Pinging 74HC4046 Users

I've not even played with one, except to measure some DC. The data sheets would seem to indicate that the non-linearity occurs at the tuning extremes. Just bound your control voltage if you're getting lock-in issues

Major deja-vu here :-) I used NXP's MS-DOS tool to cook up some values but they where way off. Fortunately the circuit I tried the HC7046 in is a one-off and not something that needs to go into production. I have used PLLs before but never had so much trouble getting the circuit going.

Or even the metal gate CD4046.

Cheers

Phil Hobbs

You are supposed to pick up the turd by its clean end.

tm

Cheers

Phil Hobbs

We build something like the charge-pump detector into FPGAs. We use an external dual schottky diode for the pump-up and pump-down blips, to avoid the deadband that tri-state charge pumps tend to create. We can also delta-sigma those outputs to control our VXCO open-loop.

The little function generator chips make nice wide-range, low frequency VCOs. Exar, Maxim?

is

nce

re

he

ve

id

ave

And there used to be a fairly wide range of chips available that were designed to be very linear VCOs, intended for use as voltage-to- frequency A/D converters. Analog Devices still seems to be in the business

Phil Hobbs probably should be using the AD650 - though I can't recommend it on the basis of personal expereience

Fifteen bucks for a 1-MHz V-F converter? Not me, especially not inside a PLL where 1% linearity is way better than good enough.

I'd happily use a metal-gate 4046 at frequencies where they work--all you need is a resistor to ground from the PD2 output to pull it off the dead zone.

Above a megahertz or so, a current-programmed triangle wave oscillator is good, or else a linearized LC VCO. You can get linearities of better than 10% in varactor-tuned VCOs by putting in a couple of off-stage resonances.

Cheers

Phil Hobbs

Those aren't too bad looking, but check out the NXP HC7046 datasheet, p.

Even the much-ballyhooed HC9046 has the same sorts of worries, see P.24 of

And check out the actual HC4046--first in a Chinese knockoff,

Compare that with the HEF4046 (NXP metal gate version)--check out the VCO linearity error plot on P. 15 of

The HC versions are all over the map.

Cheers

Phil Hobbs

Turns out that the principal difference between old CD4046 and new

Apparently their mis-directed aim was higher operating frequency. The result is extraordinarily bad linearity :-( ...Jim Thompson

That's very interesting. So they didn't just use bigger FETs for the output devices of the mirrors? How did they do it?

Cheers

Phil "Bipolar" Hobbs

Most likely bigger FET's... but I'm guessing they used raw device-size scaling, no cascoding (which minimizes channel-length modulation).

I don't have any device library information :-(

The sad part of all this is how easy it would be to design a much better VCM... not a difficult task at all.

Phil, What is your application? If you need linearity, why not just roll your own, given the high-speed comparators that are now available off-the-shelf. ...Jim Thompson

I'm not using any PLLs at the moment--you asked about the modelling problem, and the badness of the HC4046 oscillator is one of my pet peeves from many years back. Sort of like you and Democrats. ;)

I've often done as you suggest--in fact over the years my main use of HC4046es has been as acquisition aids for my real phase detectors, which are generally diode bridges because their noise is so low. Mini Circuits MPD-1s are good medicine.

Just now I'm fighting bit rot in an electromagnetic simulator that worked fine up till yesterday, and debugging a little board full of pHEMTs and 0402s.

Cheers

Phil Hobbs

Try this...

Note the frequency, ~46MHz, and the current consumption ;-)

It is immaculately linear in frequency versus control current.

Now I'm yanking your chain just a wee bit, this is on a 0.18um process, but I'll try it on a 5V process and see how it behaves.

But would anyone buy it? ...Jim Thompson

Is it bit rot or root rot ?:-)

Don't you love "worked fine up till yesterday" situations?

One of my worst pranks was on a young engineer (EE, at OmniComp/GenRad)... he's fretting over a breadboard that "was working yesterday". He goes to lunch and I sit down at the bench with my OptiVisors...

and spot a solder bridge. I solder suck it up, and the breadboard works.

When young engineer returns from lunch he's bewildered and pacing around, then testing, then pacing, then testing...

To this day I've never told him... and it's been 27 years ;-) ...Jim Thompson

As long as it was less than a buck or thereabouts, probably so. Of course the much-maligned Younger Generation might not understand. ;)

Cheers

Phil Hobbs

Not really. This one's a genuine mystery at the moment--same executable, same input files, last week it ran fine, today it produces hundreds of megabytes of NaN's. Like I said, it's obviously bit rot.

I'd be very happy if my case of bit rot spontaneously fixed itself like that. I did find that the 0402-ish board had a major goof--somebody stuffed NPNs where PNPs should have been.

I wonder if there is a useful class of circuits where you could replace PNPs with NPNs without reversing the power supplies, and have it still work.

He's probably forgotten all about it.

Cheers

Phil Hobbs

The point was that these are legacy parts, and correspondingly expensive. The LM331 which offers even better linearity, but only goes to 100kHz, turns out to be still available to, but at $6 each in small quantities. I was rather hoping to provoke a response from somebody who is still using that kind of part.

Sure. A 1GHz gbw op amp could take you quite a way above 1MHz.

Messy. I'd be thinking of a digitally controlled Direct Digital Synthesis chip, which would be a lot tidier and would probably have a lower jitter (if you low-pass filtered the synthesised sine wave properly). For a seriously low jitter option, a DDS-like system including an MC100E195 might be interesting - if complicated. Coping with the temperature dependence of the delay through the MC100E195 might require Peltier junction or a self-calibrating scheme if you really wanted to exploit the full capacity of the MC100E195.

I would have bought it up to about seven years ago. That was the last time I used a '46 (actually a '9046 with its better phase detector). I left the oscillator disabled.

These days, for the sorts of things I design, I'm more likely to replace the entire application with something like this:

Regards, Allan