[time-nuts] IC for 1 PPS Output

Jeroen Bastemeijer J.Bastemeijer at EWI.TUDelft.nl
Fri Apr 7 04:22:46 EDT 2006


Dear Time-nuts,

 From a more theoretical point of view:
The source impedance is not such a big issue in a one way distribution 
system. The most important thing is the load resistance (or impedance). 
Reflections in a cable will be caused by the load-mismatch not by the 
source.
As power transfer is not the primary goal in a timing distribution 
system, a "voltage source" or low-ohmic source will perform just as 
well. (Due to the fact that optimum power-transfer will occur when the 
load and the source impedance ar the same).  Reasons why you would like 
to have a resistor in your source output:
* current limiting, in case of a short (allthough, if you use an OpAmp 
as buffer it usually has a current limiter inside)
* to damp out reflections caused by the mismatch of the load. A 
reflection caused by the load can be absorbed by the source if it is 
well matched to 50 Ohms. If not, the reflected wave, will bounced again 
at the source, and again at the load, and again, and again..... (OK, it 
will stop due to cable-damping  ;)  ).

Anyway, if your load is well matched to 50 Ohms, "any" resistor close to 
50 Ohms will perform well. You could even consider using a variable 
resistor in order to regulate the signal level.

Good luck with your project Brooke, best regards,

Jeroen

Robert Atkinson wrote:

>Hi Brooke,
>Most of the Video buffer/driver type IC's should work with a suitably
>scaled output resistor. Even a fast (high -slew rate) OP-Amp will work.
>The output impedance should be low enough to get a good 50 ohm match
>with just a couple of 100R in parallel. The out put impedance should be
>in the data sheet for the device.
>A good technique is to set the output of your buffer amplifier to a
>higher voltage than the desired output and use a resistive divider (50R
>output impedance) to attenuate the signal to the level you need, just
>don't forget about the other 50R load! This is sometimes called a "soft"
>output as the voltage will rise if it's not correctly terminated. 
>See the TAPR TADD1 < http://www.tapr.org/~n8ur/TADD-1_Manual.pdf >,
>Maxim Data sheet for the MAX477 <
>http://pdfserv.maxim-ic.com/en/ds/MAX477.pdf >  and application note <
>http://www.maxim-ic.com/appnotes.cfm/appnote_number/1036 > or National's
>LM6171 < http://cache.national.com/ds/LM/LM6171.pdf >for some ideas.
>With the MAX477 just using a 50R output resistor will give good results,
>not that it's not essential to have an exact resistance, many commercial
>designs use 52R (preferred value) output resistors with out problems,
>it's within 5%.
>
>HTH,
>
>Robert G8RPI.
>
>-----Original Message-----
>From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
>Behalf Of Brooke Clarke
>Sent: 07 April 2006 05:14
>To: Discussion of precise time and frequency measurement
>Subject: [time-nuts] IC for 1 PPS Output
>
>Hi:
>
>I've got the TVB 1 PPS PIC circuit working and would like to add an 
>output IC to drive 50 a couple of Ohm cables.
>Need two outputs, the normal 1 PPS and also the 10 kHz output since I 
>think it would work with the SRS app note for making 1,000 TI 
>measurements in one second to better see small offsets.
>
>I seem to remember that there may be an internal resistance already in 
>some chips so putting a 47 or 50 Ohm resistor in series may result in an
>
>impedance above 50 Ohms.  Any thoughts?
>
>Thanks,
>
>Brooke Clarke
>
>  
>

-- 
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: +31.15.27.86542
Fax: +31.15.27.85755
E-mail: J.Bastemeijer at EWI.TUDelft.nl





More information about the time-nuts mailing list