1. Home
  2. Electronics Design
  3. Simple digital oscillators

Simple digital oscillators

When I think about a cheap, quick and dirty digital oscillator, I think of a CMOS Schmitt trigger inverter with a feedback resistor and a shunt capacitor at the input: (use courier or similar for ASCII art schematic)

R +--/\\/\\/\\----+ | |\\ | +----| >o----+----out = c |/ | GND

But the frequency depends somewhat on the threshold and critically on the hysteresis voltage, both of which seem to be spec'd pretty loosely.

AoE mentions this circuit:

R1 C +-----/\\/\\--------+-----||-----+ | | | | \\ | | /R2 | | \\ | | / | | |\\ | |\\ | +----| >o-------- +----| >o----+----out |/ |/

(fig 5.30)

AoE sings the praise of it for its low phase noise. But what kind of frequency spread would you expect to see? If you were trying for, say 25 Hz, would you get 25+/- 5Hz?

The other choice would be a CMOS 555.

Are there any other good, cheap, simple oscillators with digital output that can provide a frequency tolerance of, say, +/- 20% at around 25 Hz? I don't care about duty cycle or phase noise.

I don't want to draw a whole bunch of mA to do this. Let's say one mA max, but less is better, and much less is much better. The only supplies available are VCC and GND.

--Mac

Reply to
Mac
Loading thread data ...

That latter c "In practice, the operating frequency ... is subject to a change of less than 5% over the production spread of transfer voltages, and to a frequency shift of less than 2% with a 40% change in supply voltage. Another advantage conferred ... is that of excellent thermal stability. The operating frequency typically varies by only 1% over the temperature range -40°C to +85°C."

--
Terry Pinnell
Hobbyist, West Sussex, UK
Reply to
Terry Pinnell

See...

CMOS-Osc-NoClip.pdf and CMOS-Oscillators.pdf

on the S.E.D/Schematics page of my website.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.
Reply to
Jim Thompson

Thanks, Terry. I may use 4000 series logic. AoE specifically mentions 74HC logic, but I'm sure it will work with any CMOS.

Anyway, it sounds as though the performance is good enough for my purposes (assuming that R. M. Marston is a reliable source). ;-)

--Mac

Reply to
Mac

I faced similar problem and went with a crystal oscillator to minimize current draw at the frequencies I wanted. The circuit in the "1 bit logic analyzer" at

formatting link

I hope this helps, Bob Smith

Reply to
Bob Smith

Well, I think a crystal is unnecessary. I'm shooting for around 25 Hz with a +/- 20% or so tolerance. A 555 would do just fine, but I think something simpler should work. I'll probably go with the dual inverter circuit from AoE (or maybe I should say from Marsten). ;-)

Thanks.

--Mac

Reply to
Mac

Mac, If you have the resources, perhaps a small (sot23-5? or so-8 package) micros would fit the bill.

Would a 4060+xtal be over kill? The current would be low.

rob

Reply to
Rob

Thanks. CMOS-Oscillators.pdf was particularly helpful.

--Mac

-
Reply to
Mac

Hello Mac,

For that the single stage would probably be enough. Else use the dual stage or Jim's circuit if you need a really clean output.

Yech. Anyway, sounds like overkill here.

Well, if one sixth of a Hex Schmitt, a resistor and a cap ain't simple enough ... It comes to a grand total of 3-4 cents in mass production as long as you can find a job for at least some of the other five Schmitts. Else you'd be looking at around 10c and then you could think about a single transistor stage to get back into the nickel range.

And here I thought I was the penny pincher when designing electronics.

At 25Hz and a resistor in the 1M+ range you can squeak by with just a few microamps. It all depends on what kind of load you hang onto it.

Regards, Joerg

formatting link

Reply to
Joerg

Agreed.

Oh, you are. Reading through your posts is quite an education in cost-sensitive design. ;-)

I was initially just concerned that the single stage might have too large a frequency spread. After reading about the two-stage circuit (same as the one in AoE) in Terry Pinnell's post and in one of Jim Thomson's linked PDF's, I am convinced that it will do the job. And I am now encouraged to hear you say that the single Schmitt may even suffice. Maybe I will just give it a try.

Oh, right, I didn't mention that. It will be driving four CMOS inputs, I think.

Thanks, Joerg.

--Mac

Reply to
Mac

Hello Mac,

The tolerance can pretty much be calculated from the hysteresis tolerances and parts tolerances, mostly of the cap. A 5Hz range at 25Hz looks like a barn door though ;-)

At that frequency those inputs will draw next to nothing.

Regards, Joerg

formatting link

Reply to
Joerg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.