----- Original Message ----- From: "Tom Van Baak" <[email protected]> To: "Discussion of precise time and frequency measurement" <[email protected]> Sent: Thursday, December 11, 2008 7:10 AM Subject: Re: [time-nuts] GPS signal simulator
>> Tom >> >> Thanks that should be very useful for software simulations. >> >> The use I had in mind for the tool I described where for the next stage of >> testing. >> It can be used to test at the REAL Hardware level, using a 'perfect' GPS >> signal with >> no noise or with any type of desired controlled noise. >> For example it is hard and time consuming to see how many ps of error >> your phase detector is adding or what the phase margin or noise gain is of >> the H/W PLL. >> Is the power supply good enough or is it causing errors, etc >> When using real GPS data, the random noise will hide the advanced >> type of information you may want to take. >> The other use of the Hardware simulator is for someone that wants to know >> how >> accurate their GPSDO is but does not have access to a source accurate enough >> to test their GPSDO. >> >> WarrenS > > Read a paper like this to appreciate the complexities involved: > > http://tycho.usno.navy.mil/ptti/ptti2002/paper29.pdf > > /tvb > ****************************** TOM Thanks once again, The paper is very helpful, even if I can't yet say I understand it all. Agree, noise is a complex issue, GPS noise more so than most, because of all the many different types of contributing noise sources. That is the reason I like to remove those unknowns initially for my H/W test. Without adding the GPS noise to confuse the facts, it is very clear and easy to see, measure, and fix if desired the difference between say a sub ns accurate design and a micro sec unit. One of the questions I've needed to answers is just how good is good enough when it comes to making a high performance GPSDO. Some of the self build units I've heard being used can be 100 to 1000 or more times worse than desired for best performance. WarrenS _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
