2
\$\begingroup\$

This might be a newbie question, but I have always heard these terms interchangeably and never given it another thought. Recently I have seen them used differently in data sheets for a board I am designing and I am unsure what the distinction is between the two.

I assumed they both meant to just have it connected to a solder pad that goes nowhere. Is this incorrect?

\$\endgroup\$
9
\$\begingroup\$

I believe...

Leave Floating : Is meant for inputs. The pad is literally floating and is free to receive whatever noise is buzzing around your circuitry. Generally not a good idea for CMOS parts unless dictated by the part spec.

Leave Open : Is for outputs. The pad on "Leave Open" is being driven by the pin if the pin is active.

Which term you use for bidirectional pins is debateable. I'd assume because the input side is always active "Leave Floating" would be the prime suspect.

\$\endgroup\$
  • 1
    \$\begingroup\$ Yes my opinion too, upvoted \$\endgroup\$ – TonyM Apr 12 '17 at 19:22
  • \$\begingroup\$ I would think that open would be for inputs, as this is the keyword used in VHDL to indicate not connecting a vector. However this does make a lot of sense. +1 \$\endgroup\$ – Tropical_Peach Apr 12 '17 at 19:23
  • 1
    \$\begingroup\$ @Tropical_Peach in VHDL 'open' is for OUTPUTS \$\endgroup\$ – Claudio Avi Chami Apr 12 '17 at 19:37
5
\$\begingroup\$

I would consider "Leave Open" and "Leave Floating" to both mean to not make any connection to the pin. The pin would normally be soldered to a pad on the PC board, but that pad would have no connection to other components.

\$\endgroup\$
5
\$\begingroup\$

A pin left open is not connected anywhere. Some pins, such as tristate output pins on logic chips, or MCU pins can be made to float by making them high impedance (programatically or via a control pin). They are still connected but have minimal effect on whatever they are connected to.

Note that an open pin is just not connected so it's always high impedance. A high impedance general purpose pin on an MCU might be high impedance only for applied voltages within the (current) power supply range plus a bit. If the power supply is removed, the high impedance may only apply for applied voltages of a few hundred mV.

The other distinction you may run into with data sheets is between NC (not connected) and DNC (do not connect). The latter may be connected internally and connecting to the pin may have adverse effects.

\$\endgroup\$
  • 3
    \$\begingroup\$ NC pins can also be internally connected (at least in the chips my company makes!). We don't call it not connected, but no connect. \$\endgroup\$ – Justin Apr 12 '17 at 19:50
  • 2
    \$\begingroup\$ As Justin says, read the data sheets carefully to discover what the manufacturer means. In a significant fraction (maybe 5%) of datasheets you may find that it is not spelled out unambiguously and you have to make further inquiries. \$\endgroup\$ – Spehro Pefhany Apr 12 '17 at 21:06

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.