[time-nuts] 10MHz to 32MHz?
Don Collie
donmer at woosh.co.nz
Mon Apr 23 03:56:10 EDT 2007
I have learned the hard way that the 4046 needs input signals whose
risetimes [and falltimes?] are not too long , and It`s probably best to use
the inhibit function, and avoid using the source follower. The output
resistance can then be absorbed into the first resistor of the loop filter,
outside the package. I`ve played around with the 4046B, and have built a
synchroniser to overcome thermal drift in a signal generator that I own.
Funny thing is, even a 15.625Hz [ie 15Hz approx] frequency at the phase
detector is quite sufficient to practically eliminate thermal frequency
drift, and with a long loop time constant, the output signal, when mixed
with another steady frequency, sounds very clean. Lockup time is a couple of
seconds or so. Lots of fun. The thing that puzzles me is: why is the plot of
VCO voltage versus time different when locking from below to locking from
the same initial frequency difference when locking from above. It`s a pity
you can`t predict PLL lockup characteristics using linear algebra - or do
you need a particular kind of phase [or frequency] detector that would
enable you to do this?
FWIW,
Cheers!..............................................................................Don
C.
To: "Discussion of precise time and frequency measurement"
<time-nuts at febo.com>
Sent: Monday, April 23, 2007 3:24 PM
Subject: Re: [time-nuts] 10MHz to 32MHz?
>
>
>> Hi John,
>>
>> talking about 4046 PLL's: we are looking for a better Spice model
>> for the
>> 4046 PLL, but only found an HC4046 model in Pspice. That part is
>> too slow.
>> Would you have a pointer to a better chip such as the 74HC4046 etc?
>>
>> thanks,
>> Said
>
> Not sure; I've never used any SPICE other than LTSpice, and never tried to
> use a 4046 in it. I'll bet they have a model for it somewhere, though. I
> would think you could just play with the propagation delay values in the
> HC4046 model to make it work with any other logic family.
>
> I've built a ton of PLLs with the PLLatinum and Analog Devices chips, but
> only one with a 4046.
>
> -- john, KE5FX
>
>
> _______________________________________________
> time-nuts mailing list
> time-nuts at febo.com
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>
>
> --
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.5.446 / Virus Database: 269.5.9/773 - Release Date: 4/22/2007
> 8:18 p.m.
>
>
More information about the time-nuts
mailing list