General Help on Understanding Digital Inputs

Discussion in 'General Electronics Chat' started by Griz, Feb 12, 2014.

  1. Griz

    Thread Starter New Member

    Jun 19, 2013
    5
    0
    Hi, I'm fairly new to the boards, just got into the design and tinkering phase with electronics, but I've been a mediocre technician with board repair, replacement and trouble shooting experience before, with some success.

    I have been reading on and experimenting with some digital logic and digital level signals for a couple of days and need to clear up some confusion with what I've been seeing in some places.

    Brass tacks is that I've always been taught, read and reminded that a DC current, even a pulsed DC current, never reverses polarity. Okay, that's the best understanding of DC I have now. Yet in some digital applications I see waveforms that would have components both above and below the zero reference point for digital signals referenced to circuit (chassis) ground. What am I not getting here? Are these actually equivalent to AC signals or not and why, please?

    Thanks in advance.
     
  2. #12

    Expert

    Nov 30, 2010
    16,298
    6,810
    It is rather difficult to imagine the special case you describe. Rather than stretch my mind for every kind of non-standard possibility and write a book about them, how about you name one or two?
     
    Griz likes this.
  3. tshuck

    Well-Known Member

    Oct 18, 2012
    3,531
    675
    Digital doesn't necessarily mean DC, though it does denote logic levels that are, more often than not, held at some value (be it above or below chassis ground voltage) to transmit data to another device.

    In fact, most digital signals have a huge AC component to them.
     
    Griz likes this.
  4. Griz

    Thread Starter New Member

    Jun 19, 2013
    5
    0
    Thanks, you don't need to write a book here. Your very answer says that it can happen and that's what I suppose I'm actually looking for here is confirmation that my definition of AC isn't off and I'm seeing what I think I am on the scope. Example without specifics, a few circuits involving micro-controllers I'm seeing have +5 Volt / -5 volt signal levels and don't necessarily call it AC for some input pins. These are few and far between, and maybe I'm just looking at the wrong docs or the right ones the wrong way. Just trying to be sure that it CAN be that way before I go smoking out chips based on a few sources on the web.
     
  5. Griz

    Thread Starter New Member

    Jun 19, 2013
    5
    0
    Tshucks, yes, but I'm more concerned with binary data transmission, instruction codes, etc. in what I was reading. I realize that levels are important to discriminating 1s from zeros and triggering gates. But, well, maybe I'm fiddling with it in ways I'm not supposed to was running through my head while I was attempting to overreach my education in these matters. But then, that's the cost of education. I've heard of pulling up a line and pulling down a line, but a few, let me emphasize that, few, of the circuits I've seen randomly on the web and been chewing on seem to almost invariably have text that reads "1 represented by -X volts for y mSeconds pulse" then turn around and designate a pull up to zero or 'slightly positive' for binary 0. I may be just having a brain error myself. I am NEW to wire level design for micro-controllers and just questioning what it is I think I knew;-) It's a habit with me I guess.

    Thanks for that statement "In fact, most digital signals have a huge AC component to them." I think that and a long coffee break and adjust my thinking hat will fix me. Thanks again to all of you.
     
  6. crutschow

    Expert

    Mar 14, 2008
    13,016
    3,235
    Typical CMOS/TTL digital digital signals don't go below zero except for possible small undershoot on fast fall-time signals due to stray circuit inductance/capacitance.

    Be careful when you use a scope to be sure it is set for DC input. On AC input the signal may appear to go negative, but that's due to the AC coupling capacitor which blocks DC.
     
    #12 and Griz like this.
  7. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,388
    1,605
    Got a link to any of these circuits? Then we would know exactly what your question is.

    Keep in mind the terms AC and DC are almost abstract concepts when looking at signals. Yes, the power out of a wall socket is AC and any battery is DC, but a logic signal is not a "true" DC because DC never changes and a signal that never changes doesn't contain any information.
     
  8. MrChips

    Moderator

    Oct 2, 2009
    12,446
    3,361
    As others have pointed out, it is important to note that the terms DC and AC can have different interpretations depending on the context.

    In the strictest sense, DC refers to current that does not reverse direction of flow.
    AC would refer to current that can exhibit a reverse in the direction of flow.

    In electronics, and particularly analysis of circuits and signals, there is a more loose interpretation. A signal may contain both DC and AC components. A typical example would be the ripple on the power supply voltage. The DC component would refer to the average supply voltage. The ripple may be considered the AC component. Hence AC is also used to classify any component of the signal that has a frequency greater than 0Hz.

    As pointed out, it is important that when using an oscilloscope you fully understand the function of the AC-GND-DC option on the input channel. If the switch is set to AC you will be removing the DC component and will only observe the AC component. The switch must be set to DC in order to observe both DC and AC components.

    In digital circuitry, signals can be both +ve and -ve voltage. RS-232 interface signals is a common example.
     
  9. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    serial data modes like rs232 use both positive and negative levels positive for ones, negative for zeros. this helps get signals through noise.
    rs422 uses two lines of similar to rs232, but always oposite to get extremely good noise immunity.
    the odds of a noise pulse going both positive and negative at the same time is very slim. thats called diferential signaling.
     
  10. ramancini8

    Member

    Jul 18, 2012
    442
    118
    Check the old RCA CMOS data book. It contains applications with capacitors that have signal levels going below ground.
     
  11. tshuck

    Well-Known Member

    Oct 18, 2012
    3,531
    675
    Off the top of my head, ARINC-429 uses +/-5V, differential signaling to transmit data. This would be digital, as discrete values represent logical states, but since it is a bipolar return-to-zero signaling scheme, it is constantly changing states (creating an AC component).

    As others suggested, a signal without an AC component doesn't contain information, so a digital signal cannot (in the strictest sense) be considered pure DC unless it's value never changes, and it never conveys new information.
     
  12. #12

    Expert

    Nov 30, 2010
    16,298
    6,810
    I want to thank you guys for explaining several concepts that I simply didn't have time to do yesterday.
     
Loading...