[time-nuts] OCXO and fluctuations after EFC adjustment

[Bob kb8tq]

Hi

One use for zener diodes is as RF noise generators ….. the
coil probably was there to quiet things down. 

Bob

> On Apr 11, 2020, at 4:42 PM, ed breya <eb at telight.com> wrote:
> 
> Rick said:
> "In the case of the 10811, I have already posted about the reference diode of special characteristics. I don't remember all the exact details of how it was chosen, but it was based on proprietary knowledge."
> 
> This reminds me of something unusual I found in a Vectron 10 MHz OCXO many years ago. It was busted, and it was a soldered-shut type. I had nothing to lose in taking it apart, hoping to find the problem and restore it if possible. It came apart OK, and I found in the densely packed circuitry, a small RF choke that was burned up. I traced the circuitry enough to figure out that it was in series with the internal reference regulator diode - a 1N82X 6.2V TC type. This added some small DC voltage drop to the Zener voltage, and that's what went to the internal circuitry, and also out to an external pin. This pin was intended to supply an external pot for tweaking the EFC line in stand-alone applications, and apparently sometime in its past, it had been faulted to an excessive voltage, causing the choke to burn out, and maybe damage to other parts.
> 
> The only things I could figure were that the choke maybe served as a small copper resistance (+3000 ppm/K or so) to provide some extra degree of temperature compensation for some reason, or that it was to suppress possible oscillations from the Zener's negative R in avalanche.
> 
> The choke was one of those little molded types, about the size of a 1/4 W resistor, and unfortunately burned up and apart too much to read any color bands. I guessed it to be around maybe up to a few uH, and a few ohms, but that's as far as I got. I set it aside, thinking someday I'll dig into it some more, but it still sits in pieces in a box somewhere, remaining a mystery to me. All I know is, there must have been a reason for it.
> 
> Ed
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.