[time-nuts] Phase Comparator

Jeroen Bastemeijer J.Bastemeijer at EWI.TUDelft.nl
Thu Dec 8 05:15:40 EST 2005

Dear Peter,

I checked the article. It looks nice and simple, however on a second 
look it shows a number of design flaws.... The claimed temperature 
stability is more a sort of assumption: In the text (not the abstract) 
it is stated as:
"The measured temp.sensitivity is about 1ps/K for small temperature 
changes around the 23 degrees C average temperature."
The claim of the temp.sensitivity is more or less based on an 
assumption. It is not defined how "big" a "small temperature change is". 
>From this vague assumption it is extrapolated that the sensitivity is 1 
ps/K.... doesn't look like a scientific claim to me. This value is not 
measured or verified by measurements, at least it is not presented in 
the paper.
Furthermore the system uses standard non-paired 2N3904 transistors for 
the comparator, and even worse, they are used at different bias-currents 
which means different temperature coeefficients. So the switching point 
WILL change with temperature... When using these transistors, they 
should be used at fairly high collector-currents 40-50 mA for a fast and 
"fairly" stable temperature response. The output stage is not optimized 
for that, the collector-current, hence switching speed depend on the 
circuits output load. Other design-flaws:
* The supply voltages for certain parts are stabilized bij 5.1 Volt 
* The schmitt-trigger used is not very stable for temperature....
With other words, a nice design, when placed in a climate chamber. But 
the claim for the temperature stability is way too high.

Jeroen PE1RGE

pohl at orcon.net.nz wrote:

>You might look at: 
>>G. Brida 
>>High resolution frequency stability measurement system 
>>Review of Scientific Instruments, Vol. 73, no. 5, pp. 2171-2174, 
>>This article reports on the frequency stability measurement system 
>>based upon the dual-mixer time-difference scheme working at 10 MHz 
>>with 1 Hx beat note. The principle of the system is reviewed, the 
>>design approach is described, and the experimental results are 
>>presented. The measurement system shows a stability of 5.6·10-14/t 
>>(Allan deviation) and a temperature sensitivity of some picoseconds 
>>per Kelvin. An approach to reduce crosstalk problems has been 
>Sorry,I don't have an electronic copy at the moment.It specifically 
>addresses the gain/bandwidth zerocrossing detector issues.
>Peter ZL2AYX
>time-nuts mailing list
>time-nuts at febo.com

Ing. Jeroen Bastemeijer

Delft University of Technology
Department of Electrical Engineering
Electronic Instrumentation Laboratory
Mekelweg 4, Room 13.090
2628 CD Delft
The Netherlands

Phone: +
Fax: +
E-mail: J.Bastemeijer at EWI.TUDelft.nl

More information about the time-nuts mailing list