4
\$\begingroup\$

I'm looking at this schematic someone did for daisy chaining a set of 74HC595's and I wanted to know why there is a capacitor put on the pin4/latch and the GND. What purpose does it serve and why is it not required on the other 74HC595?

enter image description here

\$\endgroup\$
  • 5
    \$\begingroup\$ Looks like someone's rude hack to delay the latch signal a little, probably relativel to the clock signal. Note that it says "Arduino", so this could have been done by someone that could barely spell "EE". \$\endgroup\$ – Olin Lathrop Jul 24 '14 at 22:12
  • 3
    \$\begingroup\$ it IS connected to both chips! and also pin 4 on the arduino. its probably to debounce the signal from the arduino. is it connected to a pushbutton or reed switch or some other physical device? \$\endgroup\$ – Octopus Jul 24 '14 at 22:45
  • 9
    \$\begingroup\$ @Olin: It's spelled "relative", not "relativel" ;) \$\endgroup\$ – EM Fields Jul 24 '14 at 23:57
  • 2
    \$\begingroup\$ Maybe someone just goofed with the schematic, and meant to show it across 5V and ground, i.e., your common garden variety 0.1uF decoupling cap, just misplaced on the diagram. \$\endgroup\$ – JustJeff Jul 25 '14 at 4:51
  • 1
    \$\begingroup\$ Debouncing the (data) clock input might serve some purpose. Debouncing the latch (clock) input is nonsense: multiple pulses on this input will simple re-load the same data from the shift register to the holding register (latch). My best bet is that it was meant to be connected to the power. \$\endgroup\$ – Wouter van Ooijen Jul 25 '14 at 6:53
3
\$\begingroup\$

Likely an error

The overall quality of Arduino connected hardware designs varies widely as the community consists of a comparatively large number of hobbyists and students.

The 0.1uF cap was most likely intended to decouple the adjacent +5V power supply line. Attaching it to the latch signal line serves no unique beneficial purpose and can cause problems if your firmware operates the clock line at its maximum frequency and you have a long serial chain of registers.

Delay

If you need to delay the latch pulse to comply with setup and hold times (timing minimums) of the 74HC595 you are using, then you should address this in other ways: either correct your layout and/or add the delay in your firmware.

Debounce

On its necessity...

jippie said it best:

The type of bouncing where this configuration works, is not a digital controller output like the your diagram implies. It is more intended for a push button.

On its consequences...

Even if we believe that a bouncing source was driving this line, there is no practical consequence of such behavior.

Wouter van Ooijen, said it best:

Debouncing the (data) clock input might serve some purpose. Debouncing the latch (clock) input is nonsense: multiple pulses on this input will simply re-load the same data from the shift register to the holding register (latch). -- Wouter van Ooijen

\$\endgroup\$
  • \$\begingroup\$ Makes perfect sense. \$\endgroup\$ – DustinDavis Jul 31 '14 at 19:35

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.