[time-nuts] My GPS-Derived Standard...

Bert ve2zaz at yahoo.ca
Sun Oct 29 01:30:57 UTC 2006


Greetings to all,

Yes, it is me, Bert, VE2ZAZ. I guess you know who I am by now....

An amateur radio acquaintance forwarded one thread of this forum to me today. I appreciate the constructive comments from those who understand the objectives met by my GPS-derived standard design. I almost fell off my chair, though, when I read what some of you said about me and the work I did. When you put most of your spare time for more than a year into something, it shocks you to you read that. I honestly don’t care about who said what. What matters to me from here on is what I can get out of constructive discussion. Emotional stuff over!

My main objective was to create a circuit that would provide a better long term accuracy than a free running test instrument. The objective I set was to get better than 1x10e-9 accuracy, in a simple, easily reproducible and low cost way, with as few components as possible and with a Windows GUI. For all but the most demanding hobbyists (and these were the targeted users), this is more than enough for the kind of use they make of it. If you read the article AND my website, you will see that I inform the user of the reasons why they should not expect better than 1x10e-9. In fact, almost none of the things you mentioned in your exchange were unknown to me (I invite you to read the biographical note at the end of the article). But the point is : KISS for most hobbyists. It seems that this kind of performance is satisfactory for the 60+ people who have ordered the PCB and PIC so far.

If some of you have doubts regarding the system’s ability to achieve the above-mentioned accuracy, let me just assure you that what I say is true. Over the several months that I have measured accuracy on my system, it has consistently stayed in the 10e-10 decade. I describe my measurement technique in one of the article footnotes. If you are convinced that my design does not work, then there is nothing I can do or say to convince you.

Could I have done things in a much better way? ABSOLUTELY
Better supplies, separate supplies, external DAC, better buffering, better board layout, use of SMT for better supply decoupling, phase measurement, etc, etc…The list is almost endless when trying to get the last bit of accuracy. Again, KISS.

Will there be an improved version in the future? Maybe, time will tell…

Are there some easy and small changes I could recommend to the users as improvements to the design? Your input is welcome! I will document these to accompany the design on my website. I already have a few comments that I intend to post there.

I have no doubt that I will learn some stuff by reading this forum. For sure, it would have been great if I had joined a couple of years ago...alas.

:-)

Bert, VE2ZAZ










More information about the Time-nuts_lists.febo.com mailing list