High current drain from lithium coin cell

Go To Last Post
11 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi guys,

 

I'm in the process of designing a little infra-red keyfob transmitter that is to be powered from a CR2032 coin cell.  My concern is how best to deliver the high current pulses to the IR LED.

I have re-purposed another XmegaE5 board I have, to prototype the circuit and develop the software.  This has a little piezo beeper on baord that quite expectedly loads the battery up far too much if activated, causing the voltage to drop and trigger BOD and even POR.  None-the-less, if I don't use the piezo, I'd still like to run the IR LED at something like 20-40mA with range vs battery life the obvious trade off.  (Incidentally, its using RC5 and I have reduced the pulse widths to about 25% for power saving.)

The prototype fortunately has a decent (22uF Al Electro) capacitor connected directly across the battery which supplies the high pulse currents.  Is this likely to be the best way to do it? I'm most likely going to switch to a tantalum and an MLCC or 2 in the final design for lower leakage.  Just wondering what other folks think about this.  Is there a better cap chemistry to use? Maybe a supercap? Perhaps an alternative method?

 

Cheers,

 

Steve

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Things to have in mind:

 

when you have 20-40mA what is max AVG current over a packet?

How long is the time between (to charge in).

If you cap is too good the voltage will drop and perhaps reset the micro!

If to bad(small) the voltage drop at the end of the packet.

 

What do you do when the battery is about at the end

turn current down or have a longer pause between packets?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

What's the BOD?

For XMEGA E5 with minimum BOD, 1.8V should be reliable.

Could do a buck-boost constant current source to match IR LED during discharge over CR's lifetime.

Embedded

Coin cell voltage droop

by

September 15, 2014

http://www.embedded.com/electronics-blogs/break-points/4434661/Coin-cell-voltage-droop

...

The blue line is the battery’s loaded voltage (at t=0 for 30mA); other lines are the internal resistances during the one second interval:

...

The bottom line is that the voltage the MCU sees when it first wakes up is not that which will be there mere milliseconds later. Figure on another 10% drop, on top of all of the serious losses I’ve detailed in those earlier articles.

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

schtevo wrote:
Perhaps an alternative method?
Alternative chemistries :

  • Li-ion
  • Vanadium oxide
  • Manganese oxide

Rechargeable (secondary) cells have a greatly reduced internal resistance than primary cells for a given size.

The pickup coil for wireless recharging might fit in a keyfob; maybe the electronics won't consume too much area.


http://www.mouser.com/new/illinois-capacitor/illinois-capacitor-rjd-batteries/ (Li-ion coin)

https://na.industrial.panasonic.com/products/batteries/rechargeable-batteries/rechargeable-coin

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Perhaps your IR communication just needs you to flash the IR?

 

Maybe you could charge a small capacitor through a resistance, then use a mosfet to dump the capacitor through the led to make the flash. This might work well for systems that need 38KHz infrared pulses.

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for those responses. Sorry I didn't get back sooner; Australia Day frivolities kind of got in the way.

 

sparrow2 wrote:
when you have 20-40mA what is max AVG current over a packet? How long is the time between (to charge in). If you cap is too good the voltage will drop and perhaps reset the micro! If to bad(small) the voltage drop at the end of the packet. What do you do when the battery is about at the end turn current down or have a longer pause between packets?

All fair points. Since at design stage all this can be adjusted to suit.  At the moment, standard (or reduced duty cycle) RC5 is working fine including a couple of LEDs (<1mA) with the 22uF.  My main concern is the technique and leakage current.

 

gchapman wrote:
For XMEGA E5 with minimum BOD, 1.8V should be reliable. Could do a buck-boost constant current source to match IR LED during discharge over CR's lifetime.

I can't say I have yet experimented the BOD threshold. I believe it is set at the 1st step below 3.0V. Maybe 2.2V. Will have to check.

At the moment I am using a basic constant current driver as per AVR415 with current set to ~50mA.

I was steering away from using a switcher to reduce Iq.  As it stands the whole circuit runs directly off the battery and goes completely to sleep until a button is pressed.  Though this gives me the idea of using something like an LTC3105 MPPT regulator.  Its designed to work with high impedance sources that can collapse under load. Maybe if it has a shutdown or enable control I could get it to fire up on key-press, switch off when done, and then not use the sleep mode.

 

gchapman wrote:
Alternative chemistries : Li-ion Vanadium oxide Manganese oxide Rechargeable (secondary) cells have a greatly reduced internal resistance than primary cells for a given size. The pickup coil for wireless recharging might fit in a keyfob; maybe the electronics won't consume too much area.

Not a bad idea. I actually have plenty of real estate spare. The height is of some concern however.

 

Torby wrote:
Perhaps your IR communication just needs you to flash the IR? Maybe you could charge a small capacitor through a resistance, then use a mosfet to dump the capacitor through the led to make the flash. This might work well for systems that need 38KHz infrared pulses.

Are you thinking along the lines of the old camera flash bulbs? Novel. I wonder if its overly complex to do that in real time with the IR pulses? Perhaps more like the flash were a cap its 1st charged up (in a short time) then discharged into the LED as it pulses.  With a constant current driver I guess as long as the cap has sufficient charge to send the whole frame it could be ok. Then recharge in the time between frames? I think the batteries internal resistance may not allow the cap (bank) to change up fast enough though. I shall consider this more.

 

Thanks again guys.

 

Steve

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

schtevo wrote:
At the moment I am using a basic constant current driver as per AVR415 with current set to ~50mA.
Stout IR LED

http://www.atmel.com/devices/ATtiny816.aspx?tab=documents

...

AVR415: RC5 IR Remote Control Transmitter on tinyAVR and megaAVR

AVR415: RC5 IR Remote Control Transmitter on tinyAVR and megaAVR devices

AVR415: RC5 IR Remote Control Transmitter on tinyAVR and megaAVR
(file size: 253KB, 13 pages, revision B, updated: 08/2016)

In this application note the widely used RC5 coding scheme from Philips will be described and a fully working remote control solution will be presented. This application will use the ATtiny28 AVR microcontroller for this purpose.

...

schtevo wrote:
I was steering away from using a switcher to reduce Iq.
Most power supplies are sub-microamp Iq when disabled.

Alternate AVR415 RC5 IR LED

First try at a proposed SMPS CCS LED driver; quiescent current is too much for a CR2032 :

Discrete, Analog and Logic Semiconductor Components - Diodes, Inc.

Low Voltage DC-DC LED Drivers

ZXSC400

http://www.diodes.com/catalog/Low_Voltage_DC-DC_LED_Drivers_70/ZXSC400_1735

...

  • 4.5μA typical shutdown current

...

 

Edit : ZXSC400

 

"Dare to be naïve." - Buckminster Fuller

Last Edited: Tue. Jan 31, 2017 - 02:54 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It's very commen to add an Elco of a few 1000 uF in IR remote control gadgets.

Whether it's needed?

Do some performance tests with some partially discharged batteries...

Doing magic with a USD 7 Logic Analyser: https://www.avrfreaks.net/comment/2421756#comment-2421756

Bunch of old projects with AVR's: http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Paulvdh wrote:
It's very commen to add an Elco of a few 1000 uF in IR remote control gadgets.

 

Haven't had the fortune to see any IR remote designs that run of coin cells but I know they do exist. My heater has one! Sadly I cant open it up without destroying it.

I have however taken another close look at the eZ430-RF2500 solar energy harvesting development tool and a 1000uF tantalum is precisely what they have used to supply the high current pulses for the radio module.

 

Further research has revealed that it is not at all difficult to obtain smaller (say 10,22,33uF) tantalum caps with leakage currents in the 0.5-2uA range.  This beats the ZXSC400 switcher design (4.5uA in shut down) though I expect maybe some benefit in stepping up when the battery voltage is getting low.  Mind you, in this application, the device will be idle/sleep most of the time and I suspect the lower sleep/leakage will be more beneficial to battery life than the step-up will.

And finally, MLCCs have ridiculously low leakage currents to the point where they change to specify as insulation resistance in the order of Gohm.  Right now I think the best approach would be to stick a few paralleled 10uF 1206s there and if they prove insufficient, replace with a 3216 tantalum.

At a last resort, a simple P-ch MOSFET load switch to the cap & IR LED driver could reduce the Iq to <1uA.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

schtevo wrote:
And finally, MLCCs have ridiculously low leakage currents ...
Some MLCC are better than others on leakage.

Hardware and Firmware Issues in Using Ultra-Low Power MCUs

by Jack Ganssle

December 2014

...

5 - Decoupling, and Using a Capacitor to Boost Vdd

http://www.ganssle.com/reports/ultra-low-power-design.html#usingacapacitor

schtevo wrote:
At a last resort, a simple P-ch MOSFET load switch to the cap & IR LED driver could reduce the Iq to <1uA.
... at room temperature (approx)

FETs leak though there are some FETs that are reasonable on leakage.

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes yes. I just got thru reading that Jack Ganssle article you posted earlier. Phenomenal stuff! Thanks. (That new one looks like the same stuff reformatted, no?)

The rechargeable coin cell option is looking better all the time!  Maybe Rev2; I'm getting impatient. sad

 

Time to get serious and do some calcs / run some tests me thinks.