Register to reply

Prerequisites for Digital Electronics

by nDever
Tags: digital, electronics, prerequisites
Share this thread:
nDever
#1
Feb26-12, 02:22 PM
P: 66
Hey guys,

I have designed and built digital circuits with relatively low knowledge on circuit analysis (resistor nets, Wheatstone bridges, etc.) or analog electronics. I do know how to utilize Kirchhoff's circuit laws and Ohm's law. Is it necessary to understand analog systems and circuit analysis in order to develop successful digital systems? If so, what material should I start with?
Phys.Org News Partner Engineering news on Phys.org
Printing the metals of the future
New gadget helps the vision-impaired to read graphs
3D printing helps designers build a better brick
psparky
#2
Feb27-12, 10:06 AM
PF Gold
psparky's Avatar
P: 731
Quote Quote by nDever View Post
Hey guys,

I have designed and built digital circuits with relatively low knowledge on circuit analysis (resistor nets, Wheatstone bridges, etc.) or analog electronics. I do know how to utilize Kirchhoff's circuit laws and Ohm's law. Is it necessary to understand analog systems and circuit analysis in order to develop successful digital systems? If so, what material should I start with?
In my opinion I would say your knowledge is already sufficient.

From my experience, digital systems are ussually made from simple computer programming in various programs they will have in your school. Write the code...verify the inputs and outputs....burn the chip....wire it according to the diagram...there ya go.

But certainly feel free to learn more about anything electrical.
jim hardy
#3
Feb27-12, 11:39 AM
Sci Advisor
PF Gold
jim hardy's Avatar
P: 3,518
You should be fluent enough to appreciate the input and output characteristics of various logic families so you dont try to interconnect things that shouldn't be interconnected.

http://www.analog.com/static/importe...als/MT-098.pdf

yungman
#4
Feb27-12, 02:42 PM
P: 3,883
Prerequisites for Digital Electronics

If you just doing one project then if you can do it, you are fine. But if you want to make a career out of it, you should learn the basic circuit analysis minimum. But I would say you should go deeper.

Digital speed is going up into hundreds of MHz, they become analog. I saw many digital engineer just got lost because they have no idea how to deal with high speed system and signal integrity. You can get away 25 years ago, not now if you want to make a career out of it.

There are a lot of transmission line effect, reflection, cross talk type of issue you have to be aware or deal with, or else company has to hire a signal integrity engineer to help you, this is not going to help you marketability.

If you have to design USB circuits, you are even knocking on the door of RF. Digital logic, boolean alg. FPGA programming are the easy stuffs, you learn it in two to three month on your own, those are dime a dozen OR less.
Mike_In_Plano
#5
Mar9-12, 09:15 PM
P: 560
I wouldn't say modern digital design is easy, I don't attempt it, and I stand amazed at a few people who do. None of them could say much about analog or RF, but they know the means and methods of their trade.
It's easy to build a counter, it's more interesting to interface DDR memory. It's also pretty challenging to build equipment that meets FCC part B on emissions, or can get through ESD testing. For that matter, things get pretty interesting when you fail to perform your timing analysis and equipment fails.
So, the same rules apply as with any trade: Get the basics at school and find a job that can train you.
yungman
#6
Mar9-12, 11:59 PM
P: 3,883
Quote Quote by Mike_In_Plano View Post
I wouldn't say modern digital design is easy, I don't attempt it, and I stand amazed at a few people who do. None of them could say much about analog or RF, but they know the means and methods of their trade.
It's easy to build a counter, it's more interesting to interface DDR memory. It's also pretty challenging to build equipment that meets FCC part B on emissions, or can get through ESD testing. For that matter, things get pretty interesting when you fail to perform your timing analysis and equipment fails.
So, the same rules apply as with any trade: Get the basics at school and find a job that can train you.
The blue color part is analog in every sense, that's exactly what I refer to that a lot of them don't know how to do it. I spent years designing digital electronics, including Micro-processors boards, embedded system, Firewire, USB, FPGAs etc. They are easy if you don't deal with high speed end of things where it is......again analog. Digital design require a lot of reading, learning how to program the VLSI chips. But they are all very easy to understand. You don't get stuck like in analog particular as speed goes up.

Digital design in certain ways similar to programming, it is so easy to learn. You learn the first language and it will take you maybe two weeks to learn another language to get started. But it is complex and a lot of headache to trouble shoot....nothing hard, just a lot of trouble. Digital design has a lot of signal to keep track of, documentation has to be good. No one signal is that hard, but you have a whole lot of them to keep track.

If you are into computer and digital design, it might take you a few months and you'll have the basics of it. But it'll take years to learn analog, RF. There is no comparison. As a person that spent years designing both, digital design cannot even hold candle to analog design when come to degree of difficulty. Put it this way, I started out in digital, only take me a little over a year from first learning digital electronics to be a full blown design engineer designing and programming an Intel 8085 based system. That's how I started my career, what do I know about electronics and digital electronics were so slow you really don't need to know much electronics.

Look at how many kids playing around digital and everyone design a little things. Not as much as programming as those are even easier. And yes, I was almost a full blown programmer as I programed the embedded system in the early years. You'll find 10 times more digital engineer than analog......why?..........most got stuck in analog, EM. I talked to quite a few EE, they barely creep through EM and so glad they got through it.

Also, problem with digital design, your knowledge can be obsoleted fast. I was so good in the early 8080 system, I can write long programs in machine language....just keep typing in hex numbers!!! Then 8080 was gone, then I learned 68000 16 bits, gone!!! 10 years ago, I learned Firewire, all the PHI, Link layers etc. ...........Gone!!! All those knowledge don't mean a thing anymore. BUT we use transistors 50 years ago, we use op-amp 30 years ago.....we still use every bit of it. I am currently helping two people here on transimpedance amplifiers.......it is all op-amp theory mostly learn in the late 70s and early 80s.

Yes, timing diagram are the hardest, I draw full blown timing diagram looking a the min and max prop delay to make sure they come together. You'll be surprised how many so called digital engineer never draw timing diagram!!! This is really the necessary evil of digital to guaranty reliable design. I even put a simple timing diagram drawing as part of the test when I hire engineer or technician. You'll be surprised how many fail. Just two D flip flop one as divided by two and one is a pipeline. A lot of them can't even do it. If they fail, the interview ended.
Mike_In_Plano
#7
Mar11-12, 08:10 PM
P: 560
Yungman,

I must, respectfully, disagree. It takes a very long time to finish the education of a true analog engineer. There is a vast array of circuits, technologies, techniques, and compromises that an analog engineer must become familiar.

A digital engineer will never be confronted with 1/f or schott noise. He'll never have to deal with leakage current, gain peaking, temperature sensitivity, or nonlinear correction circuits. Most will never know what a fempto-farad or charge injection is.

Likewise, Digital engineers don't have to design antennas or their matching networks. A Smith chart is something that goes by in college and is never seen again. Designing oscillators, PLLs, or stable amplifiers is totally outside of their needs / responsibilities.

There are plenty of things that a digital engineer must learn about keeping his signals clean, protecting his I/O, and dealing with skew. Even the guys who program programmable logic are not immune to timing difficulties, and this is part of their specialty. All of this will come with time and being on the job. If he attends signal integrity classes, or is fortunate enough to have one in house, he'll learn what he needs without all the rest.

Respectfully,

Mike
Windadct
#8
Mar12-12, 02:41 PM
P: 554
As a Prerequisite - you must love the smell of fried ICs
yungman
#9
Mar12-12, 03:11 PM
P: 3,883
That too.


Register to reply

Related Discussions
Help! Course Selection: Analogue Electronics versus Digital Electronics Academic Guidance 2
Digital Electronics Electrical Engineering 6
Digital Electronics: State machine Engineering, Comp Sci, & Technology Homework 4
Need practice in Analog/Digital Electronics... General Physics 1