1Hz timebase

Check out this nice simple method of achieving a 1Hz timebase. This is basically a lesson in dividing crystal frequencies in circuits to get the desired result. In this case, they are starting with a 32.768KHz crystal and dividing it down. Instead of using an NE555 like many projects, he chose to go a direction that would yield results less prone to drifting with temperature variation. The method chosen was a CD4060 frequency divider, basically just a chain of flipflops. The divider is one step short of getting to the desired result so an additional flipflop has to be added. This is pretty basic stuff, but a great read. They go into detail as to how it all works and why you would use this method.

Pssst, hey, remember that time I told you to just use a 1Hz crystal? yeah, we can laugh at that again.

[via HackedGadgets]

16 thoughts on “1Hz timebase

  1. The theory and explanation are well written, the circuit presents some good working principles.
    But it is rather useless for a project with modern components.

  2. I so want to see a 1Hz Crystal. That would be so cool. Just need a mountain of quartz and Hover dam to power it. But that would be something to see.

  3. If you need 1Hz from 32,768kHz, why not hack the board out of a $2 “analog” clock and pick-off the pulses going to the stepper.

    Yeah, the $2 hacked clock may have a resonator instead of a crystal, but venture to guess what the Author has is really a resonator anyway. A “Real 1Hz” reference has to take into account a LOT of things, accuracy, resolution (jitter >< phase-noise) etc. etc.

    But anyway… nice to see HaD posting something like this instead of a deluge of Apple iPhone crap.

  4. Hi Guys, I wrote the 1Hz project.

    It was targeted at beginners just starting to move on from simple 555 circuits. It’s the timebase for a non-microcontroller freq counter. Interestingly, the freq counter tracks within 1 or 2 counts with my H/P bench counter from 1hz to 4MHz over a range of 50F to 100F. It’s actually more accurate than I expected, but I refuse to publish any great accuracy given the variability of clock crystals, and the lack of basic clock stabilization.


  5. It think the term that everyone has been striving for (instead of saying 1Hz crystal) is 1Hz oscillator. Very different things.. ;)


  6. If you use the CMOS 4521 you’ll only need one chip apart from the crystal (4,194,304Hz) and other small parts to acheive 7 frequencies ranging from 1/4Hz to 16Hz.

  7. While I always appreciate analog circuits, This almost seems like “overkill” when you could program a small PIC with an internal oscillator to do it for a total parts count of 1.

    Then again, some may argue that using a processor to make a 1hz clock is “overkill”. Those people don’t have boxes of free samples.

  8. @tim – I think it depends on the application, if your circuit already has a ucontroller, then you might use that. But if your circuit doesn’t, then this approach means you don’t have to batch program IC’s.

  9. @Tim: The internal oscillators in the PIC and AVR chips I’ve used have been RC affairs, so not very accurate and sensitive to temperature variations, though I suppose an external crystal wouldn’t increase the part count too significantly.

  10. If you really need an accurate, temperature-compensated 1pps timebase, the DS3231 can be configured to output a 1Hz clock and stays accurate to a minute per year under varying temperature conditions.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s