Math and electrical desgin

I design instrumentation, like hardware acquisition and simulation of aerospace transducers, picosecond timing, laser controllers, fiberoptic gadgets. Emphasis is on the analog parts, with FPGAs and uPs behind that, with a fair amount of digital signal processing. My FPGA guy does a lot more real math than I do. My embedded firmware guy probably does none.

I do a lot less math than I used to, and mostly "design" by simulation. I have to intuitively understand the concepts of circuit theory, electromagnetics, and signals-and-systems, and have to be able to estimate the order of magnitude of things, to know what will matter and what won't, but I don't have to do closed-form solutions to Maxwell's equations.

A benefit to this instinct-and-simulation style of design is that I'm hugely more productive than I was in earlier times. And the products are immensely better and more reliable.

I'm trying to teach this to the kids who work for me. Use and train your instincts, because equations only analyze things. SEE how convolution or Fourier transforms work, don't just plug into textbook equations.

Someone recently posted a link to a video where Mike Engelhardt talked about LT Spice. He said the real value of Spice is train an engineer's instincts.

You learn math so you can write the software/script that does the hard work of finding the answer for you. Have to learn how to ask the question, though.

Basic electronic design doesn't require much "advanced math" because the state space is huge, every problem is massively under-determined, the full equations of state for even many relatively simple circuits are intractable and impossible for any person to comprehend. you can get by pretty good on heuristics that work well enough in many cases. thank goodness for that or nobody could get anything done.

When they break down though being able to write some of your own scripts is very helpful

rself an applied mathematician? I would say that I do. As an analog desig ner how much do you use various mathematical concepts. Basic electronic ci rcuit design does not seem to require a lot of theoretical math, where as, signal processing requires a lot of math.

ebra. I am on my 3rd pass ( of the first 10 lectures) of Gilbert Strangs o n-line MIT course. I think I am finally getting it. I really want to know linear algebra because I want to be good at matlab which is rooted in line ar algebra. However, is linear algebra used in circuit design other than a n occasional solving of two or three simultaneous equations?

nal processing stuff is definitely math intensive and required to be good a t it.

uper proficient at maxwells equations.

cannot accomplish his task with out the math?

on"

Getting those instincts is a big part of being an "applied" mathematician.

Very few interesting mathematical problems are solve-able in closed-form. You learn linear algebra so you can write simulation software. Or at the least, understand how the physics/electronics/digital signal processing chain/weird dynamic system you're trying to stabilize-math/computer solvers work well enough to ask the computer meaningful questions.

A practical example: do you know how easy it is to find the transfer function of an arbitrary-length ladder-network of arbitrary RLCs in series and shunt using matrix multiplication and ABCD parameters, with a little help from a symbolic math package like Octave? Easy as shit! I use that one all the time.

Check out Wes Hayward's "Introduction to Radio Frequency Design" for a relatively math-intensive but also pragmatic, real-world-circuit oriented RF design book. The step-by-step walkthrough of designing PLLs, low-distortion amplifiers, and oscillators, using math, is very informative

Or a creative physicist. Visualization of concepts comes before writing equations.

Einstein had some interesting thoughts about that, as did Richard Feynman.

On Friday, March 27, 2020 at 3:29:36 PM UTC-4, snipped-for-privacy@columbus.rr.com wrote :

self an applied mathematician? I would say that I do. As an analog design er how much do you use various mathematical concepts. Basic electronic cir cuit design does not seem to require a lot of theoretical math, where as, signal processing requires a lot of math.

bra. I am on my 3rd pass ( of the first 10 lectures) of Gilbert Strangs on

-line MIT course. I think I am finally getting it. I really want to know linear algebra because I want to be good at matlab which is rooted in linea r algebra. However, is linear algebra used in circuit design other than an occasional solving of two or three simultaneous equations?

al processing stuff is definitely math intensive and required to be good at it.

per proficient at maxwells equations.

annot accomplish his task with out the math?

Math is the foundation of all science and engineering.

That said, if you aren't building the foundation, you can use fairly basic math and git 'er done!

Once in a while I actually use calculus. I've sometimes used differential equations. I use very basic forms of abstract algebra which includes Post algebras which includes Boolean algebra. But more than all the others put together what I use is basic algebra. I can't say I've knowingly used line ar algebra although some of the things I've learned may well be part of tha t.

So your work is not so much circuit design as it is system design?

A lot of what I do is (a) for a given electro-optical design concept, calcu late how good it _could_ be, (b) figure out how to get there, and (c) if th at's a win, proceed, but if not, think up something better and goto(a).

Significant amounts of math are required to get a good result, as well as a lot of lore and old fashioned crank-turning.

Cheers

Phil Hobbs

snipped-for-privacy@highlandsniptechnology.com wrote in news: snipped-for-privacy@4ax.com:

It is not an uninterruptible power supply. It is a microprocessor.

Alt-0181

Just like math, certain key combinations should be known by today's modern computer user.

Looks better. Sounds better. Feels good. Real good. Get some.

Degree symbol is alt-0176

Another way is to hit the windows symbol key and the R key to bring up the Run dialog, and enter 'charmap' in the dialog box and press return. Grab any character you want at that point, and see the key combo down in the lower status bar.

That's the first time I've agreed with T. Bit this year. ;)

Cheers

Phil Hobbs

snipped-for-privacy@columbus.rr.com wrote in news: snipped-for-privacy@googlegroups.com:

I 'apply' it all into my computer or calculator and let it do the math.

On my iPad, I apply it into my Wolfram's Alpha app and it finds out all kinds of goodies on any high end scientific subject.

One of the apps I was glad to pay for on my iPad.

You are a ass hole

And you are just intentionally setting out to annoy the Greeks by

Mike.

Very good book, it was definitely over my head when I first picked up a copy about 12 years ago (bedridden due to illness at the time ironically enough, thankfully that passed) and I had to come back and approach it repeatedly to start pulling "ahhhhh I sees" out of it but it paid off eventually.

Like jamming with a musician who's much more experienced than you but still fairly down-to-earth Hayward is the kind of author that tries to really pull you up to his level whether you want to or not.

My copy is pretty worn and dog-eared by this point but I can't seem to find it on Amazon anymore.

Mike Coon wrote in news: snipped-for-privacy@news.plus.net:

Nag nag nag... Oops

Brainus Bustus

I would say other books like that are, in the audio-world Douglas Self's books on power amp and small-signal audio design, the Bracewell book on the Fourier transform in the dsp-world, "Modern C++ Design" by Andrei Alexandrescu in the programming-world, and of course AoE III free plug! free plug! :)

If I were to count how many times a book is pulled off the shelf and used, the winner would be the Williams filter book. After that, Reference Data For Radio Engineers, for the wire size and screw thread tables. AoE fairly often too.

Imagine how it makes the Latinos feel too! ;)