Subscribe to Make Magazine Today!
youritronicsPCB.jpg

When I first got involved in digital electronics, it took me awhile to understand the concepts of pull-up and pull-down resistors and when to use up or down, and when to use internal (built in, programmable) resistors, and when to use actual, discrete component, resistors.

On Youritronics (a name that will always make me wince), new editor Laci runs through a simple demo of pull-up and pull-down resistors, using the ATmega8 protostack.

In digital electronics the engineers use the terms pull-up or pull-down resistor, this is a simple resistor usually between 10k and 100k, to define the input state when no signal source is connected, this way overriding the noise signal, common sense dictates that when you have potentially larger noise then a smaller resistor is needed, but don’t be careless about it, don’t place a 100Ohm resistor because your signal source must be able to “defeat” the pull-up(down) resistor. A rule of thumb is to use at least 10x larger pull-up(down) resistor then your signal source impedance.

The pull-up term is used when the resistor connects the pin to the + and pull-down when connects to the ground, but then which should you choose? Good question, in these days it doesn’t really mater, your application or the pcb design should dictate, the older generation digital IC’s had asymmetrical current sink/source capability, they could sink more current and because of that the pull-up resistor was more common.

ATmega8 protostack simple demo

Gareth Branwyn

Gareth Branwyn

Gareth Branwyn is a freelancer writer and the former Editorial Director of Maker Media. He is the author or editor of over a dozen books on technology, DIY, and geek culture. He is currently a contributor for Boing Boing and WINK Books. And he has a new best-of writing collection and “lazy man’s memoir,” called Borg Like Me.


  • anachrocomputer

    I sometimes use an analogy with a spring-loaded pushbutton. The pull-up resistor acts like the spring that makes the pushbutton spring back to its rest position, and without that the button would stay pressed, or end up in some intermediate position. With the spring (resistor) installed, the button (voltage) will pop back out (rise to 5V) when the button is released.

    • http://gravatar.com/achundee achundee

      good analogy

  • jammit

    I consider the resistor + switch combo as a voltage divider. The switch alternates between infinity and zero ohms.

  • Mike Goldman

    that first sentence in the original post is a doozy

    • Minh

      agreed

  • bungle

    This post would make sense if it was written in english.

    • http://twitter.com/AlecRippberger Alec Rippberger (@AlecRippberger)

      *were written in English :-)

      • bungle

        *was.

        • Dorcus

          were (subjunctive case)

  • uihjoijoioj

    i dont understand your copypasta

  • Andrew Wright

    I agree with uih. I thought further analysis would be performed to give a bit more insight. Why do we need pull up or down resistors? Are we protecting the pins from over current or are we simply setting a reference voltage?

  • fraggie

    None of the above :) as you can read, they are used to keep noise away from the digital lines. If you use a motion sensor for instance ,you would not want noise on the pin to trigger the floodlight..

    And yes, you use voltage..

    • Branden Sherrell

      What? No. This is not at all true. The pull-up resistor, as stated above, forces the state of the line to be in a definitive state nominally. If you have noise on the input line coming from your signal source then it will propagate through; the resistor do no such filtering.

  • http://gbleezy.com Graham Savai’i Hart

    What the heck is input impedance and how would we figure that out exactly for any particular pull-____ resistor implementation? Is it similar or related to audio impedance?