[time-nuts] Micro Phase Stepper: Thank you!

buehl at superlink.net buehl at superlink.net
Tue Jul 19 15:16:53 EDT 2005

Quoting Ulrich Bangert <df6jb at ulrich-bangert.de>:

> Hi Folks,
> thank you very much for your kind explanation and hints regarding micro

> I imagine very well how a dac driven constant current source sourcing or
> sinking a small capacitor in conjunction with a comparator may give a
> variable slope linear ramp allowing very fine delay steps. 

The DAC sets the reference for the comparitor, while the 'ramp' remains 
constant.  Thus you can trigger at any point on the ramp, the resolution of 
which is the accuracy limit of the comparitor and the DAC.  The max to min 
time range is determined by the slope of the ramp.  Time from starting the 
ramp (application of current source) to the detection point (and related 
comparitor & propagation delays).  

Using two comparitors/ two DACs allows a START and STOP arangement to define a 
time interval; Either two pulses, or with logic a variable width pulse 
(limited by the rise/ fall time of the logic).

Regards,  Tom Buehl

However, even
> with a arrangement like that, i would expect the device to have a
> minimal delay of say the propagation delay of the comparator plus one or
> two gates in a current semiconductor technology, so '0' to x.xxx ns is
> perhaps a bit too optimistic??? Or have those high tech guys some
> 'faster than light' technology available for compensation of propagation
> delays???  
> If ready to go integrated circuitry is available like the MAXIM parts,
> this is pretty much like a invitation to use them but i will remember
> Tom Buehl's suggestion and check their precision first before i use them
> in a real project. I have not checked Tom Clarke's links due to my slow
> private ISDN line (I live in a region of Germany where we have more cows
> than humans per square kilometer. We will be lucky to get DSL at the end
> of the year). Will check that the next day at my working place where we
> have a DSL-3000 line.
> Thank you all for your help, this group is definitely the 'high
> signal-to-noise ratio place' that John is asking for!
> Ulrich Bangert, DF6JB
> _______________________________________________
> time-nuts mailing list
> time-nuts at febo.com
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts

More information about the time-nuts mailing list