the 9V battery it is replacing.

Posted at 4:50 pm on Aug 24th, 2009 by spuffler

Ah I see yaleman posted the link on Decc 13 2007, use your browser find function.

Posted at 4:53 pm on Aug 24th, 2009 by spuffler

Maxwell’s stuff is quite pricey. The smallest BC series at 350F is $72 today at a popular online distributor….. I can go through an awful lot of alkaline 9V batteries for that.

Posted at 5:03 pm on Aug 24th, 2009 by spuffler

The link to the PDF article is a broken link

Posted at 5:37 am on Oct 15th, 2009 by Anton

Now that’s what I call a capacitor!

Posted at 2:06 am on Oct 24th, 2009 by Anton

IS THIS YOUR PDF?
http://www.scribd.com/doc/17684600/SuperCap-Battery

Meh, probably not… Though, the circuit diagram is there…

Posted at 10:31 am on Nov 4th, 2009 by Robert