Hi
On Sep 9, 2013, at 9:14 AM, "quartz55" <quart...@hughes.net> wrote: > Wait a minute, Bob. I have an LPRO with good reported bulb voltatge. Are > you telling me this TBolt is no better than the Rb standard, as far as > stability and perhaps worse? It may not or may not be any more stable than your LPRO. You can not determine from the LH ADEV if it is or isn't. > And the frequency accuracy is also no better and both have to be compared to > a Cs or H Maser to be calibrated? Frequency accuracy and frequency stability are two different things. A source that swings +/- 2 ppb a day, but *average* to zero is accurate, but not stable. > I thought the GPS put out these precise second pulses that the TBolt would > measure over time and discipline the OCXO to those precision seconds over > time resulting in a variable (very small) but precise statistical frequency The "jitter" in the GPS pulses is the issue. Your GPS may be flopping around by many ns each second. That's many ppb before you do any averaging. > (unlike the Rb which is just very stable) ….. and thus is a good way to help answer "how stable is my GPSDO" ... > and thus having a reference back to NIST to a degree. (in my simplistic > language) > > Yes, I'm not fully understanding the Tau and ADEV I'm sure. I did read it's > the sq root of Allan variance. So the reported 10000 Tau ADEV is not a > measurement of the variance of the frequency compared to the precise 1 second > marks over time? Assuming LH is reporting 10,000 second ADEV as "10000 tau" then it's the ADEV correctly calculated for the GPS looking at the GPS disciplined OCXO. Since one is following the other it's like saying your counter is exactly right because it reads 10.000000000 MHz when you plug the reference output on the back of the counter into the input jack on the front of it. In order to do it correctly you need to compare to *independent* sources. > I thought if I had 1.0e-12 that was comparable to knowing it's within 1Hz at > a billion (10^12) Hz? Since it's ADEV you have a standard deviation of frequency change between 10,000 second samples of 1 ppt. Standard deviation isn't the same as mean. > > Should I reverse the &a command I did and let it run in the mode I got it in? > The 10000 ADEV is back down to 1.6e-12 this morning. You can see the > screenshot here, > http://s251.photobucket.com/user/DogTi/media/time/20139-91229_zps7e475453.jpg.html > it looks like the temp is still jumping but for some reason the number of > sats has increased, a lot, maybe it's just the harmonic convergence. Again, you are checking your yardstick against it's self. Check it 10,000 times and average the results. The average error will be very small. That does not really say much about how accurate it is or if it changes length when the humidity goes up. > > Is that white line that runs around the center essentially the frequency > variance? And why do I have what appears to be multiple blue and red lines? White corresponds to the words in white just above the graph. It's running frequency. More or less, it's how much the frequency is jumping around. I have no idea why the dual ADEV lines. There is also a trace set to dark grey on your plot that shows the pps moving around. Bob > > So many questions, so little time. Anyhow, as soon as it warms up a bit and > I have my Earl Grey, I'm going to turn this thing off and go reset the > antenna and see what happens, I need to make a bracket which won't take long. > I just worry about climbing up the ladder these days since I had a herniated > disc removed and a spacer installed this spring. At least the pain is gone > and I'm not in a wheelchair. > > Dave > N3DT > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
