2
\$\begingroup\$

The 74hc590 http://www.ti.com/lit/ds/symlink/sn74hc590a.pdf is a binary counter followed by a register with a tri-state output.

I thought about using it an a (homebrew cpu) design, as it's wider than the 74hc161, but the output buffer means that it takes an extra clock before a counter clear CCLR# is reflected at the output.

Have I misunderstood this part?

What was the 590's original purpose? i.e. Why were people putting registers after counters, resulting in this part being made?

\$\endgroup\$
4
\$\begingroup\$

The use case is pretty well described by you: a binary counter with a latch.

For when you want something counted, but the output of the counter to only change on a latch signal.

Use cases range pretty widely – from gated clock dividers to clock-domain crossing counters to very simple IO expanders.

A very striking example is a frequency counter: assume you want to measure something in the range of 20 kHz. You have a 100 Hz clock. You use that 10 Hz clock to reset the counter clocked by the ca 20 kHz, so counts should be always in the range of around 200. Of course, you don't want your display to update while your counter is counting – only maybe 12.5 times a second. So, you latch your counter on the 8. edge of your 100 Hz clock.

(these ICs weren't all invented to build CPUs, so the assumption that "it exists" and "it must be useful in a CPU design" is unfounded)

\$\endgroup\$
  • \$\begingroup\$ Thanks for your answer. I didn't think that all 7400 ICs were made for CPUs, I was just curious as to what it was for, as my exposure to electronics is limited. \$\endgroup\$ – fadedbee Aug 12 '18 at 12:21
  • 2
    \$\begingroup\$ You're welcome! I wasn't sure you were insinuating that, but I wanted to avoid future readers would interpret your question that way :) \$\endgroup\$ – Marcus Müller Aug 12 '18 at 12:33
  • \$\begingroup\$ Marcus, your frequency counter example of this chip's target-market is supported by it having tri-state output (for MUX'd displays), a gate-able clock, and a separate latch input pin. Somewhat obsolete today - I'd use a microcontroller instead, without a second thought. \$\endgroup\$ – glen_geek Aug 12 '18 at 13:24
  • \$\begingroup\$ @glen_geek :) thanks! yeah, I can only second that, especially considering that many microcontrollers even have timer units that can double as input edge counters. But then again, these ICs are from a time where that wasn't the case :) \$\endgroup\$ – Marcus Müller Aug 12 '18 at 13:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.