[time-nuts] Clock Driver Design

[Bob Camp]

Hi

How clean is your clock source? If you have something that is -150 dbc at 1 Hz, then you probably need some fancy analog gizmos. If you can make do with "only" -110 to -120 dbc/Hz at 1 Hz, then properly driven LVC CMOS will do just fine. That's true for a square or a sine output. Since you pretty much can't find an OCXO better than -120 at 1 Hz, I'd bet you'll be ok. 5 volt logic will be a little more quiet than 3.3V. More or less faster is quieter as long as you stay with saturated silicon CMOS. Change materials and all bets are off. 

For square wave cable drive you can parallel up a couple of the '125 or '126 gates to get how ever much power you want to put into the cable. You can source or load terminate (or both). If you source and load terminate, your logic levels will be 1/2 the output. With either source only or load only termination you can get full swing logic levels. More drive will always be required with load termination (you are putting current into 50 ohms). 

Logic IC's are cheap, easy to use, and simple to find. A low voltage single supply drives them and they aren't current hogs unless heavily loaded. What's not to like?

Bob


On Sep 26, 2013, at 1:28 PM, Tom Minnis <Tom_minnis at att.net> wrote:

> I am looking into various degrees of craziness.  The source is CMOS and there are plenty of 1 in to N out parts designed to drive clocks on a PCB but not much is said about driving clocks on to a random length of coax to another piece of equipment and what additional precautions that might warrant.  I am also considering making a sine wave output and maybe other frequencies.
> Tom
> 
> On 9/26/2013 4:34 AM, Bob Camp wrote:
>> Hi
>> 
>> Standard high speed CMOS logic works pretty well. How crazy are you trying to get?
>> 
>> Bob
>> 
>> On Sep 26, 2013, at 1:48 AM, Tom Minnis <Tom_minnis at att.net> wrote:
>> 
>>> I am working on a small clock distributor and wanted to get some ideas on what works best for 10MHz and 1PPS driver circuits.  I remember sifting through the archives a year or so ago and tripped on some discussion of this but I can't find it anymore.
>>> Tom
>>> _______________________________________________
>>> time-nuts mailing list -- time-nuts at febo.com
>>> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>>> and follow the instructions there.
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
>> 
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.