[time-nuts] Frequency Dividers

Bruce Griffiths bruce.griffiths at xtra.co.nz
Mon Aug 10 00:14:58 UTC 2009


Pete

You can probably do much better with the AD9513 at lower frequencies if
you use a couple of cascaded longtailed pairs each with carefully
selected gain (series feedback R between the longtailed pair emitters)
and bandwidth (capacitor between the collectors)  to condition the low
frequency input before driving the AD9513. Using a pair of back to back
shottky diode clamps between the collectors of the longtailed pairs will
also help. The optimum gain and bandwidth for each stage depends on the
input frequency and amplitude.

Bruce

Pete wrote:
> I thought I'd share some recent results for 2 useful frequency dividers.
>
> For frequencies above 50MHz, an A-D 9513; and a 74HC4059 for
> 40 MHz & below. The A-D 9513 likes higher slew rates & works
> best above 100MHz. I cheated and bought an evaluation board for
> the A-D 9513; layout is really important.
>
> First, I arranged a best case test to benchmark the system & have a basis
> for comparison with more flexible options. This setup yields high
> slewrate
> waveforms throughout. The configuration is: (note- the 74LV4046 PD1
> input has 50 ohms to gnd)
>
> DTS-2075 100MHz Ref out (400mVp-p squarewave) to A-D 9513 (set
> to divide by 10) 50 ohm input.
> A-D 9513 10MHz out (AC coupled, 1.6Vp-p) to 74LV4046 PD1 input.
> 74LV4046 PD1 out to 74HC4059 (set to divide by 1E4 ) Cp input.
> 74HC4059 out to 250 ohm(divide by 5) probe to DTS-2075 CH1, period
> mode. The CH1 signal is 850mV p-p. The DTS-2075 reads 1ms period;
> std dev = 3ps rms (+/-10ps p-p) The DTS jitter floor is 2.3ps rms.
> This equates to 1.7ps rms additive jitter from the 100MHz ref out to CH1.
>
> Knowing we often wish to examine sinewave sources, I repeated this test
> with a BLP-10.7 filter placed between the A-D 8513 and the 74LV4046.
> This provides a very low distortion 10MHz, 1.76V p-p sinewave into the
> 74LV4046 PD1 input. The DTS-2075 jitter changed to: std dev =3.7ps
> rms (+/-13.3ps p-p) with period unchanged. This equates to 2.9ps rms
> additive jitter.
>
> Finally, measuring the jitter degradation resulting from decreasing
> sinewave
> amplitude helps establish the useful input range for the 74LV4046 PD1.
>
> Input V rms   RMS Jitter Measured     Computed RMS Jitter(2.3ps floor)
> 612mV            3.7ps                                2.9ps
> 500mV            3.8ps                                3.0ps
> 316mV            4.7ps                                4.1ps
> 200mV            6.1ps                                5.6ps
> 100mV            9.4ps                                9.1ps
> 50mV            14.5ps                              14.3ps
> 25mV            19.5ps                              19.4ps
>
> It appears that inputs above 200mV rms are enough to "overdrive" the
> PD1 amplifier since the measured jitter response is so small. The
> 200-600mV input range is a "sweet-spot" for low jitter.
>
> 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.
>






More information about the Time-nuts_lists.febo.com mailing list