[time-nuts] Updated Divider Jitter Results - 74HC390

SAIDJACK at aol.com SAIDJACK at aol.com
Sun Apr 5 03:37:59 UTC 2009

Hello Pete, Bruce,
I can confirm that the Wavecrest is sensitive to edge rise-time. It was not  
designed to measure sine waves.
With a 10MHz sine wave from a Fury GPSDO as the source I get 8 -  10ps rms 
That exact same signal run through an NC7SZ04 buffer prior to feeding it  
into the DTS-2075, then scaled to the +/-1.1Vpeak input of the DTS-2075 yields  
jitter below the units' noise floor (~3.2ps rms) on a good OCXO.
I also have a DTS-2070C that performs slightly better - about 2.9ps  rms.
In a message dated 4/4/2009 17:43:23 Pacific Daylight Time,  
peterawson at earthlink.net writes:


Your analysis conforms closely to measured results on my  DTS-2075.

With the cleanest 10MHz source I have, at 2Vp-p, the DTS-2075  jitter
reading is 11.4ps rms. Running this number back into equivalent  input
noise yields 716uV rms. The DTS-2075 input spec (assumed to be
for  -3dB response) is 700MHz, which should be nearly 900MHz
equivalent noise  bandwidth. These numbers are all reasonable &
serve to demonstrate the  need for low noise signal conditioning.

Pete  Rawson

More information about the time-nuts mailing list