xmega 32MHz internal oscillator accuracy

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

Hi,

I've been reading all I can here and in atmel's app notes about the accuracy of the internal clock sources.

I see in app note avr1003 that the xmega's internal 32MHz oscillator is accrate to 1% at 3V and 25 degs C. The doc also says "The oscillator can be further tuned and calibrated using the auto-calibration feature covered in section...".

Which brings up two questions I haven't been able to answer:

1. Does atmel spec the xmega's internal oscillator accuracy over temperature changes? I'm interested in accuracy as ambient changes from about 5 to 32 C.

2. What is the accuracy with use of the internal oscillator with use of the auto-calibration feature?

Does anyone know? And if you can please tell me where you found it I'll be better off for next time.

Thanks.

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

1) there's certainly a graph of that in tiny/mega datasheest - does the X data not have the similar characteristic graphs?

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

clawson wrote:
1) there's certainly a graph of that in tiny/mega datasheest - does the X data not have the similar characteristic graphs?

Not that I've been able to find.
Note that the xmega A family docs are still marked "preliminary".

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

steverino wrote:
I see in app note avr1003 that the xmega's internal 32MHz oscillator is accrate to 1% at 3V and 25 degs C.
When I measured an xmega128A1 on an xplain board at room temperature the frequency was drifting between 32.3 MHz and 32.5 MHz. I used two separate instruments to confirm this.

I later learned that Atmel shipped xmegas without ADC factory calibration. I didn't check if the factory calibration of the 32M oscillator was also broken. A broken factory calibration would be an explanation for the deviation I measured.

I didn't go through any external calibration.

It should get better with DFLL, but that is partly broken (see the errata for a workaround).

Quote:
1. Does atmel spec the xmega's internal oscillator accuracy over temperature changes? I'm interested in accuracy as ambient changes from about 5 to 32 C.
The information for the 32.768 kHz and the 32 kHz oscillators are somewhere in the A1 datasheet. The one for the 32 MHz apaprently not.

Quote:
2. What is the accuracy with use of the internal oscillator with use of the auto-calibration feature?
That should be in the A1 datasheet. The DFLL, if you get it to work, should be able to adjust the oscillator in 0.2% steps, plus, of course, the DFLL's reference oscillator accuracy.

Stealing Proteus doesn't make you an engineer.