Watchdog timer accuracy

Anybody have any experience with how accurate the WDT is? I was wondering about using it for a general timer, albeit limited in time choices.

Thinking about using the 125 mSec setting for multiple second time periods. A couple mSec drift on 30 second periods won’t be a problem.

Have you read the datasheet?

MikeLittle:
Anybody have any experience with how accurate the WDT is? I was wondering about using it for a general timer, albeit limited in time choices.

Thinking about using the 125 mSec setting for multiple second time periods. A couple mSec drift on 30 second periods won't be a problem.

We will be interested in the seeing the results when you have tested this, but expecting the watchdog timer to be stable to within 2ms over 30 seconds seems weird.

srnet:
We will be interested in the seeing the results when you have tested this, but expecting the watchdog timer to be stable to within 2ms over 30 seconds seems weird.

I'm just saying that I don't plan on using it for any critical timing control.

I saw someplace that a RTOS was using the WDT to generate the time slices. So it must have some kind of stability to it.

So it must have some kind of stability

Yes, and that information is presented in the data sheet. The watchdog oscillator frequency depends rather strongly on temperature and supply voltage.