Crazy thought.Could you just force a DC offset into the EFC assuming the internal varicap is not out of range. It would be simply adding a resistor to pull up or down to see if you can get a bit of pull and allow the dac to move back in range a bit. Its a band aid. I know its sort of crazy. But ripping the oven apart really is no fun at all. Regards Paul WB8TSL ************************************** Not really such a crazy thought, I've done this on a Trimble-Nortel NTGS50AA where the original 34310-T oscillator had "aged" beyond the 3 to 6 Volt EFC range. On a unit that was refusing to lock the oscillator was removed and confirmed to require an EFC voltage of approx 6.5 Volts for an output frequency of 10 MHz. As other tested samples of this oscillator, although admittedly not many, required around 4.5 volts there does remain the possibility of an actual internal fault but it did test on the bench as otherwise ok. Using a simple unity gain level shifter based on a few resistors and a TL071 op amp, just because there was one to hand, the EFC voltage from the control circuit was shifted 2 Volts high and the unit then behaved as expected. The effects of temperature, supply variation, etc on the modified circuit were not investigated as this was only a short term test but it certainly looked to be a viable proposition, to the extent that rather than fit a replaccement 34310-T that board was fitted with SMB connectors using the pads provided to keep as a test bed. Nigel GM8PZR

Hi Assuming you can break the EFC lead to the OCXO, you aren’t going to damage anything inside by feeding the EFC line with +/- 5V. If you are going to tear open the OCXO, the line will need to be pulled anyway. Without knowing what they drive the EFC line with, it’s hard to know if the driver on the line would be happy with a back fed voltage. Best to be careful …. Bob > On Mar 6, 2018, at 3:30 AM, gandalfg8--- via time-nuts <time-nuts@febo.com> wrote: > > > > > > Crazy thought.Could you just force a DC offset into the EFC assuming the > internal varicap is not out of range. > It would be simply adding a resistor to pull up or down to see if you can > get a bit of pull and allow the dac to move back in range a bit. Its a band > aid. > I know its sort of crazy. But ripping the oven apart really is no fun at > all. > Regards > Paul > WB8TSL > > > > ************************************** > > Not really such a crazy thought, I've done this on a Trimble-Nortel NTGS50AA where the original 34310-T oscillator had "aged" beyond the 3 to 6 Volt EFC range. > > On a unit that was refusing to lock the oscillator was removed and confirmed to require an EFC voltage of approx 6.5 Volts for an output frequency of 10 MHz. As other tested samples of this oscillator, although admittedly not many, required around 4.5 volts there does remain the possibility of an actual internal fault but it did test on the bench as otherwise ok. > > Using a simple unity gain level shifter based on a few resistors and a TL071 op amp, just because there was one to hand, the EFC voltage from the control circuit was shifted 2 Volts high and the unit then behaved as expected. > > The effects of temperature, supply variation, etc on the modified circuit were not investigated as this was only a short term test but it certainly looked to be a viable proposition, to the extent that rather than fit a replaccement 34310-T that board was fitted with SMB connectors using the pads provided to keep as a test bed. > > Nigel GM8PZR > > > _______________________________________________ > 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.

Someone mentioned the right non-lazy approach with a summing/offset opamp. Do agree. But then just for a try use a V supply and a 10K. I seem to recall tracing the DAC to an opamp. The EFC line is pretty easily accessed. Its one of the little coax cables. The other even lower risk approach is disconnect the coax and see what voltage is required to 0 out the 10 MHz. How far away is it from the DAC feeding the coax connector. Regards Paul On Tue, Mar 6, 2018 at 9:40 AM, Bob kb8tq <kb8tq@n1k.org> wrote: > Hi > > Assuming you can break the EFC lead to the OCXO, you aren’t going to > damage anything > inside by feeding the EFC line with +/- 5V. If you are going to tear open > the OCXO, the line > will need to be pulled anyway. Without knowing what they drive the EFC > line with, it’s hard to > know if the driver on the line would be happy with a back fed voltage. > Best to be careful …. > > Bob > > > On Mar 6, 2018, at 3:30 AM, gandalfg8--- via time-nuts < > time-nuts@febo.com> wrote: > > > > > > > > > > > > Crazy thought.Could you just force a DC offset into the EFC assuming the > > internal varicap is not out of range. > > It would be simply adding a resistor to pull up or down to see if you can > > get a bit of pull and allow the dac to move back in range a bit. Its a > band > > aid. > > I know its sort of crazy. But ripping the oven apart really is no fun at > > all. > > Regards > > Paul > > WB8TSL > > > > > > > > ************************************** > > > > Not really such a crazy thought, I've done this on a Trimble-Nortel > NTGS50AA where the original 34310-T oscillator had "aged" beyond the 3 to 6 > Volt EFC range. > > > > On a unit that was refusing to lock the oscillator was removed and > confirmed to require an EFC voltage of approx 6.5 Volts for an output > frequency of 10 MHz. As other tested samples of this oscillator, although > admittedly not many, required around 4.5 volts there does remain the > possibility of an actual internal fault but it did test on the bench as > otherwise ok. > > > > Using a simple unity gain level shifter based on a few resistors and a > TL071 op amp, just because there was one to hand, the EFC voltage from the > control circuit was shifted 2 Volts high and the unit then behaved as > expected. > > > > The effects of temperature, supply variation, etc on the modified > circuit were not investigated as this was only a short term test but it > certainly looked to be a viable proposition, to the extent that rather than > fit a replaccement 34310-T that board was fitted with SMB connectors using > the pads provided to keep as a test bed. > > > > Nigel GM8PZR > > > > > > _______________________________________________ > > 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. >