Bruce Griffiths skrev: > Mike > > A statement of accuracy is of little value unless you also give: > > 1) An estimate of the accuracy of standard used for comparison. > > 2) An estimate of the random and systematic errors in the comparison > > 3) Some details of the comparison method. > > 4) Averaging time and other pertinent info.
True, but for an hobby enthusiast we could be a little less stringent (as long as we remember we are cheating) if the goal is to say: Be able to measure and maintain my 10 GHz signals to within 1 Hz on long term. Achieving the frequency stability below 1E-10 for tau = 100s and above should not be too hard to achieve and some fairly easy means of monitoring the state of a Thunderbolt should be able to indicate wither this is met or not. It's not that we can't achieve better performance, but if we do not need to be near the cutting edge but rather have fairly relaxed requirements, let's not complicate maters more than necessary. The one thing we do want to say is that things needs to be monitored, since there may be many sources which could make the functionality be severely degraded or lost. Many forget the performance supervision aspect. GPS receivers in particular seems to be (fairly) cheap magical boxes which makes things work, and then they are forgotten all about until something breaks. The type of actions needed to handle the case when the service is degraded depends. For a ham it may very well be that he just does not really know where he is on the spectrum right now... but the years of previous measurements could at least given him some confidence on how much he could expect to be off now that he is not able verify or steer his rig. (Sorry for the stereotypic identification of a ham as being "he", I know for a fact there is female hams out there and also sharing the same concerns as any male ham on where she is on the spectrum, but it was just convenient to type it like that.) Cheers, Magnus _______________________________________________ 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.
