[time-nuts] Internal 5 volt switching regulator on some "non-programmable" FE-5680A's

David davidwhess at gmail.com
Sun Jan 8 23:01:55 UTC 2012


On Sun, 8 Jan 2012 14:23:02 -0800, Orin Eman <orin.eman at gmail.com>
wrote:

>On Sun, Jan 8, 2012 at 1:51 PM, Clint Turner <turner at ussc.com> wrote:
>
>> Hello,
>>
>> After posting a few days ago about one of my '5680A's having the voltage
>> converter installed - but not connected - I've done a bit of
>> reverse-engineering and sleuthing around and (probably) have a fairly
>> complete picture of what it would take to populate that section of the
>> board.  That information may be found here:
>>
>> http://ka7oei.com/10_MHz_**Rubidium_FE-5680A.html<http://ka7oei.com/10_MHz_Rubidium_FE-5680A.html>
>>
>> This page is a work in progress, using my LPRO-101 page as a template and
>> unashamedly stealing a good chunk of its content - most of which is
>> directly applicable, anyway!
>>
>
>Are you sure about those resistor values?   They look like 5.11K and 5.62K
>(standard 1% values) to me.

I agree. Besides not requiring rare and expensive 4 significant digit
precision resistors which just happen to match the marking of the
standard 1% resistors, the nonstandard 12.95K and 11.15K on the
schematic would yield 5.23 volts.  5.62K and 5.11K will yield 5.08
volts from the nominal 2.42 volt reference on the LT1376.  The 5 volt
resistor selection in the data sheet would have been closer but maybe
they wanted the extra 7 millivolts or were compensating for the
feedback bias pin current.

If noise was so important I wonder why they did not use a lower noise
switching regulator or the LT1375 which supports oscillator
synchronization.



More information about the time-nuts mailing list