I want to reply to Farhan and others about the PLL issue although
this maybe a little off-topic.
Two years ago I were attempting to make a "PLL in a PIC" as Farhan
state starting from the Huff-Puff design appearing in Elektor
magazine: ei9gq/stab.html .
This design used an external 8 bits counter, internal 8 bit
preescaler and the internal 8 bits main counter, so that the reading
of frequency was in 24 bits. However the process of reading was
tedious upon having to insert bits in order to read the two counters
to which access was not had, the reading could slow down to (256+256)
* 2 cycles of 400ns = 0.5 mS.
Once the reading was carried out it is necesary to make a comparison
of 24 bits and to generate a pulse with a size(width) proportional
to that difference. All this process is very slow, it can be used
only for small differences from original frequency. So, the dynamic
range of control is very small in my opinion.
I think that it is impossible to carry out this type of control with
a PIC fast or slow. In my opinion, if it is not possible to find the
145170 one could use any standard PLL like those of National
Semiconductor of the LMX series, or the MB1502 of Fujitsu, the
Phillips SAA1057 or any another equivalent. In last case a DDS could
be used. The main advantage of the MC145170 is their operation up to
180MHz.
Greetings
Manuel
EB7HPM/ EC7ALV
