[time-nuts] Trimble Thunderbolt question, splitting its output

Charles Steinmetz csteinmetz at yandex.com
Tue Mar 31 03:46:05 EDT 2015


Dave wrote:

>I'm not sure how to avoid the change in input impedance with the 
>clipper involved unless I were to increase R1 to a very much higher value.

The voltage noise density of a 3.3k resistor is about 7.5nV/rootHz, 
for a total noise of about 23uVrms in a 10MHz bandwidth (compared to 
2.7nV/rootHz and 9uV for a 475 ohm resistor).  I doubt that changing 
R1 to 3.3k would raise the jitter measurably.

With R1 = 3.3k and R5/R6 = 33k (to reduce input attenuation with the 
increased R1), the input impedance for the first ~600mV (+/-) of 
input voltage would be ~50 ohms (100||100|33k), and for input 
voltages >600mV it would be ~49.2 ohms (100||100|3.3k||33k).  (Again, 
with a transition zone of ~100mV as the diodes turn on.)

Best regards,

Charles





More information about the time-nuts mailing list