Blog
There's still a place for elegant, creative circuit design
Bill Schweber
5/18/2011 6:06 PM EDT
In the past few years, I have felt—and told many people—that the age of circuit design is over. By "circuit design", I don’t mean IC design: I mean taking ICs, discrete devices, and passive components, then combining and configuring them to create the distinctive topology which addresses multiple design issues, while meeting an application's requirements.
Why have I felt this way? Because it seems to me that the circuit-design challenge and battle has been largely, although not entirely, won by the IC vendors. That's the case even in the subtle areas of data acquisition and test/instrumentation, where the analog front end must be carefully tailored to the transducer as well as operating conditions. Much of circuit design now consists mostly of selecting the right ICs, being sure that they interface properly with each other and the I/O, and then providing software to make them execute their roles properly.
For example, you can now get high-performance op amps and instrumentation amps (in-amps) for modest cost. No longer do you need to select passives and carefully match temperature coefficients or worry about drift. In many cases, they are now embedded in the IC and thus inherently matched, or the IC has some clever scheme that cancels out many drift errors. Further, vendors offer reference designs and development kits which feature tested and verified circuits for many common, and even some unusual applications.
But as with so many of my pronouncements, I was wrong. Here's why: I recently saw a design and actual prototype done by Jim Williams and Omar Sanchez-Felipe of Linear Technology Corp. Jim's formal title is Senior Scientist, but that's somewhat misleading: he is among the preeminent analog-circuit designers in the world, with a career which spans 40+ years. I still marvel at his first major published design for a portable scale for the MIT nutrition lab, which had to resolve to 0.1 oz, never need calibration, and be built using only standard, off-the-shelf components. It was—and still is—a marvelous example of understanding every subtle source of signal-chain error and then figuring out how to minimize or cancel each one.
What have Jim and Omar done? They started with what seems to be a simple, straightforward way to measure temperature called acoustic thermometry, which is based on the consistent variation in the speed of sound as temperature changes. In principle, this should be an easy circuit: all you need is to take an ultrasonic transducer, pulse it, measure the transit time of the echo, do a simple calculation, and you are all done. What's the big deal?
But principle is not reality; it's not even close to it. The final circuit required careful management of low and high voltages, pulse timing, signal lockout, and many other factors, plus physical placement issues. You can read the full details in their Application Note 131, An Introduction to Acoustic Thermometry here, or an similar version here. While you are doing that, I will use this application and its physical realization in an actual circuit design as a constant reminder that there still is room for what we understand in our gut as The Existential Pleasure of Engineering (to use the title of Samuel C. Florman's excellent book). ♦
|
|
|
|
|
kendallcp
5/18/2011 6:32 PM EDT
It would be lovely to say that circuit design is alive and well - but that's not really true. Those of us with a modicum of experience face a constant challenge to transfer the skills we've acquired to a younger generation, lest they go and make the mistakes we learned so much from making.
The drive to higher performance, more integrated components, and also "everything built-in" user-programmable-system-on-chip parts, wasn't intended to "dumb down" the process of system design, but to make it more efficient. But we should always be alert to this de-skilling tendency. Just because high performance, highly applicable physical or virtual components are _available_, doesn't mean there will always be a trained, thoughtful cadre of engineers out there that can _use_ them. We must carry on teaching, sharing knowledge and stimulating the imaginations of our young engineers. Otherwise circuit design really will die, and the electronic industry age will exp(-t/RC) its way to oblivion.
Sign in to Reply
UdaraW
5/18/2011 10:33 PM EDT
I agree with kedallcp over there.
What has been made available for the purposes of efficiency has prompted the survival and growth of a dumber generation of engineers. In my view, this is somewhat similar to the inability of present day school kids to do their own math due to the availability of laptops/calculators from too early an age in school. Isn't it?
Sign in to Reply
Student4ever
5/25/2011 4:48 AM EDT
You are absolutely right UdaraW. Actually I would stretch that one step further... Not encouraging school kids to do math on their own mental ability has given rise to a mentally lazy generation. I wouldn't say dumber... since I doubt that they lack the inherent ability to perform math.
Sign in to Reply
hm
5/19/2011 12:19 AM EDT
Many applications do not have solution with off-the-self IC. It needs to be resolved with standard electronics parts. I strongly believe there is place for ingenuity in circuit design.
Sign in to Reply
JeffL_2
5/19/2011 1:27 PM EDT
OK let's think of it this way. When I first got into engineering, a "useful" system was sized to be a collection of relay racks. Each rack probably contained a subsystem, each rack unit contained a motherboard containing multiple cards that fulfilled a function (like processing), each card represented a function (memory, processing. I/O), each card), each card consisted of multiple chips which reflected some MSI combination or just the gates themselves, and useful "IP" started at the level of the rack unit going upward. As Moore's Law wreaked its incessant havoc over the years, each level in the hierarchy shifted upwards in function (kind of like a corollary to Moore's Law) until nowadays we have the SoC hence nothing further to integrate (once it is actually developed and deployed of course), and although this model explicitly deals with hardware similar patterns can also be observed in software development. Long term the market for "engineering knowledge" is therefore finite and constantly shrinking. It's altogether too easy to observe this in hindsight, had I been able to see "the end result of progress" would be massive unemployment in my "profession" before I got a chance to even retire myself I certainly would have given a second thought to going into this field, I certainly wouldn't advise anyone else to do so without seriously considering the consequences of this pattern.
Sign in to Reply
kdboyce
5/20/2011 1:24 AM EDT
I kind of look at like the early days of art and artists. The choices of colors and mediums were much more limited than today. Just as there were great artists in the past, there are (and can be more) great artists today. For me, the difference can be found in the interest levels of the 'engineers'. In other words, do they really want to know how things work, or just put together things that work, tinker toy style?
I am not as skilled in the 'art' as those cited in the article, but I certainly have built some complex systems out of basic parts. Not as elegant as it could be, but they worked.
Reminds me of a test for a good design: Can you build it out of Radio Shack parts to repeatable specifications?
Sign in to Reply
agk
5/21/2011 4:13 AM EDT
A correct Principle will come to reality quickly in the hands of expereinced designers. A wrong principle needs more experimentation and comes into reality by many corrections applied to the principle and the design.
Sign in to Reply
bcarso
5/24/2011 12:31 PM EDT
I agree with Bill, but am also seeing the pernicious effects of systems on chips and databook/app note "design". I was saddened and surprised by a piece of Pease's some years back, where he in essence said there wasn't any circuit design to do anymore --- unless one was designing ICs, there was no work.
Since I was at the time doing plenty of circuit and system design of the sort Bill describes, and continue to do so (when clients that need me can find me!), I thought it a quite curious misapprehension. Of course I applied the latest integrated parts where it was appropriate, while attempting to produce designs that did not rely on single source parts.
Even as far back as the mid '70's though, I was getting the advice that I ought to just take it for granted that the people who designed ICs were possessed of a greater wisdom than I could even aspire to. One earnest and annoying man told me that I should use a top-grade OP07 (iirc) for a charge preamp, because he'd been assured by reps that it was the lowest-noise opamp obtainable (this was in the heyday of the aerospace boom, and this guy was being showered with samples in hopes that he'd select something, with cost among the lower priorities). I pointed out that the part was entirely unsuitable, having in particular very high parallel noise, excessive voltage noise, and no benefits to be gained from cooled operation; his response was These parts were designed by PhDs!! I said Well that may well be, but I have a far superior solution that I'm working on. He thought that this was the highest arrogance.
Sign in to Reply
bcarso
5/24/2011 12:32 PM EDT
(continued)
But there are not too many of us left, and as I've remarked elsewhere, you can focus on a symbol-domain curriculum (i.e., "digital") and be able to do some useful work, conceivably, in several years. To gain sufficient analog and mixed-signal expertise to be productive takes far longer, and there are shockingly few good teachers out there. It's a whole lot more fun, in my opinion, but a tough go to be sure.
Sign in to Reply
Jerry.Brittingham
5/24/2011 1:25 PM EDT
Having been both a designer and a teacher for the past 50 years I am consistently amazed at the number of "engineers" who forget that the world, especially outside the chip, is really analog. The rules still apply whether on not you like them.
Sign in to Reply
Student4ever
5/25/2011 5:00 AM EDT
My comments below are in response to your article dated 24th May about the challenges of sensing parameters that do not easily lend themselves to measurement. Sorry I could not find an appropriate place to post my comments / queries and have resorted to putting them here instead.
Your editorial mentions measuring the temperature of the earth's core as one such parameter and measuring the amount of dissolved gas in a liquid as another. On the same lines I would like to ask (all readers as well) if there is any material available on measurement of quantum of heat absorbed by a body? Not by radiation methods but direct measurement.
If any of you know of such work or material, I would be most interested to hear.
Bestregards
anand
Sign in to Reply
BicycleBill
5/25/2011 11:33 AM EDT
Readers: anyone have any ideas for the above query from "Student4ever"?
Sign in to Reply
bcarso
5/25/2011 1:12 PM EDT
Perhaps a good start, the wikipedia entry for Calorimetry?
Sign in to Reply
WKetel
5/28/2011 8:32 PM EDT
First, for "student4ever", there is a method to determine the quantity of heat absorbed by an object. While it is almost direct, it is not trivial. The equation used is Q=MS(delta T). Where the evaluation gets complex is if the temperature rise in the mass M is not uniform, or if the specific heat of the item "S" is not uniform. At that point you get into quite a few rather ugly integrals that are a challenge to set up. The technique is indeed known as calorimetry, and I have seen it used to design a heatsink that had no spare capacity at all. It just exactly worked, keeping a part just below the upper limit for operation.
Addressing the need for circuit design, consider the case of some product that will never ever be produced in the HUGE quantities needed to make a custom IC economically sensible. This would include almost all of the manufacturing equipment in the whole world, and a fair amount of the more specialized scientific equipment, and probably a lot of other types of things that I can't call to mind right now. Two more catagories are commercial aircraft and military electronic devices, ( except ordinance). Not every product needs to be crammed in a package half the size of the previous model, with an additional 297 useless features added. Not all companies make cell phones, i-pods, or laptop computers. For many years I have designed custom circuits for a wide range of industrial test systems, and if we ever had to build 20 of them, that was a "huge" production run. So the truth is that circuit design is not dead, it is just a bit smaller now, but also more demanding.
Sign in to Reply
electronicsdemystified
6/1/2011 7:18 AM EDT
Growing up in the 80's and 90's I greatly admired my uncle for being able to take apart just about anything and make it work again. Early in high school I was subjected to 130+ dB's in the back seat of a car for the first time and immediately fell in love with subwoofers and audio amplifiers. I would take apart amplifiers and marvel at all the parts completely at a loss for the purpose of the various components. This motivated me to become an electrical engineer to find out how to build a bigger and better amplifier. Unfortunately discrete amplifier design was not offered as part of the curriculum where I went to school. Talking to my intro to EE professor he pointed out that audio design was figured out in the 70’s and only a small handful of people actually design audio amplifiers today. “However”, he went on to explain, “you can study for analog IC design and when you learn how to make an RF amplifier or integrated amplifier circuit the principles are largely the same”. Honestly I recoiled at the notion that I would not be taught how to make an amplifier at the time… A few years, hardware design jobs and a masters degree later I am a mixed signal IC design engineer. The shameful thing for me is that while I know how to make an audio amplifier, I have yet to actually sit down with the components that I have on hand and cobble one together… I’m hopeful that I’ll have the opportunity soon.
Sign in to Reply
Duane Benson
6/3/2011 4:57 PM EDT
From where I sit, here at Screaming Circuits, I'd have to say that circuit design is alive and thriving. We see an awful lot of incredibly complex designs, analog or digital. Boards with ten BGAs and 200 support components are not at all uncommon.
Sign in to Reply
GREAT-Terry
6/16/2011 3:21 AM EDT
integrated circuit just helps to build more complex function and this won't limit the circuit design. The key is the specification of the design. For a complex system specification, usually people start off with discrete circuit using op-amps (IC though), ADC (IC again!), reference etc. However, the circuit structure and system architecture evolves. SoC is just for a well know standard design that people have spent lots of energy to build it discretely.
Sign in to Reply