Re: [time-nuts] Jitter Test on Dividers

2009-09-21 Thread Poul-Henning Kamp
In message 4ab6b76d.1010...@rubidium.dyndns.org, Magnus Danielson writes: Seems less likely. I was pondering if the power line frequency was doing some nice beat frequency with the system... 1250 seconds at 50Hz = 62500 periods. Assume one extra period over 1250 seconds: 62500 + 1 = 62501

Re: [time-nuts] Jitter Test on Dividers

2009-09-21 Thread Magnus Danielson
Poul-Henning Kamp wrote: In message 4ab6b76d.1010...@rubidium.dyndns.org, Magnus Danielson writes: Seems less likely. I was pondering if the power line frequency was doing some nice beat frequency with the system... 1250 seconds at 50Hz = 62500 periods. Assume one extra period over 1250

Re: [time-nuts] Jitter Test on Dividers

2009-09-21 Thread Brian Kirby
And a FYI - it was 60 hertz power here Magnus Danielson wrote: Poul-Henning Kamp wrote: In message 4ab6b76d.1010...@rubidium.dyndns.org, Magnus Danielson writes: Seems less likely. I was pondering if the power line frequency was doing some nice beat frequency with the system... 1250

Re: [time-nuts] Jitter Test on Dividers

2009-09-21 Thread Magnus Danielson
Brian Kirby wrote: And a FYI - it was 60 hertz power here Never let flimsy details like facts derail a perfectly good theory. (That's about 13,33 ppm BTW) Cheers, Magnus ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to

Re: [time-nuts] Jitter Test on Dividers

2009-09-20 Thread John Miles
-nuts] Jitter Test on Dividers I ran a 24 hour test on the async dividers (74HC390s) that Tom Clark designed and they basically have a triangular peak to peak jitter of 500 picoseconds over 22 minutes. The baseline drift started at reference 0 ns and made a negative parabola that dipped

Re: [time-nuts] Jitter Test on Dividers

2009-09-20 Thread Magnus Danielson
John Miles wrote: Interesting! 500 picoseconds is a lot of drift. Can you try 74AC390s as well? While that may be interesting, I think we should stop here and think a little... where would a seemingly stable oscillation/beat frequency with a period of about 1250 seconds (4 cycles in 5000

Re: [time-nuts] Jitter Test on Dividers

2009-09-20 Thread Bruce Griffiths
...@febo.com [mailto:time-nuts-boun...@febo.com]on Behalf Of Brian Kirby Sent: Sunday, September 20, 2009 2:19 PM To: precise time Subject: [time-nuts] Jitter Test on Dividers I ran a 24 hour test on the async dividers (74HC390s) that Tom Clark designed and they basically have a triangular peak

Re: [time-nuts] Jitter Test on Dividers

2009-09-20 Thread Bruce Griffiths
Magnus Danielson wrote: John Miles wrote: Interesting! 500 picoseconds is a lot of drift. Can you try 74AC390s as well? While that may be interesting, I think we should stop here and think a little... where would a seemingly stable oscillation/beat frequency with a period of about 1250

Re: [time-nuts] Jitter Test on Dividers

2009-09-20 Thread Magnus Danielson
Bruce Griffiths wrote: Not surprising, given that there is typically about 30ns clock to output delay per HC390 (divide by 2 and divide by 5 asynchronously cascaded) with 7 asynchronously cascaded 390's between the 10MHz clock input and a 1PPS output having a typical total clock to output delay

Re: [time-nuts] Jitter Test on Dividers

2009-09-20 Thread Magnus Danielson
Bruce Griffiths wrote: Magnus Danielson wrote: John Miles wrote: Interesting! 500 picoseconds is a lot of drift. Can you try 74AC390s as well? While that may be interesting, I think we should stop here and think a little... where would a seemingly stable oscillation/beat frequency with a