Hi Sounds like a lot of fun getting everything to zero out :)
One thing to watch for is hysteresis. The tempco may not be quite the same going hot to cold as going cold to hot. Bob > On Aug 9, 2017, at 2:34 PM, [email protected] wrote: > > HP5065A +20V supply and C-field circuit mods > > I'm doing some more work on the 5065A and thought I'd share the results > of a > modification (one of several that will be done on this unit). > I measured the +20V P/S temperature coefficient and got -13.5ppm per > degree C. > This equates well with the 1N938 10ppm/C and the voltage set resistor TCs > > of plus/minus 10ppm/C. However the TC varies if the C-field pot is > changed. > The original circuit has 11.9ma thru the reference diode series resistor > (925 Ohm)for a diode current of 6.8ma and a C-field pot circuit current > of > 5.1ma at midrange on the pot. (zero TC on the diode is specified as > 7.5ma) > The diode current ranges from 5.3 to 7.8ma depending on the C-field pots > position. > Since I am replacing the C-field circuit, the C-field pot circuit has > been > removed. The 925 Ohm resistor was replaced with a 1ppm/C 1K resistor and > a > 1K precision pot. This allows the Reference diode current to be adjusted > to compensate for all the elements in the circuit. (The TC of the diode > changes depending on the diode current) > After adjusting the diode current and temperature over many iterations > and monitoring the +20 Volt output I was able to find a diode current to > bring the +20 Volt output coefficient to below 1ppm per degree C! > (-.6ppm/C) > So that part of the mod is done. > Next I will be installing a new C-field current source that has a TC of > 2ppm > per degree C. > Then I'll do some long term plots against the Maser to see if the long > term stability has improved at all. > > More to come! > Cheers, > > Corby > > _______________________________________________ > 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.
