My observations are that the RS232 offset on these needs to be considered as very much a fine adjustment only, and presumably just intended to pull the frequency on to 10 MHz rather than to provide a usable offset. I did try pushing my first unit to the extremes of the adjustment range shown in the manual and it didn't like it, the frequency was highly unstable and at one extreme it just gave up and rejected the offset. Observations on three units so far indicate that none seem to have had any offset programmed in the field but all are less stable the further they are from being set to 10MHz. I haven't recorded any long term data as yet, just observed the short term stability on an HP 53132A and Tek FCA3100 clocked against one of three T'bolts, but am getting a definite impression that 10MHz is a sweet spot and that anything either side is just begging to be corrected:-) Regards Nigel GM8PZR In a message dated 12/01/2012 01:04:03 GMT Standard Time, smit...@c-c-i.com writes:
More observations. My unit is S/N 0316 61953, UN 63983. If I try to change the offset count (using RS232) by more than about 131,000 counts (0x20000) at a time it sometimes loses the 1PPS and the lock signal goes high. By bringing the offset count up slowly (increments of 0x10000) I recorded the following drift rates versus offset count: 0 usec / sec at 0 1 usec / sec at 1441792 (0x160000) 2 usec / sec at 2949120 (0x2D0000) 3 usec / sec at 4390912 (0x430000) 4 usec / sec at 5898240 (0x5A0000) The offset count was incremented in units of 0x10000 (65,536) counts so the drifts are approximate. A plot of these results is here: http://c-c-i.com/node/161 My unit appears to have an adjustment resolution of about 6.8E-13 / count - close to what has been reported here before. _______________________________________________ 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.
