PCB mounted coin cells

Digikey?

Only vendor is keystone.

MPD:

There are edge-on holders that are tall but small footprint.

What's wrong with lithium?

He could solder down a Tadiran lithium cell. The self-discharge time is 40 years or something crazy like that.

Supercaps are another option that uses very little PCB footprint.

"Silver oxide batteries become hazardous on the onset of leakage; this generally takes five years from the time they are put into use (which coincides with their normal shelf life). Until recently, all silver oxide batteries contained up to 0.2% mercury. Silver-oxide battery - Wikipedia

Are all the vertical-orientation holders THT, and you can't handle that in the project?

(You didn't mention a vertical restriction, so I'm assuming this would work.)

On a related subject, cell holders get rapidly more expensive as you drift away from CR2032. (The next most common smaller size, CR1220, is typically over double the price. DK prices, mind.) If you're cost driven, it's got to be CR2032.

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website:

=======================================================

HP used coin-sized mercury cells in voltage divider chains in some of their voltage "standards", such as the model 740. As I remember, they were mounted to switch terminals by nickel straps welded to the cells in NiCd battery pack fashion.

A suitable welder for this purpose isn't hard to build, and I think they're available off-the-shelf at reasonable prices. All you need to assemble one is a few large capacitors, a charging circuit, a high powered SCR and a couple of small spot welding tips. It's an added labor expense, but it does the job neatly.

Dave M

You got me thinking about supercaps.

Everything I have read is very positive. Seen them used in dash cams.

How do they compare with regular batteries?

Can they be charged with reg chargers?

Their capacity would be much less than batteries.

Since they are not used in cars, which would really reduce the weight, they must be a reason.

Andy

Never mind.

Found the answer here.

Andy

Badly. Like 1/100th the energy capacity badly.

But the power capacity is 10-1000 times, which is where the advantage comes in. They fill that odd niche between electrolytics (much faster, even lower density) and batteries (high density, very slow).

Most of the time, it's cheaper to just use more battery (to match the power requirement), and enjoy the extra lifetime.

Much as it's often cheaper to use a fat pile of electrolytic capacitors instead of a few (but bulky and expensive) film capacitors, to satisfy the same sort of tradeoff on shorter time scales (us to ms).

They can be used to start a car. Once or twice. After that, you'll be wishing you kept that battery. :)

Something that comes to mind this time of year: I wonder if they might serve well as a bypass, to improve a weak battery. But supercaps (and electrolytics) are just as bad in this sort of weather (it's been about -15C for the last couple weeks), so probably not.

Tim

Many thanks.

Sorry, these are the holders (Keystone 2988) I was referring to with a width of over 13mm

I had a good look at this website and they have quite a range of styles. Unfortunately they seem to largely cater for the larger cell size

The customer doesn't want lithium for a number of reasons. There are a number of issues.

I had looked into this but the self discharge rate is significant. I can't recall the leakage rates but I thought they would self discharge in less than a day. BICBW

Varta and others claim improved sealing. I'm also hoping the battery should only leak when discharged.

This has to fit in a tube so vertical space is also restricted.

Cost is secondary but not to the point of bespoke holders!

Size is very important, hence the choice of SR41 / SR48.

I tested one that seemed to have a complex time constant, more like weeks.

How much power do you need for how long, between charges? And what is the lifetime requirement?

Weld 2 wires to the cell and solder it to your boars. They do it every day on a large scale with NiMH, so why not with AgO?

Best regards, Piotr

s/s/d or the situation might get pretty interesting, train mad sprinting. You seem to have too small Hamming distance in English...

Best regards, Piotr

The datasheets imply a much shorter time. I also need 3V which limits the start and endpoint discharge.

uA with weekly intervals between charges, hence the steer to a primary cell.

I was wondering if there was a good reason why they weren't available on a commercial basis?

What's it powering? Many switched-capacitor boost ICs are at their most efficient at input voltages in the range of 2.5-3 volts, nearly 90%.

A supercap can do that. Self-discharge rates are numbers like hundreds of thousand of hours per volt.

There are 3.3 and 5 volt volt supercaps that would cost a fraction of a battery and holder. They would last the life of the product.

Any supercap that can supply uA over a few days, ie 1-2F so to get a modest voltage drop, has a leakage measured in mA, certainly the datasheets I've looked at. They are also extraordinarily dependent on past history of charge.

I'm not saying that some might last 10 years per volt, but I haven't found any hint that's the case from datasheets. Most are very cagey indeed.

Otherwise I would have like to have used one.

Small secondary cells also led to a dead end.

You might check out a few more. Digikey is a good start. I'm seeing caps that leak under 10 uA per farad.

Since leakage is about exponential on voltage, two lower-voltage supercaps in series would greatly reduce self-discharge at your voltage, and might even be good economics.