[time-nuts] Replies: Jitter Test on Dividers

Brian Kirby kilodelta4foxmike at gmail.com
Mon Sep 21 01:33:38 UTC 2009

1)    John Miles wrote:
> Interesting!  500 picoseconds is a lot of drift.  Can you try 74AC390s as
> well?
> -- john, KE5FX
Brian replies:    Unfortunately, I did not use IC sockets....

2)    Magnus Danielson wrote:

While that may be interesting, I think we should stop here and think a 
little... where would a seemingly stable oscillation/beat frequency with 
a period of about 1250 seconds (4 cycles in 5000 seconds) come from? 
Also, it is being triangular as noted by Brian. The only thing I can 
come up with from the top of my head is power or temperature related.

The relative quietness of the TADD-2 was nice to get confirmed. I have 
not yet had the time to measure that with mine.


Brian replies:    I am also recording the voltage every 10 seconds and 
it does not coordinate.  I live in Guntersville, Alabama and I am about 
3 miles from the Guntersville Dam, where our power is generated.  I 
monitor the power, as the voltage is quite high.  The lowest voltage has 
been 124 volts (typically around 5 PM) and it normally cycles ups to 131 
volts (around 4 AM).

3)   Bruce Griffiths wrote:

Diurnal temperature variations convolved with A/C cycling?

Alternatively modulation of the 5V Vcc supply (150mV pp) with a 22
minute period would also produce the same effect.

Brian replies:    I live in a townhouse apartment and the timing room is 
on the second floor.  Literally every air conditioner vent is shut off 
in the apartment, except for the timing room, which the A/C outlet is 
wide open.  I also run a ceiling fan 24 hours and a small 12 inch 
oscillating fan on the floor.  This keeps the timing room about 75 
degrees and the rest of the apartment upstairs is 72 degrees (the 5370B 
is quite a heater, along with a 2.4 ghz pentium 4 computer).  The air 
conditioner does not run much at night.  I need more instrumentation if 
I want to watch the A/C and refrigerator...!  And not only can I watch 
the refrigerator, if I am not careful, I'll eat everything in it !

I have about 4 more 12 hour test segments to run in series, when they 
are through, I will try to heavily insulate the 74HC390 divider box 
inside and outside and run the jitter test again on the Clark divider 
box.  The divider is in an aluminum box about 10 inches wide, 3 inches 
tall, and 8 inches deep.  It has a 5 volt 1 amp linear power supply 
inside, I believe by Power One.  The divider is wired on glass epoxy 
perf board and is stood off the chassis by about 1 inch.

Brian KD4FM

More information about the time-nuts mailing list