replace battery with super cap

[Dr_Doggy]

So i hear about these super caps and decide to get some, my first attempt of use is to replace my 4.5v AA battery pile, i realize now that i need to be careful with volt limits of 2.5v per cap,

so after putting 2 in series and charging to 4v , one is reading 2.8v and the other at 1.2v, i know that this is prolly not good.

so then i decided to put a 2.4v zener diode in parallel to each capacitor,
maybe its just me but now i m worried, originally when i would hook with pickit supply @ 4v the supply would notify me of current error, so i put a resistor between caps and supply. BUT now with the zeners, caps are connected direct to supply but not creating current error.

so now I am here wondering how to do this properly, and would diodes work for regulators? maybe i should just go back to my rechargeable AA x 3 batteries?

maybe i should use resistors? but need help determining values.

 

[cowboybob]

However you decide to charge the caps (there are 5VDC supers out there - I have some) keep in mind that the voltage level rather rapidly falls off with a load on it. The heavier the load, the more rapid the reduction (just like any cap).

Thus, nothing like a battery and, as such, they are not always the best substitute for a regular battery.

What was your 4.5 VDC battery pile powering?

And how many Farads are your caps?

 

[Dr_Doggy]

in summary a tsop IR receiver, microcontroller and some leds, <100ma furshure as that is pickit supply limit, its hard to tell accurate reading as i would need to tap in to the bus.

uC min is 3v-ish

100F each cap... so 50 with both in series.

 

[cowboybob]

so after putting 2 in series and charging to 4v , one is reading 2.8v and the other at 1.2v, i know that this is prolly not good.

Different ESRs perhaps but the combined charge is the 4VDC you provided, which is why it is generally best to charge series supers separately (or with active balancing circuits) since when one has reached its rated voltage level, it will no longer conduct.

You have to play with them to see how their combined voltage holds up under load.

 

[Dr_Doggy]

hmm. could a doubler circuit help here, where i charge caps in parallel @ 2.4v and discharge in series!?

https://www.slottechforum.com/sloti...ipler,Voltage Quadrupler circuit_files/vd.gif

i have feeling that these avalanche diodes arnt working how i assumed , suspicious cause the "top" cap never really gains charge , and not getting current overload errors that i usually would when i hook caps to pickit..

 

[Mikebits]

Without current limit on your zener's, you probably exceeded their power ratings and blew them up. You need to have a series resistor with your zener to control the amount of current.
**broken link removed**

Be careful connecting power to your PIC project so not to damage it...

 

[Nigel Goodwin]

hmm. could a doubler circuit help here, where i charge caps in parallel @ 2.4v and discharge in series!?

Why on earth to you want to try and use poor technology for a power supply?, a simple rechargeable battery will be cheaper, hold it's charge for massively longer, and most likely have a longer service life as well.

I'm totally bemused why people keep trying to use these pretty naff devices as power supplies?, presumably because the advertisers used the name 'super' for them?

 

[JimB]

I'm totally bemused why...

I agree.
A "super cap" may be ok for a low current memory hold-up supply, but the basis physics of it are that it is a capacitor, and the discharge curve of an RC circuit is exponential (OK inverse exponential), the voltage just drops off quite quickly because of the exponential thing.
A battery on the other hand has a relatively flat discharge curve, because of all the chemistry doing it chemical thing in the electrolytes, and only when the chemicals are used up does the voltage drop in a big way.

Newer is not always better. Especially when applied inappropriately.

JimB

 

[spec]

So i hear about these super caps and decide to get some, my first attempt of use is to replace my 4.5v AA battery pile, i realize now that i need to be careful with volt limits of 2.5v per cap,

so after putting 2 in series and charging to 4v , one is reading 2.8v and the other at 1.2v, i know that this is prolly not good.

so then i decided to put a 2.4v zener diode in parallel to each capacitor,
maybe its just me but now i m worried, originally when i would hook with pickit supply @ 4v the supply would notify me of current error, so i put a resistor between caps and supply. BUT now with the zeners, caps are connected direct to supply but not creating current error.

so now I am here wondering how to do this properly, and would diodes work for regulators? maybe i should just go back to my rechargeable AA x 3 batteries?

maybe i should use resistors? but need help determining values.

Hi Dr_Doggy

That's just the sort of thing I would think about. It's good to have these ideas; that way you often find something new, but as the other members say, this one really is a non-starter. But to give you an idea why here is some info:

Series Capacitors

Putting capacitors in series tends to be problematic. I see that you have found this:

Unless the capacitors have identical actual values they will have different voltages. The lower value capacitor will have the higher voltage. Supercaps have a wide open spec -20% +80% for example. That is most probably why you had such a wide difference in the voltages on your two supposedly identical 100F supercaps.

You were on the right track by by putting zeners in parallel with the supercaps. Normally two resistors are used, but that wastes power, which would not be good in your application. So the zeners seem to be a solution. One problem is the sharpness of the zener breakdown voltage. You could design a circuit to sort this out but, quite honestly, it would not be worth it. The other problem, as Mikebits said, is the the zener power dissipation on charge.

How Long will Supercaps or Batteries Power Your Circuit

You want to replace batteries with supercaps... OK, but do you know how long a 50F capacitor would power your system. If you would like to know, this is how to work it out:

The formula is:

Q=CV=IT ......... (f1)

where:

Q is the charge in the capacitor in Coulombs (1 Coulomb = 6.24x10^18 electrons)

C is the capacitance in Farads

V is the voltage across the capacitor in Volts

T is the time in Seconds

I is the current in Amps (see note below)

As we want to know T, transpose formula (f1) to get

T=CV/I ....... (f2)

NOTE: this assumes that the supply current to your circuit is constant. This will not be true in practice but it is good enough to give you a pretty good idea of how long a supericap or battery will power your circuit.

Substituting the actual values for your circuit from what you say:

C=5F

I= 0.1A

V= the voltage drop your circuit can tolerate. For example, suppose the normal supply were 4V and your circuit would stop working at 3V. You get 4V-3V=1V, so 1V is the value to go in the formula.

Thus: T=5*1/0.1 = 500 seconds or 8.33 minutes

Compare that to a typical AA 2A/Hr NiMh battery which would give you 2/0.1 Hours= 20 hours

In summary, as the other members have advised, stay with batteries. They will give a far superior performance.

About batteries

If I understand correctly, your objective is to replace non rechargeable batteries with some kind of rechargeable source. I would think that rechargeable batteries would be the best approach. If you are worried about self-discharge, go for the Panasonic (was Sanyo) AA NmH Eneloop types. They hold their charge much longer. If you want a smaller size go for AAA batteries they are about 900mA/Hr. I would advise Eneloop types for all applications anyway.

LiIon batteries are worth considering. They have a high capacity for their size, and a voltage of around 4.2V when fully charged, 3.7V nominal, compared to the 1.25V nominal for NiMh types. The CR123A would be a good choice I suggest. They typically have a capacity of 1.3 AHr and are widely used in cameras. You would need a matching LiIon charger. A typical camera battery charger could be slightly modified to do the job or if the camera charger is for CR123As no modification would be necessary.

It would probably be best to use a voltage regulator with any batteries, including non-rechargeable types.

 

[spec]

Could you let us know the voltage limits of your microprocessor or better still tell us what the device is.

I would also like to know, just out of intrest, the part number of your supercaps.

 

[Dr_Doggy]

was trying to save money , and according to some rounded online calculations i could pull it off, Going by spec's specs i would say my draw is closer to 25-50 ma since it dies after about 2 days on batteries. Hopefully with 50F on the caps they will last most of the over night.

I am running the 16f688 so it is able to go from 2-5v , the tsop receiver is 2.5-5v, and im not too worried about the leds going dim. so it can run to half charge, here is ebay item no for caps: 391146753734

I read about the esr issues which is what lead me to the zener diodes, I chose this over the resistors since idk how to calculate the proper resistance required to overcome the drift with out too much leakage. and if i need advanced balance circuit it breaks budget.

It is funny that I started this circuit with 2.5v solar feeding a 1.5v battery then used boost converter to step it up, but had problem with boost converters starting up at such low volts, so i upgraded pannel to 5v and 4.5v battery stack.

just another part in the learning curve!

Nigel Goodwin said: ↑
I'm totally bemused why...
Im happy i found out that i could buy them whilst i was reading how to make em!
I realize I do seem to run in to problem with power supplies and choosing which work best for my apps , I wish there was a chapter or 2 i could read.


So is it still consensus that I should stick with batteries?

 

[cowboybob]

From what you've described as your power needs, I would say yes.

Although I should say that I did, once, have a situation that required a large power surge of a very brief (few mS) duration, with plenty of time to recharge AND I wanted the provider source to be completely isolated from the main battery powered bus. So I used a 4F super cap that fit the bill perfectly. The exponential loss in voltage level at discharge was within the Vcc power requirements of the device being powered (an Iridium system satellite modem).

 

[Mikebits]

Add a generator to one of these babies and you all set

 

[spec]

__________________________________________________________________________________________________________________________________

Hello Dr_Doggy,

This is what I would recommend for your micro power supply:

The three batteries are AA NMH

Here are some links for them. The Eneeloop types are recommended (you would have one spare battery):

cheap AA
https://www.amazon.co.uk/BTY-Rechar...3194&sr=1-37&keywords=battery+aa+rechargeable

eneloop AA
https://www.amazon.co.uk/Sanyo-1900...d=1447623627&sr=8-33&keywords=battery+eneloop

eneloop AA
https://www.amazon.co.uk/Panasonic-...=1447623879&sr=8-118&keywords=battery+eneloop

The graph shows the discharge voltage characteristics of NMH batteries which seem to be designed for your application. They probably have a more stable voltage with discharge than the batteries you have been using.

To get the best from the batteries, and optimise their life, you need to use a smart -dv/di charger.

You say you have a 2.5V solar cell. If you had a further two maybe a light powered charger could be worked out.

I think you will be suprised how long the Eneloop batteries will last before they need recharging. They like a couple of discharge/charge cycles to reach their full capacity.

A 100nF ceramic capacitor across the battery stack would add the last bit of finess (if in doubt decouple).

You could even put your supercaps in series across the batteries- not sure if that would make much differnce but it would make a really low impedence supply. It wouldnt do any harm anyway.

 

[spec]

...
It is funny that I started this circuit with 2.5v solar feeding a 1.5v battery then used boost converter to step it up, but had problem with boost converters starting up at such low volts ...

Could you post the circuit diagram of the bost converter or a web link- just intrested. Also could you give info on solar?

 

[spec]

I realize I do seem to run in to problem with power supplies and choosing which work best for my apps , I wish there was a chapter or 2 i could read.

I know how you feel. We have all been there I'm sure. You would not be able to find an explanation in books for every problem you would like to solve in electronics or any other field. The way to solve a problem is to be very familiar with the basic rules and make sure you apply them correctly. The most basic rule is Oms law and after that Kirchhoffs first and second laws. You can go a long way with just Ohms law if you apply it correctly.

The other thing is you need is the basics of physics. The most important is that energy cannot be gained or lost. That may sound odd but it is true. It's also important to use approximations. For example with the explanation about how long the micro would run off a super cap or battery I simplified just one thing: that the current drawn by your micro was constant. It would be a very unusual circuit t that were true. But for the purpose in hand it is the way forward and allows you to get a pretty good approximation simply and without too much work.

The other thing is to get any information you can about the requirement before you start looking at solutions. Please don't think I am criticising, I m not. I'm just using this as an example so here goes: The first thing you should have done was to establish two things.

(1) what input voltage range will the micro operate with. When you told us that it made the analysis simple.

(2) what input current does the micro take. We still don't really know. I was wondering why you didn't measure it but perhaps you don't have a meter. (by the way you can get a pretty good basic meter for £5, in England anyway)

You might be interested to see the graph below which shows why the capacitor is a bad choice for this application. JimB pointed this out. Once again I have made the assumption that the micro takes a constant current. This is way off at the lower voltages but it illustrates the principle and to analyse it would serve no purpose. The black line shows what the battery would do and the blue line shows what the capacitor would do. Once again I cheated and changed the scale for the capacitor to illustrate the point. If I didn't the capacitor characteristic would be invisible because on that scale the capacitor would be discharged probably within the thickness of the y axis line.

I hope you didn't mind the lecture; I was just trying to explain why your question was on the wrong lines.

Although I have been preaching don't think electronics isn't fun and even when you know the rules you can still speculate and invent all you like- there are no restrictions.

 

[Dr_Doggy]

thanks all the help is great , i will stick with the aa s

I was just hoping for a indiana jones situation where i would just fit under the door.

the 5v panel i bought were ebay too, 500mW, not too sure about the 2.5v smaller ones, they were (salvaged).

here is link for boost converter:
https://www.electro-tech-online.com/threads/1v-boost-converter.142170/

I also tried one with discrete components, which was poorly designed and a bust as well

 

[Mikebits]

I was just hoping for a indiana jones situation where i would just fit under the door.

I thought that was Alice in Wonderland? So you did not like the Wind generator on a cap idea? People just don't get my humor...

 

[spec]

thanks all the help is great , i will stick with the aa s

I was just hoping for a indiana jones situation where i would just fit under the door.

the 5v panel i bought were ebay too, 500mW, not too sure about the 2.5v smaller ones, they were (salvaged).

here is link for boost converter:
https://www.electro-tech-online.com/threads/1v-boost-converter.142170/

I also tried one with discrete components, which was poorly designed and a bust as well

No probs,

You remind me of some of my exploits from way back

The AAs are by far the best way to go.

It seems to me that to get what you want solar panels charging the AA batteries would be a good avenue to explore. I have a load of old garden lights in a box in the garage. I suspect they have the bits to make a solar charger for your batteries. If you lived nearby I could have given you as many as you like.

Mikebits was having a bit of fun, but a wind generator is not a bad idea. You wouldn't necessarily need a conventional generator, just a coil of wire, a pemeanant magnet and a diode so it shouldnt cost you a thing if you can salvage the bits.

By the way, you say I helped you but you helped me in return: my knowledge about supercaps dates back about 15 years. By checking on them, I found that they have come a long way since then and are much cheaper too. I have also meant to find out about solar cells and now know a little about them. I will have a look at the link you posted.

(thank you for the like votes)

 

[spec]

I thought that was Alice in Wonderland? So you did not like the Wind generator on a cap idea? People just don't get my humor...

I did!