[time-nuts] 74HC4046 receiver

Bruce Griffiths bruce.griffiths at xtra.co.nz
Thu Apr 9 01:47:57 UTC 2009

Pete wrote:
> Bruce,
> Quite correct, fumble fingers win again. As they say "here's
> the rest of the story".
> I was able to measure just 1 part. It is a NSC MM74HC4046N,
> date code 9018, presumably still available from Fairchild. The VCO
> was disabled (INH=H), the COMP input was grounded & I looked
> at the PC1 output. Transition times were <3ns & abberations were
> <5%. The resulting jitter measures:
> Input Vp-p        Jitter ps rms
> 2.36                  8.6
> 2                       8.8
> 1.5                    9.2
> 1                       9.5
> 0.7                    10.2
> 0.5                    11.1
> 0.3                    14.5
> I tried comparing this with a 74AC04, but my setup must be too
> noisy as the best the 'AC04 could do was 12ps rms. The input to
> the 'AC04 was AC coupled & yielded 49/51% duty cycle, apparently
> switching cleanly with the 2.36Vp-p input. Ground bounch & ringing 
> were <10%.
> Pete Rawson
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.

What were you using as the timing reference?
Were you driving the start input with a reference and the STOP input
with the 4046 (or AC04) output?
Or were you using another setup?

The Fairchild version of the 74HC4046 has lower sensitivity inputs than
the corresponding Philips/NXP or TI parts.
You would need to look at the corresponding National datasheet to see if
their 74HC4046 its similar to either the Fairchild part or the others.

With 2.36V pp sinewave input at 10MHz the input slew rate is ~ 75V/us at
the threshold crossing of the AC04.
It would only take about 1mV rms of noise at the AC04 input to produce
an output jitter of 12ps.
If the AC04 noisebandwidth is 100MHz this corresponds to ~ 100nV/rtHz
which is possible.

The way to test this is measure the AC04 output jitter as a function of
the input signal slew rate and try to deduce the equivalent input noise
from the results.
Driving 2 AC04's from the same input signal may also be useful as to
first order the source jitter will cancel if the time difference between
the output transitions of the 2 AC04's is measured.

The 1ps jitter that I quoted for an AC04 is the limiting value achieved
when the AC04 input is driven by a high slew rate input signal not a 
~2V pp 10MHz sinewave.


More information about the time-nuts mailing list