Supercapacitor disconnected by software? Possible & Usef

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

Hi all, I recently got an idea/problem concerning the use of supercapacitor and battery.

I have a system wich will be powered by a high capacity battery. Its capacity is dependent on the current you draw from it, so I'd like to keep it as stable as possible: since my system consumes ~20mA I'd like to charge a supercapacitor previously for some time so the current drawn from the battery is lower (~5mA or less).

The problem is that the system is, most of the time, in idle/sleeping and its consumption current should be less than 1µA and since the supercapacitor draws ~8µA of current I'd like to disconnect it as soon as the system gos into 'sleep state'.

I'm using an AVR in deep sleep mode.
Is this feasible?
Is it even a good idea?

Alternatives?

The whole point is to achieve the maximum working time from the battery.

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

How long does the MCU have to draw the 20mA continuously? Can you chop that usage into 5 msec intervals with short delays in between that would allow a low leakage capacitor to charge back up?

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

After reading this http://www.embedded.com/electron..., I'm starting to think that using a capacitor is not a good idea.

And I have to consider the decoupling capacitor which I haven't.

dak664 wrote:
How long does the MCU have to draw the 20mA continuously? Can you chop that usage into 5 msec intervals with short delays in between that would allow a low leakage capacitor to charge back up?

It's only during transmitting wirelessly, so yes, with the right software algorithm, it shouldn't be hard to do so.
But then, it's the same problem: how do I disconnect the capacitor to prevent the impact of it's leakage current when the device is in sleep stage (keeping the consumption to a minimum).

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

A p-channel mosfet.

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

Quote:
to prevent the impact of it's leakage current

Look at the spec's carefully.

Disconnecting the cap won't impact its internal discharge rate.

Be careful of what you are calculating, also. If the cap is disconnected, for whatever reason, then it will still take battery energy to charge it back up to the battery's voltage every time it is reconnected.

JC

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

tglaria wrote:
... a high capacity battery.
What's the battery's estimated self discharge current?
tglaria wrote:
Alternatives?
Use an external RTC and regulator/PMIC/FET/etc to duty cycle the AVR; AVR controls the RTC for the duty cycle's period.
The RTC's current is about one to two orders of magnitude less than the AVR's sleep current.
Abracon makes these.
Any others?
http://www.abracon.com/news.php (search for January 30, 2013 and the RTC announcement)

"Dare to be naïve." - Buckminster Fuller

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

DocJC wrote:
Quote:
to prevent the impact of it's leakage current

Look at the spec's carefully.

Disconnecting the cap won't impact its internal discharge rate.

Be careful of what you are calculating, also. If the cap is disconnected, for whatever reason, then it will still take battery energy to charge it back up to the battery's voltage every time it is reconnected.

JC


Yes, I'm aware that it takes energy while charging, but I'm considering that the time the system spends 'sleeping' makes up for that wasted energy.

gchapman wrote:
tglaria wrote:
... a high capacity battery.
What's the battery's estimated self discharge current?

According to Tadiran (which is the battery I pretend to use) their self discharge rate is less than 1% per year.Here
Yet, don't know how to convert that into a self-discharge current.

gchapman wrote:
tglaria wrote:
Alternatives?
Use an external RTC and regulator/PMIC/FET/etc to duty cycle the AVR; AVR controls the RTC for the duty cycle's period.
The RTC's current is about one to two orders of magnitude less than the AVR's sleep current.
Abracon makes these.
Any others?
http://www.abracon.com/news.php (search for January 30, 2013 and the RTC announcement)

I'll start reading about this, thanks.

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

tglaria wrote:
... their self discharge rate is less than 1% per year.
Some data sheets state a lifetime at a low discharge of 10 microamps.
Likely simply connect the AVR to the cell.
Cell may have a large enough pulse current capability that a large capacitor is not needed (iow use Atmel's recommendations for an AVR power supply).
The leakage of a large MLCC might be acceptable for this kind and size of cell; might be able to use a few kinds of supercapacitors but maybe not.
Ramping the AVR's frequency might work.
http://www.powerstream.com/p/ER13150.pdf
Running fast to save power by Jack Ganssle (embedded.com; May 05, 2014)

"Dare to be naïve." - Buckminster Fuller

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

gchapman wrote:
tglaria wrote:
... their self discharge rate is less than 1% per year.
Some data sheets state a lifetime at a low discharge of 10 microamps.
Likely simply connect the AVR to the cell.

Well that's the idea, to reduce to a minimum the consupmtion while the device is sleeping.