AVR PicoPower announced

Log in or register to post comments
Go To Last Post
16 posts / 0 new
Author
Message
lfmorrison
lfmorrison's picture
Level: Raving Lunatic
Joined: Wed. Dec 8, 2004
Posts: 4643 View posts
Location: Nova Scotia, Canada
#1
Posted by lfmorrison: Fri. Mar 17, 2006 - 07:12 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Just caught sight of this on the AVR website... They've announced a new "picoPower" line of AVR microcontrollers.

http://www.atmel.com/products/av...

Apparently it is a subfamily of AVR's will be released that offer significantly reduced power draw, both in Active and Sleeping modes. Active power saving schemes will apparently include both allowing more control over shutting peripheral clocks down (we've already seen that) and new ideas, such as shutting down the Flash for any periods of time within a given instruction cycle where the fetch is already completed.

One interesting claim is that a new "sleeping BOD" will allow the BOD to be automatically disabled during deep sleep modes, shaving 20 uA off the nominal power-down power consumption without compromising system integrity, brown-out detection response times, etc.

Another claim is that an AVR using picoPower technology will be able to drive 32 kHz RTC crystal in Power-Save mode resulting in total consumption (with a sleeping BOD) of 650 nA.

Planned devices will be drawn from the current ATmega family, with a P suffix after their names:
ATmega164P
ATmega324P
ATmega644P
ATmega48P
ATmega88P
ATmega168P
ATmega165P*
ATmega325P
ATmega3250P
ATmega645P
ATmega6450P
ATmega169P*
ATmega329P
ATmega3290P
ATmega649P
ATmega6490P

Sounds promising...

Tags:

AVR Microcontrollers, megaAVR and tinyAVR, ATmega16, ATmega164P, ATmega165, ATmega165P, ATmega168, ATmega168P, ATmega169, ATmega169P, ATmega32, ATmega324P, ATmega325, ATmega3250, ATmega3250P, ATmega325P, ATmega329, ATmega3290, ATmega3290P, ATmega329P, ATmega48, ATmega48P, ATmega64, ATmega644, ATmega644P, ATmega645, ATmega6450, ATmega6450P, ATmega645P, ATmega649, ATmega6490, ATmega6490P, ATmega649P, ATmega8, ATmega88, ATmega88P
Top
ka7ehk
ka7ehk's picture
Level: 10k+ Postman
Joined: Fri. Nov 22, 2002
Posts: 21554 View posts
Location: Tangent, OR, USA
Posted by ka7ehk: Fri. Mar 17, 2006 - 07:31 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Mouth is watering already!

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

Top
Guillem Planisi
Guillem Planisi's picture
Level: Raving Lunatic
Joined: Fri. Jun 11, 2004
Posts: 2405 View posts
Location: Vilanova (Barcelona)
Posted by Guillem Planisi: Fri. Mar 17, 2006 - 07:40 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yesterday I had received the visit of Atmel's sales man, and he told me that the 'official announcement' will be next monday. They plan to begin with LCD AVR's then extent to all of the family in the next months, including ATmega64P, ATmega128P, ATmega16P...

Promising, but with many more registers to play with to control power: shut off totaly the digital inputs, sleeping BOD, "zero power" RTC, new sleep modes, new wake up options, down to 0.6 or even 0.4 nA. at 1V8, etc.

Alkaline cell manufacturer's wouldn't be happy.

Guillem.

Guillem.
"Common sense is the least common of the senses" Anonymous.

Top
JBecker
JBecker's picture
Level: Hangaround
Joined: Tue. Nov 20, 2001
Posts: 255 View posts
Location: Germany
Posted by JBecker: Fri. Mar 17, 2006 - 08:56 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sounds good! Power consumption would then be in the region of an MSP430. If only Atmel would improve the ADC to at least higher speed or better 12 bit resolution, I would completely switch over to AVRs for extremely low power projects.

Jörg.

Top
Bronze Marktheusch
theusch's picture
Level: 10k+ Postman
Joined: Mon. Feb 19, 2001
Posts: 39897 View posts
Location: Wisconsin USA
Posted by theusch: Fri. Mar 17, 2006 - 09:43 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Power consumption would then be in the region of an MSP430.

Actually, a rough comparison of AVR "V" chips against MSP430 is very favourable to me. Are there any particular subsystems where the MSP430 beats the pants off the AVR in low power consumption?

[The one I know of is the brown-out detector.]

The AVR has areas that are better--such as a wider available supply voltage range.

One can open the old 8-bit vs. 16-bit thing, but in micro>>controller<< work (compared to some micro>>processor<< work where the 16-bitness is an advantage) the AVR compares quite well, especially with the ability to crank up the raw clock to higher frequencies.

Lee

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.

Top
nleahcim
nleahcim's picture
Level: Raving Lunatic
Joined: Wed. Feb 19, 2003
Posts: 2191 View posts
Location: Redwood City, CA
Posted by nleahcim: Fri. Mar 17, 2006 - 09:49 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hmm - instead of low power stuff I just wish they'd release faster AVRs, smaller AVRs with more RAM, and more CAN AVRs.

Top
Mikeharrison
Mikeharrison's picture
Level: Hangaround
Joined: Thu. Jul 26, 2001
Posts: 201 View posts
Posted by Mikeharrison: Fri. Mar 17, 2006 - 11:32 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

..so just like the nanowatt PICs then..;-)

..oh not quite, the PICs are actually available.

Top
lfmorrison
lfmorrison's picture
Level: Raving Lunatic
Joined: Wed. Dec 8, 2004
Posts: 4643 View posts
Location: Nova Scotia, Canada
Posted by lfmorrison: Fri. Mar 17, 2006 - 11:53 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Mikeharrison wrote:
..so just like the nanowatt PICs then..;-)

..oh not quite, the PICs are actually available.


Well, considering that the announcement hasn't even officially been made yet (according to Guillem) I think Atmel can be forgiven for not having any units available for purchase.

In fact, there are only datasheets available for two devices in the family.

And this section of Atmel's webpage was literally only put online *today* according to the "what's changed" section of the AVR Product line.

:P

Top
Plons
Plons's picture
Level: Raving Lunatic
Joined: Tue. Nov 1, 2005
Posts: 7035 View posts
Location: Hilversum - the Netherlands
Posted by Plons: Sat. Mar 18, 2006 - 12:54 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The sun always shines in the AVR-world

Nard

A GIF is worth a thousend words   They are called Rosa, Sylvia, Tricia, and Ulyana. You can find them https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

Top
JBecker
JBecker's picture
Level: Hangaround
Joined: Tue. Nov 20, 2001
Posts: 255 View posts
Location: Germany
Posted by JBecker: Sat. Mar 18, 2006 - 09:46 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Are there any particular subsystems where the MSP430 beats the pants off the AVR in low power consumption?

Maybe I am missing something, as I have never worked with a V chip, nor have I possibly read the datasheets thoroughly enough.
Is it possible to run a V chip with a 32kHz oscillator at a 'sleep' mode with less than 1uA? ('sleep' has a different meaning for different chips)
I have some applications where I use an MSP430 to simulate an RTC. This means that the 32kHz oscillator will be kept running, the chip will wake up once every second and increase the timer variable. Apart from that it will react to changes in some ADC-measurements and log data, but most of the time it will only do the RTC stuff. In these applications I get an over-all average consumption of <2uA at 3 to 3.2V supply voltage (Lithium cell) for an MSP430F133. The new MSP430F20xx should do this with less than 1uA and can be clocked up to 16MHz (although it will _not_ do anything near 16Mips at this speed! TI, please stop advertising that, it will only do NOPs at 16Mips.) .
A practical question, as you seem to have experience with the V chips: is it really possible to do the same stuff at a comparable power consumption with the AVR? I would appreciate an honest answer based on practical experience (no flames, we have had enough of that most recently!).

Jörg.

PS: But even if the answer is 'Yes, can be done with an AVR', I would still very much like to have the 12-bit ADC!!!

Top
pykedgew
pykedgew's picture
Level: Posting Freak
Joined: Mon. Jun 18, 2001
Posts: 1057 View posts
Location: Brisbane Queensland Australia
Posted by pykedgew: Sat. Mar 18, 2006 - 10:32 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Including all the above & would like to see a better internal VRef for the A/D.
Can it be more in the line like 1.024V, 2.048V or 4.096V with a 11bit or 13bit resolution without having to rely on an external chip & make it selectable internally by software.

Ken

Top
dl8dtl
dl8dtl's picture
Level: Raving Lunatic
Joined: Fri. Dec 20, 2002
Posts: 7506 View posts
Location: Dresden, Germany
Posted by dl8dtl: Sat. Mar 18, 2006 - 01:08 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

> Is it possible to run a V chip with a 32kHz oscillator at a 'sleep'
> mode with less than 1uA?

That should be possible, yes. Usually, you'll run the AVR on a higher
clock though while it is awake, and use the asynchronous timer to wake
it up between sleeps. The integral power consumption ought to be the
same (as while you are awake, you get the job done proportionally
faster), assuming you can use the internal RC oscillator for the main
clock so you don't have any startup delays. (Crystals oscillators
tend to stabilize very slowly, so you'd require a larger startup delay
for them, consuming power without really using it to compute
something.)

I did something similar (just for a test) with an STK504, that's the
new LCD starter kit, running an ATmega3290. I can't tell you for sure
anymore (I didn't record the measurement anywhere), but I think I got
the current consumption *including* the LCD well under 100 µA. The
scenario was basically the same, the 32 kHz timer (which is connected
to the LCD controller there) woke up the chip every once in a while.
Keep in mind that most of the current was needed to recharge the LCD
segments, and that the wake-up rate is related to the LCD frame rate,
so it's much higher than once per second. Given that, I think you can
already get into the 1 µA range without one of those `picopower'
devices right now.

What I did, however was to aggressively turn off everything inside the
AVR that was not needed for the job. Modern AVRs allow you to turn
off the clocking of various subsystems internally, thus conserving
more energy.

Jörg Wunsch

Please don't send me PMs, use email if you want to approach me personally.

Top
JBecker
JBecker's picture
Level: Hangaround
Joined: Tue. Nov 20, 2001
Posts: 255 View posts
Location: Germany
Posted by JBecker: Mon. Mar 20, 2006 - 09:07 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

My Butterfly draws about 10uA in power-down mode. What is it doing then apart from keeping the 'RTC' running?

Jörg.

Edit: I should not have mentioned 'Butterfly' here. This could easily start a new flame war :-)

Top
lfmorrison
lfmorrison's picture
Level: Raving Lunatic
Joined: Wed. Dec 8, 2004
Posts: 4643 View posts
Location: Nova Scotia, Canada
Posted by lfmorrison: Mon. Mar 20, 2006 - 06:28 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Jörg:
Nobody has experience with the new P parts (obviously) since they don'y exist yet. But apparently they will be able to operate in their "sleep" mode, still driving the 32kHz crystal, at 0.65 uA with 1.8V supply. Or 0.9 uA or so, at 3V and room tempearture.

We'll all be interested to get our hands on the real thing to verify these results I'm sure.

Contrast this with the Butterfly. According to the ATmega169V's datasheet, the 32kHz oscillator by itself consumes around 7.5 uA at 3 V. Then there are several voltage-divider resistor networks, etc on the butterfly that probably make up most of the rest of the drain.

Apparently the "new" 32kHz oscillator circuitry in the P series of AVRs is going to be much more energy efficient.

Top
Gold Markclawson
clawson's picture
Level: Moderator
Joined: Mon. Jul 18, 2005
Posts: 100401 View posts
Location: (using avr-gcc in) Finchingfield, Essex, England
Posted by clawson: Tue. Mar 21, 2006 - 10:56 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

SteveN wrote:
Quote:

ATmega164P
ATmega324P

I wonder if this means the mega164/324 are not dead after all (although their death was never confirmed anyway - I believe) or they will only be available in the "P" version?


Steve more evidence for the existence of 164/324 comes in SP2 for AVR Studio that was released a few days ago - as you'll see in the release notes (http://www.atmel.no/beta_ware/as4/412sp2/Releasenotes.txt) one of the changes is added 164P/324P support.

Cliff

Top
dl8dtl
dl8dtl's picture
Level: Raving Lunatic
Joined: Fri. Dec 20, 2002
Posts: 7506 View posts
Location: Dresden, Germany
Posted by dl8dtl: Tue. Mar 21, 2006 - 01:28 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

> Nobody has experience with the new P parts (obviously)
> since they don'y exist yet.

Well, samples do. I've got a sample of an ATmega165P, but
I'm not sure whether I could really measure something
related to the `P' device within an STK501. Anyway, if you've
got something you want me to measure, tell me (preferrably
in privte email).

Jörg Wunsch

Please don't send me PMs, use email if you want to approach me personally.

Top