AVR-Chat

Hello, David -

No direct experience on these chips, but with some RF hardware, generally.

If you measure the output frequency, you will need to insure that none of
the frequency hopping is going on. That MAY require different code for
testing or a special "test mode" with all that turned off. Testing the
output does insure that the PLL multiplier is what you expect it to be,
however, I would think that prototype testing should insure this and that
production testing of this parameter might not be needed.

Testing the output frequency could also be problematic if you have a fixed
antenna. You would probably need a second antenna and maybe a broadband
amplifier to get enough signal to count. If the antenna port has a
connector, then its not so much of an issue.

Measuring the 16MHz reference would depend on whether or not there is a
buffered output version of that signal. I would NOT connect directly to
one of the oscillator pins as that will definitely shift the frequency.
10ppm is 10Hz/MHz so, applied to 16MHz, that means tuned within 160Hz.

Jim Wagner
Oregon Research Electronics

> I'm interested to know, from anyone who actually has this device in
> production, how you are doing frequency setting.
>
> I'm thinking in terms of a counter driven by IEEE488, and some sort of
> automated process to tweak the on-chip cap values to dial in the
> oscillator.
> It's not clear to me whether it is better to measure the 16 MHz test
> pin, or the RF carrier, or both.
>
> Achieving 10ppm accuracy out the door is the end goal.
>