[time-nuts] GPSDO on the cheap

David davidwhess at gmail.com
Thu Dec 6 19:01:49 UTC 2012


In past designs I just included an EEPROM so in the event of a cold
start, the last settings would be known.

On Thu, 6 Dec 2012 10:45:48 -0800 (PST), "M. Simon"
<msimon6808 at yahoo.com> wrote:

>I would use a digital pot for coarse setting. Or a manual trimpot. That way your control signal "holds" even if your comparison goes away or if for some reason your loop comes out of lock. Something like that also reduces the noise contribution of the DAC.  
>
>Simon
>
>=============
>
>Date: Thu, 6 Dec 2012 01:17:52 -0800
>From: Chris Albertson <albertson.chris at gmail.com>
>To: Discussion of precise time and frequency measurement
>    <time-nuts at febo.com>
>Subject: Re: [time-nuts] GPSDO Alternatives
>Message-ID:
>    <CABbxVHs_XiBAX15h27c38L4gEvmMGNzWO+9Fw2J-BQG9k5Z_jQ at mail.gmail.com>
>Content-Type: text/plain; charset=ISO-8859-1
>
>Don't need anything so complex.   A GPSDO depend on an OCXO that is VERY
>stable.  It can be controlled with a very low bandwidth analog signal.
>
>The output from a OCXO is divided down and then the phase of the divided
>down 10MHz RF is compared to the PPS and you don't need to even know the
>how far apart they are.  All you need to know is "led or lag"  just a one
>bit answer.   An XOR gate or a flip flop can tell you that.
>
>If you want to get more fancy you can connect a few temperature sensors to
>the Arduino's ADC lines and push those over the USB port along with other
>statistics.   Youcan also do things like control the time constants the
>software uses via USB also.  But you don't need this.   It can be added
>later or not.



More information about the time-nuts mailing list