Hi Probably a question for John Miles (who designed them and is a list member) or Symmetricom (who now sells them for $10K to $15K).
Quick and dirty description - takes the two inputs and digitizes them against an internal clock. The result is fed to a PC via USB. The data is auto-correlation processed on the PC and after a bit of work by the CPU, you get phase noise information. Bob > On Nov 20, 2014, at 5:34 PM, Alex Pummer <[email protected]> wrote: > > > > Hi Said, > do you have any information about how that "TimePod 5330A works" any > principal description? > 73 > KJ6UHN > Alex > > On 11/20/2014 2:08 PM, S. Jackson via time-nuts wrote: >> Hello Mike, >> attached is a 10MHz DIP-14 TCXO Phase Noise plot from a random LTE-Lite >> unit. >> I had sent out a 20MHz typical phase noise plot some weeks ago, and >> comparing the two they are almost perfectly 6dB apart as would be expected >> from >> the 20log(n/m) relationship. There are variations from unit to unit of >> course, but it does not seem like one version of the board or the other has >> advantages in terms of phase noise. >> I had also sent out a superimposed plot of the 20MHz and the divide-by-2 >> 10MHz output of the same board at that time, and again the relationship was >> almost perfectly 6dB lower at 10MHz versus 20MHz. >> While phase noise follows theory, it does seem that the DIP14 metal shield >> has a beneficial effect on the ADEV stability though. The plots we are >> getting are pretty darn good, and I want to test more boards before I post >> any >> ADEV, because its quite a bit better than our specification and I want to >> make sure its real. >> The 10MHz DIP-14 boards do not have an isolating buffer like the 20MHz >> boards do, on these the TCXO drives the output directly, so one must be >> careful >> to set the equipment to 1M Ohms input impedance or use a buffer >> externally, which is what we did to measure the attached PN plot. >> Bye, >> Said >> > > _______________________________________________ > 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. _______________________________________________ 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.
