Supercapacitors

I've never used a supercap before, though I have had one kicking about here for a long time, a 5v 1 Farad. I have an old code entry lock, which I am looking to install, which runs on 5v and needs a battery to maintain its code memory at 5uA.

I don't expect the mains supply to ever be off for long, but with no supply at all available to the lock, it reverts to a not very secure entry code of '0'.

Would the cap do the job? Anyone any experience of using them?

Reply to
Harry Bloomfield
Loading thread data ...

The R/C constant for voltage decay doesn't magically change just because you're using "supercapacitors" - just do the math as usual and work it out for yourself as if you were using common-or-garden electrolytics.

Reply to
Cursitor Doom

Erm what is wrong with a good old battery like the ones in computers for the comes?

Brian

Reply to
Brian Gaff

Brian Gaff submitted this idea :

I just thought I had one on the shelf, put the cap to some use.

Reply to
Harry Bloomfield

Cursitor Doom formulated on Sunday :

I'm just not sure how the self discharge works, I have no experience of these caps.

Reply to
Harry Bloomfield

Well you need to do the sums. How much charge is stored in it when its up to 5v? Then, at what lower output voltage will the code memory lose its data? How much charge does that equate to? So your "available charge" is the difference, and given you said 5µa, you can work out how long you can get that current before the output voltage drops too low.

I expect the formulae you need are in Winky somewhere.

Reply to
Tim Streater

formatting link

Reply to
dennis

Charge in joules is, IIRC, ½CV²...

Reply to
The Natural Philosopher

Er, that's the stored energy. The charge Q (in coulombs) is simply CV. For the example given, the available charge is C(Vmax - Vmin). As a first order approximation, and ignoring any self-discharge, it would seem reasonable to assume a linear decay rate, so I = C(Vmax - Vmin)/T. (From I = dQ/dt). Rearrange to get T = C(Vmax - Vmin)/I.

Call yourself a 'natural philosopher'? Sheesh!

Reply to
Custos Custodum

He can't. First it's necessary to know V_min for continued remembering and what v the cap charges to. I don't suppose he knows those figures.

NT

Reply to
tabbypurr

Hmm well the cmos batteries seem to be being sold in packs of 2, seems a little odd since by the time you need the second one either the machine will be too old or died.

Brian

Reply to
Brian Gaff

but all that is trivially skirted. Fit your supercap & see how long it remmebers

NT

Reply to
tabbypurr

I don't think it matters if his drain figure is correct as it will stay at 5V -5% for days. It will probably drain more if the weather is humid than the device uses.

Reply to
dennis

The CMOS battery in my laptop has just gone and that's a usable i5 two core machine. So they don't outlast the machines these days.

It was only a 1822 or similar battery though, not rechargeable.

Need to buy one and solder some wires on or keep it in standby mode so it doesn't forget.

Not that it matters if it forgets its still boots and windows corrects the time from the internet.

Reply to
dennis

I have used them, but not for a while. From memory, the self-discharge rate is quite high. The main benefit of supercaps is the fast rate of charge and the high number of charge/discharge cycles.

You say it is 1 Farad at 5v - from memory again (not always reliable these days!) supercaps operate at 2.7v.

Reply to
Farmer Giles

1F @ 5F is 12.5J or 12.5 watts for 1 second. Of course, we don't know how much of that can be used as we don't know what voltage drop your lock can bear. It uses a battery at present which is likely to be less than 5V I'm guessing.

So let's say we can only use 4% of that energy which seems very conservative and leaves us with 0.5 Ws.

5uA at 5V is 25uW, so 0.5Ws would give you 20,000 seconds or 5.5 hours.

I've no idea what the self discharge of these things is, but it will be much less than 25uA, so it sounds like a useful thing to do.

Cheers

Reply to
Clive Arthur

I assumed what he was inteersted was the energy in the charge not the couloumbs

In my whole career as an electronic engineer I never once used coulombs for anything.

Golly, what a superb skill in cut and pasting you display..

Yes, I do.

Reply to
The Natural Philosopher

snipped-for-privacy@gmail.com explained on 22/07/2019 :

V_min is 3v, V_max is 6v, but it will have and charge the supercap to a rail of 5v. I'm thinking connect the supercap directly across the rail. I'm just trying to avoid the need to introduce a battery for backing up the settings. Its a UA3730, I am trying to repurpose as a code entry lock system, on my garages side entrance - my original code lock has become unreliable and lacking a circuit, I have not been able to work out why..

Reply to
Harry Bloomfield

Clive Arthur wrote on 22/07/2019 :

I suppose I could just charge it up to 5v and see how long it takes to drop to the 3v V_min of the IC.

Reply to
Harry Bloomfield

Yes, nothing like a good experiment, after all it may be duff. Make sure your meter doesn't load it too much, and be aware that it'll probably discharge more quickly at higher temperatures.

Cheers

Reply to
Clive Arthur

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.