Oscillator based on a crystal problems. / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Oscillator based on a crystal problems.

1 2 next

Move Thread

LAN_403

Arcstarter

Wed Feb 10 2010, 04:49AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

I made a quick oscillator with a 4MHz crystal resonator with a 10megohm resistor in parallel, across one of the inverters from a 74hc14. The output was fed to another inverter for buffering (which was pretty much unnecessary).

Well, i took the old circuit apart and made another one, exactly the same in every way. It is made on top of some copper clad, with little squares of copper clad glued to the top for all electrical connections, other than ground. The copper clad everything is on is, of course, the ground plane.

Now while using a 4MHz crystal, the output is 11MHz. I tried a 3.5MHz crystal, and it was pretty much the same, except in the first quarter of a second or so after starting up, it was about the right frequency. Maybe still too high, i cannot tell.

Does this sound like a 74hc14 problem? Before i had absolutely no problems and my uncalibrated DMM read exactly 4MHz.

Thank ya

Myke

Wed Feb 10 2010, 05:51AM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

I've also had problems with getting the right freq from crystals using the same oscillator as you. I built it on a breadboard so the capacitance between the rails might have an effect on how well it can oscillate.
What topologies have worked for everyone?

Dr. Slack

Wed Feb 10 2010, 08:44AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

There are many things that could be giving problems. It sounds like the crystal is going off at a "harmonic". I put that word in quotes, as they are mechanical modes which a generally, but not exactly, N times the fundamental, and frequently cause oscillation problems. A crystal that is designed for harmonic operation will have these higher modes well specified.

That said, almost everything you are doing could be unintentionally exciting the higher modes, in a fashion which varies from build to build.

a) A crystal will resonate with the Cs present at input and output. Generally we use real Cs which swamp the parasitic Cs present on the input and output of a random gate. If the only C it sees is those parasitics, then it is very likely to go off somewhere higher. You will often not notice the difference in frequency between it oscillating with its intended C loading and oscillating with a much reduced loading of only parasitics, because that's the beauty of a crystal, its frequency is very insensitive to changes in loading.

However, the strays seen on the pins of a buffer will vary 100% from one buffer to another, and strongly with supply voltage. It is entirely possible that one buffer will tune the fundamental nicely, and another the harmonic. Use of real Cs on both terminals of the crystal to ground will reduce the influence of variable buffer parasitics.

b) A 74C14 is totally the wrong gate to use for sustaining a crystal. A 74C04 would be better, best would be a 74U04 (the unbuffered version). They do work, or appear to, but they are doing too much and upsetting the crystal, which needs to be tickled rather than thrashed if it's to behave.

c) Driving the gate output straight into the crystal is yet another problem, the high drive will happily excite higher modes. Put a resistor (100R to 1k) in series with the gate output to tame things (it may stop a high loss crystal, experiment).

Zeus

Wed Feb 10 2010, 09:41AM

Registered Member #2316 Joined: Tue Aug 25 2009, 03:04AM
Location: Bendigo, Australia
Posts: 107

Harmonics jumped to my mind at first, but could it be a overtone crystal.

Avi

Wed Feb 10 2010, 01:18PM

Registered Member #580 Joined: Mon Mar 12 2007, 03:17PM
Location: Melbourne, Australia
Posts: 410

Maybe you could try adding the 2 something picofarad capacitors on each side of the crystal to ground, but I don't know how to calculate their value.

101111

Wed Feb 10 2010, 02:45PM

Registered Member #575 Joined: Sun Mar 11 2007, 04:00AM
Location: Norway
Posts: 263

Avi wrote ...

Maybe you could try adding the 2 something picofarad capacitors on each side of the crystal to ground, but I don't know how to calculate their value.

These values are sometimes listed in the crystal datasheet.

kilovolt

Wed Feb 10 2010, 03:03PM

Registered Member #2018 Joined: Tue Mar 10 2009, 09:56AM
Location: Switzerland
Posts: 74

I had a similar problem with the oscillator for my 4MHz HF-SSTC. I made an oscillator with a 7400. When I replaced the 7400 by a 74HC00 it didn't work anymore...

Greets kilovolt

Proud Mary

Wed Feb 10 2010, 04:06PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

It's not very sound to use a quartz crystal without some degree of LC selectivity in the feedback path, and/or a low-pass or bandpass filter at the output to attenuate harmonics and spurious emissions.

A trimmer cap together with a fixed L, or a fixed silver mica cap with a slug-tuned L is good. Temp compensating cap arrangements are good to maintain frequency stability and thermal drift. Anchor all the frequency determining elements with epoxy to reduce microphony.

GeordieBoy

Thu Feb 11 2010, 11:30AM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

For an HF-SSTC I'd recommend buying a ready-made crystal oscillator module like this:

They are fully screened, and designed by experts to be stable for a wide range of supply voltages and output loading conditions. Using one of these is much easier than studying datasheets and then trying to design and test your own crystal oscillator.

-Richie,

Steve Conner

Thu Feb 11 2010, 12:40PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I don't like to sound hobbyist-unfriendly, but this is true. If you try to make your own crystal oscillator, sometimes it just decides to oscillate on some random frequency totally unrelated to what it says on the crystal!

You need to delve deep into circuit theory to explain why, but for some "cookbook" guidelines on making crystal oscillators work well, look in an old ARRL handbook or similar ham radio book. Just watch out for the really old vacuum tube oscillators that pump enough current to shatter a modern miniature crystal!

The little capacitors are usually 22pF or so, and their purpose is to provide loading so the crystal can oscillate in a parallel resonant circuit. The exact value will be in the crystal datasheet, and when ordering custom crystals for parallel resonant operation, you have to tell the factory what load capacitance you want them cut for.

But the exact value doesn't really matter, all it does is affect the frequency slightly. It certainly sounds like the circuit didn't have enough loading capacitance: try a 22pF from the input of the inverter to ground, to provide a path for the crystal current. Dr. Slack recommended one on the output, too.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob