[time-nuts] frequency sources
Lux, Jim (337C)
james.p.lux at jpl.nasa.gov
Fri Oct 16 03:10:18 UTC 2009
On 10/15/09 7:20 PM, "Don Latham" <djl at montana.com> wrote:
> Some list members indicated a need for a well-controlled easily set
> frequency source. Another list I get has to do with software defined radio
> (softrock40). A member of that list mentioned a Maxim chip, the DS1077
> oscillator/divider. The chip plus breakout board is close to ten bucks
> from Sparkfun. This chip will put out two frequencies and is I2C
> programmable. Over its temperature range, it is goodto +- 1.4% of its
> fundamental frequency.
We have some similar parts here at work, and the output waveform may not be
very symmetric or even consistent from cycle to cycle. It's sort of like
the problems with clock oscillators designed to reduce EMI by dithering the
period to "spread the spurs"...
Even running it into a divide by 2 (which would make a symmetric output)
doesn't necessarily work if it's using some sort of funky divider scheme
where only the "average period" meets the requirement.
I'm not familiar with that particular part, but it's something to be aware
More information about the time-nuts