ATtiny861A low freq. power consumption

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

Hi,

Does anyone around here have experience with attiny861A current consumption, when running at very low clock frequency?

In my "invention", I use a tiny861A, clocked with a 32.768 kHz crystal, and running on 3 V (actually two AA cells in series).

When the microcontroller is active, and no other peripheral devices are switched on (there aren't that many anyway), the current consumption is around 93 - 97 µA, which is about 50 percent more than I'd expected.

When looking at the datasheet, page 251, the graph shows that the current consumption at 3 V, 100 kHz, should be about 40 µA. I'd be very surprised if the microcontroller would consume more current at a lower clock frequency.

The only "I/O module" that is switched on, is Timer 0, which should increase the current consumption with about 1 µA (next to nothing). Also, BOD is on, which can further increase the current consumption with 18 µA (page 259). Pull-up resistors are switched off.

All in all, I would expect a current consumption of 60 µA or less. Certainly not above 90 µA. What could I have forgotten?

When I put the microcontroller in deep sleep mode (BOD off), the current consumption is well below 1 µA, so that is fine.

Br, E

You're absolutely right. This member is stupid. Please help.

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

Perhaps run the same test using internal oscillator divided way down? Maybe it takes about 50uA for the crystal circuit?

BTW, when I do the power budget for AVR8 apps, it usually ends up better to "run like hell" (e.g. 8MHz) to get the work done and get back to sleep faster, versus plodding along "forever".

If you need the accuracy of the clock crystal, people have had good results on models with timer2 clocked from TOSC pins, but I don't think your model has that feature.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

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

Quote:
Perhaps run the same test using internal oscillator divided way down? Maybe it takes about 50uA for the crystal circuit?

Yeah, I've had the same thought. Haven't seen any info on the crystal circuit current consumption, though.
Quote:
BTW, when I do the power budget for AVR8 apps, it usually ends up better to "run like hell" (e.g. 8MHz) to get the work done and get back to sleep faster, versus plodding along "forever".

I agree. But this device is actually doing something all the time (except when it is turned off).

I need the accuracy of an actual crystal. Otherwise, I'd definitely used the internal RC ocsillator.

Ah, well. Luckily, it is not a show stopper. Just need to change batteries a bit more often than I'd hoped.

You're absolutely right. This member is stupid. Please help.

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

Quote:

I agree. But this device is actually doing something all the time (except when it is turned off).


Let's explore that for a minute. (And this is in fact the test case for "run like hell"...)

What is the something it is doing all the time? Continuous A/D conversions? That isn't a recipe for lowest power consumption, but in any case it is some hundreds of microseconds per conversion--good time for a nap.

Watching pins for events? Again, good time for a nap. I use a 10-ms debounce period, and wake up on the timer tick. Instant response? Pin-change interrupts.

The calculations with the above? Run like hell, then take a nap.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

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

Sorry about leaving you in the dark for more than a month, Theusch. It is a sort of advanced "alarm clock" that our production people need. Can't tell you for what, though.

That's why I need the crystal. I need an accuracy in the vicinity of an ordinary clock. It doesn't need to be extremely accurate, but an RC oscillator is not good enough.

And no, an ordinary alarm clock can't be used for this purpose. Otherwise we'd use one. :wink:

Anyway, the stuff seems to work nicely, and production likes it. So I'm happy.

Br, E

You're absolutely right. This member is stupid. Please help.