Jim Williams

This is a different Jim Williams from the Audio Upgrades Jim Williams, right?

-- Les Cargill

sage

..

Jim has some video on the LTC website, also on youtube. I only met him once or twice. He was a friend of a friend. Of course he was a regular at the old Foothill, sometimes with Dobkin. [Geez o Pete, Dobkin could buy his gear new!]

You have to give Jim credit for working his way into MIT strictly on demonstrated ability. He didn't have a degree IIRC.

message

thur

e on

lia...

the

ho

ic

It depends what kind of engineering you are doing. I don't look at knowledge as being a bad thing. But look at it this way, do you really need to know Maxwell's equations in point form to design analog circuits? If you are designing chips, a little solid state physics never hurt anyone, but if you are buying components off the shelf, just knowing how input relates to output is fine. Even as a chip designer, you rarely get to design your own devices. Sure you size them, array them, etc, but you don't fiddle with implants often.

(...)

(...)

How much more valuable Jim would have been, had he gotten his degree, instead of teaching classes at MIT...? no, wait...

:)

--Winston

It depends. I use Maxwell's equations a fair amount--for my antenna work I had to write my own 3D EM simulator, because none of the ones I could buy had the optimizing features I needed. In optics you need to use the wave equation frequently, which is three lines of algebra from Maxwell's equations.

You probably also use Kirchhoff's laws, which are special cases of Maxwell, applicable in the limit of low frequency, high conductivity, and size small compared to a wavelength.

If you use transformers, somebody had to use Faraday's law to design them. Of course there are different levels that one can work at, but the best engineers I know can work at very many of those levels, from device physics to circuit design, to protocol stacks and FPGAs, to application programs. You're also much less likely to go stale or get obsolete that way.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics Electro-optics Photonics Analog Electronics

55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058

email: hobbs (atsign) electrooptical (period) net
http://electrooptical.net

Maxwell? Isn't that some sort of coffee or decaf?

Best one I ever heard about Kirchhoff's law during my consulting career but this was from a biochemist with some EE background:

"No, Sir, we can't do that because per Kirchhoff's law the sum of all currents into and out of this node must be zero" ... some silence ... "Dang! Isn't there a way to get around this darn Kirchhoff law?"

But today life is so much easier. I have become rusty in some of the highfalutin academic matters. But whenever I need some of it I go on the web and, tada, it's all there. The Internet is like a giant cheat sheet :-)

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

You can. Kirchhoff ignores displacement current and charge accumulation. Antennas and Van de Graaf generators wouldn't work if Kirchhoff were really true in all cases.

And if you use it that way, you get the same result. ;) Simulation is not an adequate substitute for thought.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics Electro-optics Photonics Analog Electronics

55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058

email: hobbs (atsign) electrooptical (period) net
http://electrooptical.net

You can go a long way with instinct and simulation.

John

[...]

This was a regular mesh network thing that ran with stimulus signals in the millihertz range. But I bet some investment banker could have derived a scheme to do it anyhow :-)

Actually I have to say that the web has been a real enabler for me. For example, years ago I knew nearly nothing about fiberoptics but a client needed to get things done. Etalon lockers in a noisy environment, femtometers, "impossible" laser diode bandwidths, auto-search, low noise signal tap-off, TECs, and so on. Normally I'd have said "Call Dr.Hobbs" but you were still at IBM. So I spent some non-billed time as sort of a continuing education, did the project, it worked with healthy margins, client happy. Even if I had access to a major university library I seriously doubt I'd ever gotten to that result in a reasonable time frame.

This non-billed ed time then paid a small unexpected dividend and it probably won't be the last. A new client called, IR comm project, rather adverse path conditions, didn't work right. So I made it work reliably for them.

App notes and articles by people like Jim Williams are perfect examples of how to foster education via the web.

Simulation is similar. Yes, it can't replace thought. But it sure has enabled a ton of new opportunities. Starting tomorrow I'll have to design a filter that has to work with some extreme, really odd and not at all constant terminations. When dealing with several variables in that sort of game there's only so much you can do with a Smith chart or it'll literally disintegrate on you. The computer, however, can crunch through millions of combinations in a matter of hours.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

Yep. Instinct is what is seriously missing in many younger lads and lasses in engineering these days.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

I think it would be wise if simulation was not taught in the universities. Force everyone to learn how to design by instinct and pencil and paper first. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

Instinct can really only come by way of a soldering iron. And that's where we have a problem.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

Not, "...let me go pull my copy of Hobbs' book off the shelf and read up a little?" :-)

I agree that the web is fantastic for getting information on pretty much any topic you want, although typically the information isn't in as much depth as a good book is... and when it is, it's seemingly often in little isolated "island," as opposed to one cohesive tome.

But finding some good PDF files can be a gold mine, of course.

---Joel

I'm not sure. I built a lot of stuff from schematics when I was a kid, only vaguely understanding most of it.

I then grunted thru some math on my own, but MIT fully honed it.

I never simulated a circuit until I was 37 years old... long after I had created maybe 30+ ASIC's that worked. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

To be honest I didn't even know about Phil's book at the beginning. But the millisecond I found out I told my client to get it, and they did.

With fiberoptics I found such a large volume of top notch R&D papers that it gave me most of the ideas I needed. After that it was mostly grunt work. Then later we found lots of unresearched stuff where we had to continue on our own. Kind of like the sign "Pavement ends in 1/2mi" on a steep mountain road where you shift to 4WD and the knuckles turn white :-)

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

:-)

That's because you didn't do ham radio? At a few hundred watts of RF power stuff that wasn't fully understood tends to makes itself understood with a loud *KAPOOF* :-)

I went pretty much without sims the first 4-5 years of my career. Then I got Microsim PSpice and another world opened up. And then they were bought :-(

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

Sure. From age 22-24 I was building and running my own (transistor and varactor) 2 meter rigs.

I didn't have too many *KAPOOF* in my repertoire... maybe I'm both better at math and more careful ;-)

Still exists if you'd follow my instruction.

What do you think I'm running? ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

I mean _real_ ham radio, shortwave, and lots of power.

Nah, simple: In Germany they had this rule that the highest license class can use up to 150W rated plate dissipation. Someone (call sign DL1AH) found out that there is a deflection tube for color TVs (PL509) that is rated at 30W but can really take several times that. Probably so the sets would survive a 10-day soccer world cup marathon session at

100% duty cycle. So ... everyone used five of these. I got a sustained 1200W of RF power out of my amp, 100% legit. So I figured if I made the balun with one Amidon T200-2 core it's gonna be fine even with a slight mismatch. I didn't have the dough for two of these cores and knew the calculations were dang close to kabloiue. One fine day a muffled boom and ... THWACK ... Antenna was laying on the ground.

Then after a 470uF/350V electrolytic in there turned into a spacecraft and passed my right eye at 2" distance upon launch I learned that a capacitor's ESR does have to be reckoned with, and that if a datasheet doesn't contain an entry for ESR this doesn't mean it's zero.

Another thing I learned was that even the big ceramic caps have their limits, after one exploded and the whitish ceramic had turned into bubbly green glass.

Well, I still have it, including the nice cloth-covered binders. But after LTSpice came out I became spoiled by its fancy GUI and all that.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

At one point there I was running a pair of 4CX250's into a "hairpin" made from copper pipe, driving a multi-wavelength V-beam... toasted TV sets for a block away ;-)

Huh? PSpice certainly has a more user-friendly GUI (MicroSim front-end, NOT Crapture). ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.