I had to specially order the copy I have (23rd edition), but it is not so common on this side of the pond. The ARRL books are found here in libraries. For other electronic issues, it really depends upon what your level is right now, and if you want to learn or just jog your memory on the finer points.
For op-amp circuits, you could download some of the NS application notes for their ICs, they give all kinds of info, including circuit examples. As for transistors, if you don't already know enough for a basic test, you may as well forget trying to study up as you don't have that much time.
Personally, I cannot see you having time for any meaningful studying, you cannot cover such a large scope in such a short time. Do you know exactly what they do, anything about their products, etc? This is a good starting point, as it can tell you where to concentrate your efforts.
Unfortunately some radio receivers and transmitters appear as if they were designed by firmware or digital hardware guys :-(.
Knowing something about RF is a good when dealing with EMC and transmission line issues. For instance the firmware guy could frequency modulate a long external clock line with a dither signal and thus, keep the radiated noise density low enough, to hide it into the thermal noise.
You admit to knowing your own limitations? I find that is an admirable quality in a potential employee, business partner, supervisor, etc .... this point alone distiguishes you from many others. If I were your interviewer and heard only this, and I believed you could learn new things as required -- you'd be hired. I don't need to test your "basic" skills with cheezy questions about superhets.
IME, most new RF applications are heavy on software / firmware ... especially after the second IF (if not using direct conversion). In many applications, software radios are cheaper and perform better than their "analog" counterparts.
General RF constraints and properties in communications systems:
1) Link Margin - used to predict range and reliability given power, antenna and sensitivity constraints.
2) Dynamic range / linearity - the un-desired vs. desired signal levels at a reciever antenna may change by 1,000,000 to 1 or more. The amplifiers, downconverters and the rest of the linear system must accomodate this, or reliability and range will be compromised.
3) rho / quality factor - this is much faster to measure than BER. Rho is a composite of many factors such as EVM (error vector magnitude), Group delay, and intermodulation.
4) antennas - most have a resonant electrical length. Radiation resistance of the antenna is the real part of the load. This power is radiated. Any incident power not radiated is reflected back to the source, causing problems in some systems.
All this fuss because when you apply a signal on a wire, the same but less signal appears on a similar wire. Example: If you apply 1V RMS to a resonant 50 ohm 18" dipole at 320 MHz, you will get about 0.125 VRMS on a similar antenna placed 3 feet away and loaded with 50 ohms.
My grampa showed me this when I was 9 years old. Still haven't got a good grasp on it ;-)
as have I. I have done worse though - written out a couple of careful, well thought out answers to questions, then deleted them after realising that the ideas are good enough to make money from....I tend to think about that whenever post info about solving the practical problems that usually make the difference between success and failure of a circuit - after all, a fair chunk of my money comes from fixing things other people screw up, so educating others is not necessarily a good idea. OTOH there are plenty of idiots to provide me with work, many of whom are so dumb they wont understand things even when carefully explained, so what the hell.
A very good question. Since having one of the ideas about a week ago, I've clocked up around 25hrs chargeable time, so yes, I have made money from it, just not a fortune. This particular idea arose from thinking about how to test circuits in spice, and realising that expensive test gear, whose behaviour is easy to describe mathematically, can be incorporated into a spice simulation very easily.
Not a revolution at all, but a good enough thought at the time that I dropped what I was doing (writing an s.e.d. reply) , and started developing a few goodies, to help with a little 10kW smps project I am working on. Not only do I get paid for doing it, but it all helps me get closer to "right first time" designs, saving me time (in practice, allowing me to use that time to fix other problems ;) and my customers money (in practice, I'll keep the money ;)
The other idea - well, since having it, i've found myself a business partner, organised some finance and set up a company. Next step is to flesh out the product range, before separating the research and design elements, identifying critical paths and developing project plans (without, I might add, a single gantt, pert or MS project chart). Then, detailed costings, competitor analysis and then the real work begins. Worst case, we'll lose a little bit of money (perhaps $10k - its not a capital intensive product, and NRE is mostly IP, rather than, say, tooling). Best case, in 5 years time the company will be worth a few million. That, to me, seems like a risk worth taking, and is after all the reason I dont work full-time.
Other than that, I do have to agree that:
except, of course the "easy to come by" bit. Mediocre ideas are easy to come by, and bad ideas abound; good ideas are less common, and great ideas few and far between.
Actually clock signals might not be the biggest noise source. Many MCU got build-in crystal oscillator and the clock signals on those pins are quite near to sine wave (quite low harmonic and hence low interference). It is outputs like LCDs, LEDs, buzzer /speaker that generate most noise that reduce receiver sensitivity. When doing PCB layout, put those things away from your antenna, and you might need EMC suppressors (RC network, or filters) at the MCU outputs.
Joe
P.S. Lewin, good luck with your job interview! :)
It maybe too late now, but I like the book Electronic Communications Modulation and Transmission (2nd Edition) by Robert J. Schoenbeck ISBN 0-02-946552-4
Some examples are a bit outdated but it has got good illustrations on many circuits.
And what happens to the spectrum of those radiated signals when you frequency modulate the main clock or constantly vary the prescaler to these signals or add a random number of NOP instructions into the software main loop driving the LEDs etc :-).
You realize that the majority of college students today seem to view most of their entire college education likewise, correct? But this fits in well with the general trend of companies today that rattle of a litney of experience 'requirements' when very few are often truly needed (does the guy designing some GUI for your product REALLY need a BSCS? Does your administrative assistant really need a degree in anything at all?).
But I'm just bitter, I've never had particular difficulty getting jobs, it's just sad to see the general state of employment affairs in this country today. There are, of course, many notable exceptions -- truly outstanding companies that ask for what they really need and only hire those who possess such skills or can reasonably be expected to learn them on the job.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.