4 thoughts on “m600_ocxo_gps_tdev”

  1. Hi,

    I am wondering, how can you tell the measurement resolution from looking at the plot? Is this because the graph "levels" out at longer taus?

    1. Yes, to me it looks like the internal OCXO is more or less free-running up to an averaging-time of 100s or more. After that it is disciplined to the external reference (either an externally supplied 1PPS or 10MHz, not sure which one was used here). Since the TDEV levels out at ~few nanoseconds it means that is how well the M600 is able to lock onto an external signal.

    1. In frequency counting mode you can estimate the instrument floor by feeding an external 10MHz to the REF input at the back and the same 10 MHz (through a splitter or distribution amp) to CH1 at the front. Or use 10MHz for REF, and then a good signal generator locked to the same 10MHz and the signal-generator output at some suitable frequency into CH1. I have some graphs here:
      http://www.anderswallin.net/2015/04/keysight-53230a-noise-floor-test/

      By default the 53230A does internal averaging which means it's hard to interpret the results and get reasonable ADEV plots. The RCON-mode is included by default in the counter but it is undocumented. I think it can only be used when controlling the counter remotely (GPIB, USB, or Ethernet). I have some graphs over here:
      http://www.anderswallin.net/2015/06/cont-vs-rcon-mode-on-the-53230a-frequency-counter/

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.