[time-nuts] ngTADD-1

[Chris Albertson]

The problem with analog 10MHz is the non-linear parts at each end of
the fiber optic.  It would be hard to get a nice sine wave at the
output.

I think if you have a requirement for very clean sine wave at the far
end you'd best try to phase lock a local XO.  Send a 10 MHz square
wave down the fiber and use that to phase lock a local XO.    I think
almost all fiber systems use a PLL for clock recovery.

Problem is the parts built into the jacks.

On Thu, Apr 25, 2013 at 12:08 AM, Attila Kinali <attila at kinali.ch> wrote:

> Sorry, i was a bit unclear. What i meant with modulation was to
> use an analog scheme that never saturates the transmitter or receiver
> both in the max (fully on) or min (fully off) direction.
> Ie do not send 1's and 0's as it is common in the digital domain of
> networking, but an "analog" 10MHz sine. With that you get around
> of the non-idealities that come with on-off transmissions.

--

Chris Albertson
Redondo Beach, California