Cheap small 12V Li or Alkalines?

[Joerg]

Hello Folks,

I am looking for a battery that can deliver a few mA and would offer in
excess of 12V/50mAh. Should be small, <2" long and about 1/2" diameter
(or square).

The usual A23 is a bit wimpy here and the Energizer 504 is a bit high in
price. Ideally it should be <$1 in >1k qties and solder tabs (which
these two don't have). Low self discharge. Did anybody come across one?

Regards, Joerg

 

[[email protected]]

I am looking for a battery that can deliver a few mA and would offer in
excess of 12V/50mAh. Should be small, <2" long and about 1/2" diameter
(or square).

I am thinking of a stack of watch batteries. Since you want solder tabs,
there might be a problem in the application of keeping them in contact
with just spring pressure. On the other hand, I have seen batteries
that do appear to be made out of several coin cells, so maybe the
manufacturers have a jig that can reliably spot-weld the cases.

Browsing through Digi-Key, lithium coin cells might work:

Price (USD per assembly for 1000 assemblies)
mAh Qty Cell D x H * Pana Energizer
48 4 BR1225 12.5 x 10 2.45 ---
55 4 CR1616 16 x 6.4 2.12 2.13
55 4 CR2012 20 x 4.8 2.19 2.24
90 4 CR2016 20 x 6.4 0.64 0.69
165 4 CR2025 20 x 10 0.61 0.80

*Of entire stack

Keystone and MPD both have holders that will hold 2 of these coin cells.
Two holders will add anywhere from $1.08 to $2.32 to each assembly,
depending on how many you buy. With the CR2025 batteries you might get
away with about $1.70 per assembly.

Is this the one going in series with the power wire to the LCD, or
something else?

Matt Roberds

 

[Joerg]

Hello Matt,

Browsing through Digi-Key, lithium coin cells might work:

Price (USD per assembly for 1000 assemblies)
mAh Qty Cell D x H * Pana Energizer
48 4 BR1225 12.5 x 10 2.45 ---
55 4 CR1616 16 x 6.4 2.12 2.13
55 4 CR2012 20 x 4.8 2.19 2.24
90 4 CR2016 20 x 6.4 0.64 0.69
165 4 CR2025 20 x 10 0.61 0.80

Unfortunately most coin cells are for low drain applications only. Their
internal resistance is so high that the voltage begins to collapse at
low currents, sometimes at less than 1mA. Maybe for liability reasons, I
don't know.

Is this the one going in series with the power wire to the LCD, or
something else?

No, just for simple electronic circuits that can't easily work from a
low voltage cell. Sometimes a converter can be provided but other times
it's so cost critical that this extra 50c or so in cost isn't in the
cards. Often it is stuff that will only use one battery over the
lifetime so it won't pay to run through hoops to accept a CR123 or
something like that. Also, in most places a CR123 can cost as much as a
12V remote control battery and be less available at local stores.

Regards, Joerg

 

[redbelly]

Unfortunately most coin cells are for low drain applications only. Their
internal resistance is so high that the voltage begins to collapse at
low currents, sometimes at less than 1mA. Maybe for liability reasons, I
don't know.

How much voltage drop can you tolerate? I've measured a 10% drop for a
single 2016 Li battery drawing 19 mA. If you're drawing 3 mA (you said
"a few mA" in your OP), that works out to just a 0.2 V drop for a 12V
stack of 4 batteries.

You can also get a holder for 4 2016's from Digikey. Its 0.89"
diameter is bigger than you wanted, you can decide if you could live
with that or try a smaller diameter battery. It's overall height is
0.62", considerably less than the 2" you wanted.

The holder is Digikey part # BH800S-ND, cost is $0.43 each in
quantities of 1000. It has tabs that fit into a pc breadboard; they
are not what is known as a standard "solder tab" but they can be
soldered to nonetheless.

http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?Ref=325231&Row=270037&Site=US

Mark

 

[Joerg]

Hello Mark,

How much voltage drop can you tolerate? I've measured a 10% drop for a
single 2016 Li battery drawing 19 mA. If you're drawing 3 mA (you said
"a few mA" in your OP), that works out to just a 0.2 V drop for a 12V
stack of 4 batteries.

The 2016 would be ok. It was one of the batteries under consideration
but it would take four of them.

You can also get a holder for 4 2016's from Digikey. Its 0.89"
diameter is bigger than you wanted, you can decide if you could live
with that or try a smaller diameter battery. It's overall height is
0.62", considerably less than the 2" you wanted.

The holder is Digikey part # BH800S-ND, cost is $0.43 each in
quantities of 1000. It has tabs that fit into a pc breadboard; they
are not what is known as a standard "solder tab" but they can be
soldered to nonetheless.

http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?Ref=325231&Row=270037&Site=US

It seems this holder is for two so 12V would require a couple of them.
Not necessarily a problem but when I look at the data sheet it seems the
cells can easily slide out sideways or up.

Anyway, thanks for the hint. I'll check out holders some more.

Regards, Joerg

 

[redbelly]

It seems this holder is for two so 12V would require a couple of them.
Not necessarily a problem but when I look at the data sheet it seems the
cells can easily slide out sideways or up.

Anyway, thanks for the hint. I'll check out holders some more.

Regards, Joerg


It seems this holder is for two so 12V would require a couple of them.
Not necessarily a problem but when I look at the data sheet it seems the
cells can easily slide out sideways or up.

Anyway, thanks for the hint. I'll check out holders some more.

Regards, Joerg

Hello Joerg,

The holder is for two 2032's. 2016's are half the thickness, so twice
as many should fit.

More to the point: since I happen to own a couple of these holders, I
have now gone ahead and inserted four 2016's into one of them. The
holder and batteries are sitting here in front of me as I type this;
the batteries are held securely in place. (A thin wall surrounds most
of the holder's circumference, preventing sideways slippage. A metal
clip on the top prevents slipping out that way.)

I might as well test the assembly under load while I'm at it ...

Open circuit voltage is 12.75 V.

I added a 4k load (to draw 3 mA):
Right away, the voltage is about 12.5 V and continually dropping.
After 5 minutes, it's at 11.61 V.
After 10 minutes, it's at 11.39 V.
(Note, this has used up 0.5 mAh of the 80 mAh life rating.)

Don't know if that voltage drop is acceptable to you; does your
application have a continuous or intermittent load? Guess you'd need
to test them in your own device of course.

Mark

 

[Joerg]

Hello Mark,

The holder is for two 2032's. 2016's are half the thickness, so twice
as many should fit.

That's cool.

More to the point: since I happen to own a couple of these holders, I
have now gone ahead and inserted four 2016's into one of them. The
holder and batteries are sitting here in front of me as I type this;
the batteries are held securely in place. (A thin wall surrounds most
of the holder's circumference, preventing sideways slippage. A metal
clip on the top prevents slipping out that way.)

Thanks for the info. That wasn't very clear in the drawing. They showed
a thin wall but only half way around.

I might as well test the assembly under load while I'm at it ...

Open circuit voltage is 12.75 V.

I added a 4k load (to draw 3 mA):
Right away, the voltage is about 12.5 V and continually dropping.
After 5 minutes, it's at 11.61 V.
After 10 minutes, it's at 11.39 V.
(Note, this has used up 0.5 mAh of the 80 mAh life rating.)

I wonder why this droop occurs. Strange, because it seems that after
less than 5 minutes they went under their spec'd voltage.

Don't know if that voltage drop is acceptable to you; does your
application have a continuous or intermittent load? Guess you'd need
to test them in your own device of course.

It's mostly continuous. The voltage drop would be ok but not if this
continues at 0.3V every 5 minutes. Looking at the data sheet it could
just be the initial droop they all have:

http://rocky.digikey.com/WebLib/Energizer/Web Data/CR2016.pdf

Regards, Joerg

 

[redbelly]

Thanks for the info. That wasn't very clear in the drawing. They showed
a thin wall but only half way around.

It's a little over half way around, enough to hold things in place.

I wonder why this droop occurs. Strange, because it seems that after
less than 5 minutes they went under their spec'd voltage.

Yes, most batteries I've ever looked at run below the spec while under
load. With compactness and portability comes a lack of regulation on
the voltage.

It's mostly continuous. The voltage drop would be ok but not if this
continues at 0.3V every 5 minutes. Looking at the data sheet it could
just be the initial droop they all have:

http://rocky.digikey.com/WebLib/Energizer/Web Data/CR2016.pdf

Since the drop rate had slowed considerably between the first and
second 5-minute intervals, and given the graph in the Energizer
datasheet, it seems likely that the drop does not continue at this
rate. A test to verify this would be necessary, of course.

Good luck!

Mark

 

[Jeff L]

It's a little over half way around, enough to hold things in place.



Yes, most batteries I've ever looked at run below the spec while under
load. With compactness and portability comes a lack of regulation on
the voltage.



Since the drop rate had slowed considerably between the first and
second 5-minute intervals, and given the graph in the Energizer
datasheet, it seems likely that the drop does not continue at this
rate. A test to verify this would be necessary, of course.

Good luck!

Mark

I've abused CR2032's pretty heavy before, and they seem to last OK.
Application was a small demo card driving 16 LED's sequentially, where when
the button was pressed, each LED lit one at a time and stayed on. Cycle time
for all the LED's to light and then shut off a second after the last one lit
was about 4 seconds. A small AVR MCU controlled the thing, which drew just
under 200 mA when the last LED lit. Power was from 2 CR2032's in series, in
one of those 2 cell holders you mention. They lasted for hundreds of cycles,
the only thing that really drained them is if we forgot to take out the
batteries after use, since when I wrote the code, was having problems
getting the MCU to go to sleep and have an interrupt vector wake it up when
the button was pressed. No time to debug, so we just accepted a 4 mA
continuous drain.

 

[redbelly]

I'm using that same "power supply" (two 2032's) for a small home-built
photosensor. It provides the reverse bias to a photodiode. Everything
(batteries, photodiode, resistors, and 4-position switch) fits on a
1.5" square PC breadboard. All that's missing is a display for the
readout; must use a separate DVM for that.

Mark

 

[Joerg]

Hello Mark,

I'm using that same "power supply" (two 2032's) for a small home-built
photosensor. It provides the reverse bias to a photodiode. Everything
(batteries, photodiode, resistors, and 4-position switch) fits on a
1.5" square PC breadboard. All that's missing is a display for the
readout; must use a separate DVM for that.

In case you need a panel meter that can operate without the usual tight
tolerancing of 5V +/-0.25V check these out:

http://www.lascarelectronics.com/products.cfm?area=2&attribute=a&CFID=13128307&CFTOKEN=29465149

For example their SP200 runs from 3.5V to 5.25V and the SP400 from 3.5V
to 7.5V. Someone here in this NG had suggested them to me.

Regards, Joerg

 

[redbelly]

Thanks Joerg. It seems whenever I do a search on displays like this I
can't find anything under $50. At $18-20, these are more attractive to
a home hobbyist like myself. Making a small compact voltmeter just
might become my next project ...

Thanks,

Mark

 

[Joerg]

Hello Mark,

Thanks Joerg. It seems whenever I do a search on displays like this I
can't find anything under $50. At $18-20, these are more attractive to
a home hobbyist like myself. Making a small compact voltmeter just
might become my next project ...

When you call them they speak perfect rightpondian. Nice, haven't heard
that in a long time.

Regards, Joerg