Electronic Design

Bob's Mailbox

Dear Bob:

Sure enjoyed your Spice article in Electronic Design.  I've followed your musings for years, gosh, decades.  Please don't ever stop.  There is another aspect of the simulation thing.  It is a matter of philosophy, I guess, and it has concerned me for a long time:  The sense of accomplishment that comes from doing something the first time.  Once you've done it, you've done it, you've done it (how many endure the agony of climbing Mount Everest a SECOND time?).  In the real world, this concept is intuitive, I think.  But in the world of simulation, we confuse what we have accomplished with what we have imagined; worse yet, what the computer has imagined!  Dreams and reality mingle, and we are led astray.

Case in point:  Stirling engine development (one of my loves) has gone high tech.  The talent and money is flowing toward the computer, not the lathe.  We simulate, learn, simulate, etc.  Each time we make an improvement it is rewarding, and provides the psychological boost needed to continue.  Because it isn't so rewarding to do the same thing again, we don't bother to repeat in metal what we've already "built" on the screen.  Soon the simulation errors compound until there is no way to distinguish dreams from reality.

I sometimes wonder if we could have experienced the industrial revolution at all, if somehow the computer had already existed.  It is very tempting to simulate a thing rather than constructing it.

Simulation ain't the same as really doing it.  Perhaps a close parallel? Or, to say it another way, is it possible that the boob tube in the office takes us on the same sort of journeys as the boob tube at home?

Darryl Phillips, President, The Airport Corp., Sallisaw, Okla.

I couldn't agree more.-RAP

Dear Bob:

I just finished reading your column "What's All This Spicey Stuff Anyhow? (Part 1)" in the Nov. 22 issue and I thought I'd share my own experiences with circuit simulation and EE education with you.

I am 33 years old, and was trained in math and computer science.  Two years ago, I decided to pursue a Masters degree in EE at a well-known northeastern university.  I told the boss that I was going to specialize in the mathy end of EE, like Control Theory, Signal Processing etc., but I had a hidden agenda.  I wanted to learn electronics.

It's all worked out because I have learned electronics, but only sort of.  You see, at this institution's outlying campuses, where they crank out BSEEs and MSEEs by the score, there are no labs.  I repeat, no labs.  No breadboards, no scopes, no components, no nothing.  Any design that takes place is on the computer using a circuit simulation program.  I have suggested to several professors that a real live lab would be very helpful, but they just "hurrumph" about the computer being just as good as a learning experience.  So you see, it's not just lazy (young or otherwise) engineers copping out on the design process, it's the educational system taking the path of least resistance.

I have improved the situation for myself somewhat.  I designed and built a dc power supply (you're probably saying "big deal," but that's pretty good for a mathematician!)  I bought a breadboard, a selection of components, and a digital meter.  With this setup, I can do quite a bit of experimenting on my own.  My father, an "old tyme" electrical engineer, is so glad that his mathematician son has seen the light that he's salvaged an old scope from who knows where for me.  I'm looking forward to its arrival.

As far as my experiences with the circuit simulation program are concerned, I concur with your opinion.  We didn't use Spice, but some other Spice-like program.  It was graphically oriented and very user-friendly, but it had the same crazy convergence problems that you describe.  One project we had was to design an analog multiplier.  Changing the quiescent current through the Gilbert cell ever so slightly made the difference between fast convergence for steady-state analysis and sitting for 15 minutes watching the cursor blink.  Then, to add insult to injury, the case that converged for steady-state analysis didn't converge for Fourier analysis.  I finally was forced to give it up.  Interestingly enough, my professor accepted "Program Did Not Converge" as a correct answer.

Christopher Lennon, Bedford, Mass.

Ouch!  This guy's heading in the right direction, but the schools aren't.-RAP

Dear Bob:

I feel that your most recent column (Spice, Part II) needs some comment.  You say, "I don't think you can beg or steal or borrow or buy a model of a transistor that's guaranteed."  First, to what kind of transistor are you referring:  bipolar, JFET, MOSFET, small signal, large signal, power, microwave, discrete, integrated devices?

Second, your statement implies that a single "transistor" model should be guaranteed to "run" under all conditions.  Suppose I suggest that you should be able to design a single OP AMP that works equally well for all applications? I suspect that you would not only laugh in my face but that you would think to yourself that I am a fool for having made the suggestion in the first place.  Well, to suggest that a single transistor model should "run" under all conditions is equally laughable.  Am I hitting below the belt?

Third, you seem to want to place the total responsibility for the success of the model upon that person who developed the model.  However, I get the impression that you have no sympathy (or respect?) for a user of one of your op amps who misuses your op amp.  Below the belt again?

Fourth, you seem to believe that those "transistor" models which have been built into the circuit simulation program Spice represent the complete set of models from which one can choose.  The sad truth is that those "transistor" models built into Spice represent just about the worst set of models from which one can choose.  Why is that the case?  Because industry gets those models for free from the Univ. of Calif. at Berkeley as the result of student slave labor and typically is not willing to pay the price for truly physical and accurate models.

....You suggest that you may wish to BUY a guaranteed transistor model.  Well, you can.  My company has developed and markets Spite (SPI Transistor Emulator), the world's best CIS/SIS (MOS if you prefer) transistor model.  The Spite model does employ physically meaningful parameters and parameter values.  It also simulates, with unprecedented accuracy, a transistor which is properly used.  Would you like to guarantee me that your op amp will always "run" for me no matter what I do to it?  There just isn't any point in building a lot of unnecessary complications into the model to protect the unknowledgeable user from himself.

So what do you want your guarantee to say?  Send me a reasonable specification and a purchase order and I will write your guarantee for you.

Dr. James E. Smith, President, Semiconductor Physics, Inc., Escondido, Calif.

I use mostly bipolars, but for CMOS this looks tempting!  -RAP

 

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish