Electronic – Thought Process on Designing Circuits

analysiscomponentssynthesis

I'm currently in my second year of computer engineering. I've taken circuit analysis courses and have a good general understanding of components, and a little bit of digital design, however, I'm very confused on how one actually goes about designing circuits to complete a task. For example, what if I wanted to make a circuit that would be used for some sort of door-lock that takes a 4-digit combination (inputted through a 10-digit number pad) along with a physical key to open the door – how would one design something like this?

Thoughts that come to my head:

  • I suppose the easy part would be the physical key and hardware
    • this would include the actual key and lock mechanism
  • Now, what I'm very curious about is the actual circuit that controls the 4-digit combination
    • (correct me if I'm wrong) there has to be touch sensors within the keypad that has values for all numbers 0-9
    • of course, this circuit will have resistors, sensors, voltage source, etc
    • Some processing must happen to determine whether or not the 4 digits entered are correct or not

More specifically, my question is (are): What is the thought process behind designing this circuit? How does one choose the correct components? How do you create a schematic through reverse-analysis (essentially synthesis)? Not all of these have to be answered, but I just want a general understanding. I'm curious as I'd like to begin creating circuits that do certain things, but I'm lost on how to think about them.

Best Answer

Think: Well first you think about what you want to do, why you want to do it and for who are you doing it for. Is it just for yourself to learn? Or are you hoping to take the combination lock market by storm with your new iphone controlled padlock?

Research: Then you do a little research. What kind of electronic combination locks are out there today? What are their features that you'd like yours to have. Are there combination lock chips with app nots as @pjc50 says? Have people written papers on other interesting information about electronic padlock design? If this is a commercial product who is your competition, maybe you should buy a bunch of locks try them out and take notes. Then rip them apart and see how they work, maybe you'll learn something you didn't expect.

Requirements: Now that you understand what you want to do write down your requirements, this could be a simple list of features and needs, or a more elaborate document. If this is just you a list will suffice. Maybe more people are involved, maybe your boss or marketing asked you to make this, and you need to go over the document with them so you are on the same page. Might make sense to ask some customers for feedback on the features of your new lock.

Spec: Now you write the spec. You'll be researching parts you can use in your final solution, power, mechanical parts for the lock, chips etc. At this stage you are thinking through and most importantly writing down what you are going to do and how you are going to do it. When you are finished you will review it and think it over. Maybe you have more work to do to calculate your battery requirements for this lock. Did you think about how much heat will be generated, any idea on the cost? Might want to put that in the spec.

You should also be thinking about who is going to make this or where it will be made.

Optional Test Critical Circuits: At this point maybe you have some circuits you should simulate that are smaller parts of the whole. Maybe there are things you don't understand that you should build up in the lab first. Such as will this solenoid I picked be powerful enough to unlock the lock? Put these results back into your spec.

Schematics: Ok time to go you can create the schematics now based on your spec. When you find something that won't work the way it was specified make sure you go back and update the spec. Go through the design process picking parts connecting wires. Review, have a peer review it, or your whole group. Update the design and review it again. Oh and make sure you can buy the parts you picked and there isn't a 20 week lead time.

Layout: I won't go into detail, but here you physically draw out the circuit board.

Send to Fab: Now send the board out to be built, wait for it to come back. Pay for electrical test and solder mask, and silkscreen...

Assembly: For you personally just assemble it by hand if you can. If you're at work, send it out.

Test on your bench: Get it back, check for shorts, power up, check voltages. Now start testing your functionality. Keep a record of all the things you did wrong that you will fix in your next revision.

A little bit over simplified especially in the end there, and the proto to production process is another story. There are different requirements if you are just designing a circuit for yourself or doing it for work.

And don't go down the road of saying, "I don't need a spec or requirements this is just a simple circuit". Every design benefits from thinking through the requirements and the design before you start "designing". So even if you just write down 5 requirements, and take the time to write one page on how you will do the design, you will already be way ahead of the game.