+ Post New Thread
Results 1 to 4 of 4
  1. #1
    Member level 5
    Points: 634, Level: 5

    Join Date
    Jan 2018
    Posts
    87
    Helped
    0 / 0
    Points
    634
    Level
    5

    Steps for first cut design

    Hi, When I have a circuit to design, I usually choose some topologies from literature, jump to simulation and keep playing with simulator without even knowing how far I can reach specifications with that topology. If small circuits are there, for example up to two stage cascode amplifier, I can work out things on paper and know how much gain and band width I may get.

    How do you people design first cut design, either using MATLAB or some other procedure, before even starting with simulation. Please provide relevant documentation if you already know.

    I am exactly talking about what this guy is saying in this page, but I couldn't really find answers to my questions in there.
    https://electronics.stackexchange.co...s-using-matlab

    •   AltAdvertisement

        
       

  2. #2
    Advanced Member level 4
    Points: 5,610, Level: 17
    kripacharya's Avatar
    Join Date
    Dec 2012
    Location
    New Delhi
    Posts
    1,017
    Helped
    173 / 173
    Points
    5,610
    Level
    17

    Re: Steps for first cut design

    No we don't do all that.

    First we absorb all the requirements stated, and try to ensure we understand the intent behind those.

    THEN, based on our own personal current experience and knowledge of the state-of-the-art and our own experience/ expertise level we choose a variety of approaches (all mentally) which might do the job. There are many parameters against which this is tested (again.. mentally). Like ease of development/ confidence in outcome, cost/ viability, development time, peripheral equipment/resources required for development. Maybe a quick research on some tricky points we are unsure about.

    Next we sketch out .. (on paper Gasp ) a few ideas of the critical elements, and just think about whether they will work, or something else is required. MAybe look online for the latest black-boxes which could do the trick better than we could. Maybe have a few alternate methods outlined, along with the pros and cons of each.

    Then we sketch out a fuller method based on our most best feel design selection. Flesh it out with real world components, see if those exist, and if they do then at what price/ spec.

    Only as a last stage do we go MatLab, or Spice or whatever, to try understand where the screw-ups could be. And maybe tweak some values.

    Choosing a topology and jumping direct into simulations is usually a waste of time without understanding what the limits of a particular design are. It's equivalent to the million monkeys at typewriters situation.



    •   AltAdvertisement

        
       

  3. #3
    Member level 5
    Points: 634, Level: 5

    Join Date
    Jan 2018
    Posts
    87
    Helped
    0 / 0
    Points
    634
    Level
    5

    Re: Steps for first cut design

    Well, that helps. But I am a beginner, what do you suggest to me, how to come to the conclusion that which topology works before even getting to simulations.



    •   AltAdvertisement

        
       

  4. #4
    Advanced Member level 4
    Points: 5,610, Level: 17
    kripacharya's Avatar
    Join Date
    Dec 2012
    Location
    New Delhi
    Posts
    1,017
    Helped
    173 / 173
    Points
    5,610
    Level
    17

    Re: Steps for first cut design

    Quote Originally Posted by circuitking View Post
    Well, that helps. But I am a beginner, what do you suggest to me, how to come to the conclusion that which topology works before even getting to simulations.
    Read. Experiment. Calculate. Do. Read some more. Try and understand what a particular design is capable of and what the limits & constraints are. Experiment again. Run simulations & compare the results with real life results. Try and reconcile why there are any differences. Those differences teach you a lot. PAY ATTENTION TO DETAILS. Read and study some more. Get on forums and learn from your peers/ elders advice and guidance. Repeat this process for several years.

    And soon you will develop that sense of what should work and what might not. You live in an era which has the internet... and Google. The ability to acquire information is so huge & quick now it is unprecedented. You can do in 1 year what it took some seniors here maybe a decade.

    That's all.



--[[ ]]--