Writing my own PCB designing software

Thread Starter

eecurious

Joined Jun 23, 2015
1
Hello All,

I am new to the wonderful world of PCB designing. I am the kind of person, who believes in doing everything from scratch... which for me as a computer programmer means writing my own software. I want to learn the black magic of how schematics are actually converted to detailed PCB layouts by software. I want to understand why the final PCB layout has all those lines and strange expanding shapes... why is everything the way it is? Are there any good books you can refer me to? Many thanks in advance for your help.

Regards,

eecurious
 

GopherT

Joined Nov 23, 2012
8,009
So, what hobby will you have for the next 5 years? Writing code for pcb design vs using a pcb design software and then designing a circuit?

Did you write the browser that you used to make this post? I smell a troll or a ...?
 

ScottWang

Joined Aug 23, 2012
7,397
Sometimes people just do things for the experience of doing them.
To experience the experience is a good thing, as what I did is to used VB to design a software run in windows xp environment and using the software to reading the results from 8051 compiler and link function which their original working environment are in DOS.
 

#12

Joined Nov 30, 2010
18,224
Sometimes people just do things for the experience of doing them.
Yebbut...re-inventing the wheel uses so much time, and humans are time limited. Would you go back to using a club and a fire and invent the works of Issac Newton by writing with charcoal on the paper you made from bamboo? No.

I usually refuse to design circuits for products you can already buy cheaper than you can build them, so I am at the other end of this belief system. I think a lot of people don't realize how many decades of human effort were invested in things they see every day. They tend to think, "If I can see it, I can build it". Those days ended when I was a child.

Before you can design an auto-router, you must understand how to design a circuit board by hand. That will require a few years, and you will still be studying RF techniques and gigahertz propagation speeds for a few more years. Or...you can learn (most of) an auto-router program in a few weeks. Nobody can know everything. Will you stand on the shoulders of those who went before, or will you spend your life (figuratively) playing in the mud while the world passes you by?

You don't start to build a house by building a mud fire pit and gathering some black sand with which to forge a hammer, you just go buy a hammer.
 

GopherT

Joined Nov 23, 2012
8,009
You don't start to build a house by building a mud fire pit and gathering some black sand with which to forge a hammer, you just go buy a hammer.
In my case, I have skipped the hammer as well - I just buy newly built houses from guys who buy hammers.
 
Some folks do things for a challenge or they are just passionate or feel artisanal. Perhaps they are just bored and need a long project.

Circuit design is a good example of supplying all those things. There are those who would design tube amps or transistor amps and not use IC's.
It's a balance. Yesterday I abandoned an NCP3063 SMPS 1.5A design I had done (cost me 3 days to design, build, integrate & test) and 'upgraded' to an Ebay 5A SMPS module cuz it represents a single component in my BOM rather than a few new ones AND it was cost competitive with the much lower power NCP3063. I learned a couple things along the way though. Our ability to integrate, modularize and then mass produce gives each generation greater capabilities.
 

crutschow

Joined Mar 14, 2008
34,280
...................
Our ability to integrate, modularize and then mass produce gives each generation greater capabilities.
That's so true.
My last standard-def TV I bought many years ago was a high end 25" Sony that I paid over $700 for and it used a very heavy and bulky CRT .
Now I can buy a light-weight 50" high-def flat screen TV for about that same price. :D
 

sirch2

Joined Jan 21, 2013
1,037
I take it the OP who is "kind of person, who believes in doing everything from scratch" has grown their own silicon crystals, made a processor, boot-strapped it with a patch panel, coded their own compiler and used that to build an operating system? These things are just a question of degree, if you really did everything from scratch would would never get further than collecting water and fire-wood and growing food, there would not be time for anything else.

There is nothing wrong, in fact there is everything right, with recreating something that already exists to either learn the nitty-gritty or improve upon it. However the best place to start is being honest with yourself and that means saying "I want to learn CAD programming" or "I want to learn PCB routing algorithms" or "I want to leant the ins-and-outs of Gerber".

Please start by downloading and using more than one of the freely available packages. Then decide if you really want to write your own or just learn how to use which ever of the free packages your prefer.
 

WBahn

Joined Mar 31, 2012
29,976
Our ability to integrate, modularize and then mass produce gives each generation greater capabilities.
As you say, it's a balance. And (I believe) it's something that we haven't done a very good job of balancing (which is admittedly hard). In many ways we have become victims of our own technological success. Each generation increasingly uses things that they simply do not understand and that a decreasingly smaller portion of that generation could reproduce starting from the point of a prior generation. Worse, the educational establishment tracks that trend and figures that people that have access to these astonishing capabilities have no need to understand how they work or how you could get by without them.
 
. Worse, the educational establishment tracks that trend and figures that people that have access to these astonishing capabilities have no need to understand how they work or how you could get by without them.
Shadows of Isaac Asimov's 'Foundation' series, in which is reflected the Rise and Fall of the Roman Empire, as a result people suffered thru a 'dark ages' where prior technology had to be rediscovered/developed.
 

DickCappels

Joined Aug 21, 2008
10,153
Think about it: If nobody ever undertook a project because it had been done before we would still be listening to music on Sony Walkmans.
 

WBahn

Joined Mar 31, 2012
29,976
I strongly support anyone that wants to do something from a more fundamental level for the purpose of learning how it is done and what is involved. I would recommend that the scope of such a project, however, be carefully constrained. Do just enough to learn the concepts (and tricks and pitfalls and whatever else), but don't get in love with the idea of reinventing the entire thing unless you are serious about sinking a huge amount of blood, sweat, and tears into it. Learn the concepts, and then use what others have polished ahead of you -- you'll not only learn a huge amount along the way, but you will be in a much better position to more effectively use the polished tools that are available.
 

#12

Joined Nov 30, 2010
18,224
It's a balance. Yesterday I abandoned an NCP3063 SMPS 1.5A design I had done (cost me 3 days.
I would recommend that the scope of such a project, however, be carefully constrained.
And, that's it in a nutshell. You aren't talking about a 3 day study session, you are talking about 3 years. When you get that much progress, there will still be better versions available, probably for free.
 

wayneh

Joined Sep 9, 2010
17,496
I cannot imagine investing blood, sweat and tears to write software that, if you're lucky, reproduces what can already be done with cheap or free software from multiple sources. Especially in such an arcane and specialized art as PCB design.

There are SO many new technologies coming in the next few years that would be here already today if the software was caught up to the hardware. Why not work on that?

I've been struggling along making an app for myself that uses my own control algorithms to operate smart-home devices via Apple's HomeKit. It's hard - mostly because I'm a noob - but also because there are no examples to follow. This is new ground. If I succeed, my app will be the only one of it's kind, ever.

My point is, Why not work on something like that? I could use a little help! ;)
 

RichardO

Joined May 4, 2013
2,270
I cannot imagine investing blood, sweat and tears to write software that, if you're lucky, reproduces what can already be done with cheap or free software from multiple sources. Especially in such an arcane and specialized art as PCB design.
Good point ... but.... How did those man different versions come about? Someone thought the could do it better and and put the effort into doing it. Having said that, the effort was huge -- maybe even from an experienced team.
 
Top