Electrical Engineering Texts?

It still wasn't much good despite the hype. It was written by Physicists and not engineers. There are far much more engineering analogue texts that are better. For beginners there is the Malvino book for example.

Hardy

Some physicists are rumoured to know a little circuit design. ;) And Win Hill isn't a physicist AFAIK.

Another Hardy mystery, I suppose...where's Franklin W. Dixon when you need him?

Cheers

Phil Hobbs

Doesn't mean the old books are necessarily good however, Modern graphics and teaching methods surpass some of the old. For example, look at old books on electrical machines? Hardy to read and the diagrams are crap. Now we have on-line flash videos of rotating magnetic fields etc. The second problem is that there is little point in spending a whole semester on bipolar transistors when they are rarely used in the majority of todays products. Analogue MOS is the only way... Maths texts are just about the same and control books. Old DSP books have little Matlab code for students to experiment with. Now get with it!

Hardy

No comparison. If you do a B.E in electronic eng vs a Physics degree at a good university like UCLA. The Phsicists tend to do work on devices foremost rather than circuit theory. yes there are exceptions of course but most of em aren't much use as you can see from this ng.

Hardy

Whistling in the dark, there? You certainly don't appear to have worked with many decent experimental physicists.

Of course, for an engineer the basic professional qualification is a BSEE, whereas for a physicist it's a doctorate, so most folks with only a bachelor's degree in physics either (a) wanted to be schoolteachers, or (b) washed out of an advanced degree program.

Cheers

Phil Hobbs

Dear oh dear, Hardy's pushed the Troll-O-Meter past half scale. Can large-scale plonking be far away?

At least the usual suspects can be amusing and sometimes try to substantiate what they say...but not Hardly.

Cheers

Phil Hobbs

You have a good point, but even when I was an undergraduate a couple decades back, there was a significant chunk of time spent on BJTs, and then you switched over to FETs but it went rather more quickly: The vast bulk of what you learned with BJTs applies to FETs as well, you know?

Since models initially used in schools don't tend to include various parasitics, I think one could make a case for teaching with the "inferior" BJT first (since you're forced to deal with the non-zero base current and impedance) before moving to the "superior" MOSFET... since then it's nowhere near as much of a shock when, oh, hey, at high frequencies your MOSFET's input doesn't look at all like an "open" either and there is plenty of feedback from output to input to help turn your amplifier into an oscillator.

Historically the idea was that you had a separate lab class and that's where your Matlab code would be; the shift to combining "theory and practice" into the same book is largely due to the shift in four-year institutes shifting their curriculums somewhat away from cranking our "research scientitsts" and more towards "practicing engineers." Or something like that...

---Joel

Horowitz is the Physicist... teaching circuit 'design' to physics majors at Harvard :-) As far as I can ascertain, Win Hill has no degree.

...Jim Thompson

Just try and make a bandgap on a CMOS process without using the parasitic bipolar devices :-)

...Jim Thompson

Most electronic circuit designers don't need to understand exactly how transistors work; we only need to understand their behavior. Only device designers need to spend entire semisters on semiconductor theory. Most of the kids nowadays seen to do some CMOS device design in school, even though few will ever use it.

Heck, I know for a fact that I understand the behavior of some semiconductors better than the people who designed them!

John

Here's a book by a physicist that has a bunch of electronics inside.

Show us your book.

John

IBM could have used the 68K. Rumor I heard was that IBM felt they'd have more control over a little company like Intel, than over a big one like Motorola. Tragic.

John

Until you want to do something unusual, at which point knowing the physical origin of some of those effects can help a lot. For instance, photodiodes speed up a whole lot when you get within a few volts of their breakdown voltage--most of the series resistance comes from the undepleted carriers, so going from mostly-depleted to fully-depleted makes a huge difference to the speed.

Your Grekhov diode circuits are another example, as are my laser noise cancellers, JT's current mirrors, and so forth and so on.

The OP seemed to want to broaden his expertise--if HS is happy pushing buttons, that's okay with me!

Cheers

Phil Hobbs

Similar thinking by IBM put Bill Gates's DOS in the first PC ... they would have preferred CP/M from Gary Kildall of Digital Research, but they didn't like his price.

-- Bill Sloman, Nijmegen

Kildall would not talk to them. There *was* no price.

-- Les Cargill

According to "Fire in the Valley", IBM wanted to buy CP/M outright for $250k, whereas Kildall wanted $10 per copy.

Cheers

Phil Hobbs

On Jan 29, 10:31=A0am, Jim Thompson wrote:

ades

what

Didn't say you didn't need them at all. It's just the proportion. Older books spend most of the time on them. You can make an argument of saying 'well it doesn't matter, just do the basics and then pick up all the other stuff later in industry". Doesn't really wash, you get people (like many on this ng) complaining about what little students know if you do that. Some of the old books on more mathematical subjects are great however, they rarely have to be changed. I do have a gripe about old machines books though. Some of them are totally unreadable. Trouble with books for students is that some of the people who write them forget that they have years of experience and the student doesn't! They need all the help they can get and something that looks ridiculous and easy to an expert may take some time for a novice to grasp.

The Horiwitz and Hill book is one such example. It's not really a student text. Many teachers recommend it just to conform. A far far better book is Malvino's book

Not sure if he has updated it but it was the best for bipolar transistor introductions and explained all the steps.

DSP books are another area where there are many books which pretend to be for students but are far better when you actually know the subject and just want more detail on certain areas that you have forgotten! Papoulis is a good example. brilliant book but not for a beginner. Rick Lyons book is pretty good however but not used so much in academic circles unfortunately. There is a certain oneness about control textbooks, most are very similar in fact and probably do the job. They do spend a lot of time on things which are of only academic interest and of little engineering design use. For example Root Locus! I love root locus but nobody does serious servo-design with it! Same for Nyquist plots and Nichols charts. Then so much time is spent on optimal control without saying that most of the approaches are little use as they don't have any integral action in them! (there are some books which include this though). The very old books on 'sampled data" systems used to go on and on about zero-order holds and how to model them. This was very important in the early days when the sample rate was so low that the inherent phase delay was significant. With higher sample rates this is much less of a concern however.(though still important of course) Then there is coms, a subject that is moving so fast it is hard to keep up. Wireless sensors,essentially a new subject. Anyway, to say that a Physcist can come even remotely close to an engineering grad is just silly. If I want to know about basic energy equations,matter or black holes, maybe even device Physics I would go ask a Physicis, otherwise I would steer well clear.

Hardy

Why? Just use half of the data bus for periphs. Most of the i/o boards on the original PC were MSI logic anyhow.

I'm sure Moto would have given them a good price on 68Ks

John

Interesting. So Cringeley is just flat wrong? My source was "Triumph of the Nerds". Maybe the "wouldn't talk to them" came after.

-- Les Cargill