Understanding Analog-to-Digital Conversion for Digital Computing and Processing

  • Thread starter Thread starter Elecomputer
  • Start date Start date
  • Tags Tags
    Analog Digital
Click For Summary

Discussion Overview

The discussion revolves around the necessity and extent of knowledge in analog circuitry for computer engineers, particularly in the context of analog-to-digital (A/D) and digital-to-analog (D/A) conversion. Participants explore the relationship between digital and analog systems, the importance of understanding real-world signals, and the implications for practical applications in digital computing and processing.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested
  • Homework-related

Main Points Raised

  • Some participants express a preference for digital systems but acknowledge the necessity of understanding analog concepts for practical applications, particularly in interfacing with sensors and converters.
  • Concerns are raised about the complexities of real-world signals, including issues related to common mode voltage and grounding, which can affect the performance of A/D converters.
  • One participant mentions the importance of basic DC circuit analysis and AC circuit analysis, suggesting that these are foundational for working with analog signals.
  • Another participant argues that detailed knowledge of analog is not strictly necessary, as datasheets for ADCs and DACs often provide sufficient information for interfacing.
  • There is a suggestion that understanding the analog nature of digital signals is crucial, especially when dealing with high-speed processing or real-time digital signal processing (DSP).
  • Some participants question the value of additional analog courses, weighing the cost against the perceived necessity for their specific interests in digital systems.
  • One participant recommends studying transmission lines and impedance matching for those designing circuit boards at higher frequencies.

Areas of Agreement / Disagreement

Participants generally agree that some knowledge of analog is beneficial for working with digital systems, but there is no consensus on the extent of that knowledge or whether additional coursework in analog is necessary. Multiple competing views exist regarding the importance of analog understanding in practical applications.

Contextual Notes

Participants express varying levels of experience and familiarity with analog concepts, leading to differing opinions on the relevance of analog knowledge in their work. Some mention specific challenges faced in real-world applications, while others focus on theoretical understanding.

Who May Find This Useful

This discussion may be useful for computer engineering students, professionals working with digital systems, and those interested in the integration of analog and digital technologies.

Elecomputer
Messages
8
Reaction score
0
Right now I am a computer engineer. I really love digital and want to focus on it because it uses logic.

Analog is a bit more complex in that signals are not only 1 or 0, but sampled with many values.

Anyways, I took digital logic and design courses, but I am unsure if i should take any analog. My concern is taking in real world signals (via sensors, for example). How much analog would I need to know to convert real signals into digital signals for digital computing/processing?

Sorry if my wording is weird.
Thanks in advanced!
 
Engineering news on Phys.org
Real world digital electronics is in not way divorced from analogy circuitry techniques. You can design the LOGIC of digital circuitry without analog but if you actually want to make things work you need to understand what's going on when, for example, you get edge-spiking causing ringing problems that turn gates on and off in a non-digital-logic kind of way.

Also, A to D and D to A converters require knowledge of analog. I personally hate analog and love digital but I learned a lot of analog and found that it was all needed in real-world debugging situations.
 
I figured the same. I am currently taking signals and circuits. But really find it too boring and question whether i should pay more for analog. And i agree, I am talking about A to D and D to A converters.

Not sure if you'd understand this, but would it be worth taking a second circuits course or go into electronic circuits. I want to at minimum be able to grab real signals and convert them to digital, and vice versa. Thats pretty much all I want from analog.
 
Check out these two links and see if any of that stuff looks interesting. :)

http://www.faqs.org/docs/electric/

http://ecee.colorado.edu/~bart/book/book/contents.htm

Regards
 
Elecomputer said:
I figured the same. I am currently taking signals and circuits. But really find it too boring and question whether i should pay more for analog. And i agree, I am talking about A to D and D to A converters.

Not sure if you'd understand this, but would it be worth taking a second circuits course or go into electronic circuits. I want to at minimum be able to grab real signals and convert them to digital, and vice versa. Thats pretty much all I want from analog.

It's been almost 50 years since I took EE so I don't really know what's in the various courses given these days and can't advise you on that.
 
Elecomputer said:
My concern is taking in real world signals (via sensors, for example). How much analog would I need to know to convert real signals into digital signals for digital computing/processing?

Sorry if my wording is weird.
Thanks in advanced!

My only computer interfacing experience is in process control where 10 hz was quite high frequency.

Where I've seen computer folks get into trouble is almost always right at the interface, and by two mechanisms:
They fail to appreciate the concepts of "Common Mode Voltage" and grounding;
and the concept of "loading" the analog circuit they are measuring.
Latter is particularly troublesome when their A/D converter loses power so the analog circuits they tapped into get loaded down and possibly tied together through the multiplexer. Most muxes need power to maintain high input impedance. It's embarrassing when your computer makes the control room indicators go haywire.
Former is troublesome when the analog circuit exceeds the modest common mode capability of most solid state multiplexers, typically ~15 volts, and the computer cannot read them.

These are simple problems solved by Kirchoff's laws.
So you need to know basic DC circuit analysis and work lots of practice problems.
I presume your computer science curriculum teaches about anti- aliasing filters, so you'll need some AC circuit analysis as well.
These are both introductory level EE courses, or were in my day.

Doubtless there's another whole world of DSP, maybe someone from that community will offer more insight ..

old jim
 
You really don't have to understand much analog to interface to ADCs and DACs and the like. Usually what you need to know is described in the datasheets. If you need to know more, Analog Devices has some stellar tutorial information in their free text on dataconversion. Check it out:
http://www.analog.com/library/analogdialogue/archives/39-06/data_conversion_handbook.html

Pretty much everything you need to know to successfully interface ADCs into your system is in there.

I'm a professional analog circuit design engineer and I highly recommend this book.
 
@Elecomputer
You really need to appreciate that all so - called Digital Signals are ANALOG. The Digital stuff is the information they carry and the 'next stage' in any 'digital' circuit always needs to dig the digital information out of the analog signal it has been handed. We are lucky, these days, that so much digital circuitry allows us to be really sloppy about the analog nature of many designs - that is, until you want to stretch your processor speed or do real time DSP - or even just get your comms link to pay for itself!
Get into some serious Analog basics. It will repay you, in the future.
 
  • #10
If you are going to be designing computer circuit boards particularly at higher frequencies, you would do well to study transmission lines, especially with respect to impedance matching.
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 10 ·
Replies
10
Views
2K
Replies
13
Views
4K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
Replies
7
Views
4K
  • · Replies 26 ·
Replies
26
Views
8K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 4 ·
Replies
4
Views
7K
  • · Replies 14 ·
Replies
14
Views
6K