Ever since I "discovered" the CD4046 PLL circuit (to accurately control the speed of a large DC motor) back in 1981, I've been amazed at the versatility of this building block.
I have personally used it in many projects, and seen it used in dozens more.
But I have never seen, or myself designed an application, in which the zener diode function is employed.
To me it is the least useful pin. I always see it as not connected.
Some 4046 variants have actually eliminated the zener and substituted with a type-III phase comparator, lock indicator or some other useful thing.
Thus the question: Have you ever used or seen that function being employed?
I am only curious. I am not planning to use that functionality ever.
I've used the 4046 in lots of projects, but always ignored the "Zener" part. Completely useless.
... well, you could build a simplistic stabilizer with it. But I agree, even a 3 terminal regulator is better.
This IC is pretty old. RCA App. Note
1972 ICAN-6101 pg. 715The RCA datasheet is really good. TI does a hack job on what they publish in comparison. RCA CMOS IC Databook 1983, CD4046B datasheet is on pg. 184
I thought three-terminal voltage regulators got popular after 1972, like the LM340, and they took over in circuits.
Otherwise, it's just a freebie (zener) for regulation when using the VCO on higher V power supplies.
I have that book. Pure gold!
And as you mentioned, that app note is a must-read for anyone wishing to understand the 4046 functionality.
That app not is no good. It has the same mistake as the TI and ON notes:
you can NOT use the phase comp II with a simple RC filter, you need an active filter. Gardner (the PLL guru) states this unequivocally, with the maths to prove it.
Interesting.
I know Mr Gardner to be the PLL guru. As a matter of fact, the APP note mentions him.
Where did you read PC II's active filter requirements? I would like to be educated.
About 10 years ago I bought Gardner's "Phaselock Techniques" (3rd edition). Expensive (>100 $), but worth every cent if you want to work with PLLs.
You can either follow his maths in the earlier chapters, or go to chapter 11, "Loop Filters".
I quote:
"Only a type 1 PLL can be realized with a passive filter,..."
He then goes on to explain that a more desirable PLL is type 2 (0 phase error), which needs a second-order loop response, hence the active filter.
He also writes about early misconceptions on how to analyze the PD output stage, which still plague the 4046 data sheets and application notes (they've never been revised).
Hope this helps. Cheers.
Additionally:
Motorola's application note:
AN-535 "Phase-Locked Loop Design Fundamentals"
Is much, much better and has the maths right. Old, but highly recommended. It does not refer to the 4046, but to the MC4024 and MC4044, which are somewhat older TTL parts.