The book is good, but I wonder the eagerness of the magnetics people to wade in the swamp of the CGS unit system. Half a century ago, I nearly lost my temper with Kittel's solid state physics book due to CGS.
-- -TV
There's a nice pdf version of Snelling's book here.
George H.
I'm a big fan of Gaussian units (rationalized CGS electrostatic units)--I use them in all my longer EM calculations. That gets rid of all the mu-zeros and epsilon-noughts. The units of current and voltage are a bit odd, but a gauss is a much more reasonably-sized unit than a tesla, and a centimetre is a better sized unit for capacitance than a farad (it's 1.15 pF). (Besides, Gauss was way smarter than Tesla.) ;)
Permeabilities and dielectric constants are dimensionless numbers like 1 instead of 1E-6 or 8e-12, and capacitance comes out in picofarads automatically. (Well, with that 15% adjustment it does, anyway.)
In doing calculations it saves a lot of blunders IME, and it's easy to convert it to MKS at the end when you want to plug in numbers. (One of these days I should build a statvolt-seconds per centimetre - statammeter.)
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
I can imagine that when you grew up with a units system you get proficient in it and have found practical ways of using it. I've been very happy with epsilon and mu 'noughts' in that those are the values in vacuum.
How a centimetre can ever be a useful value of capacitance is totally beyond me and my comprehension, and no, I don't even want to know. :)
On the other hand, after a S.I. life I've been introduced to the use of inches, ft & etc. and must say, those measurements _are_ very useful in daily rough construction work with wood and GI sheets and pipes and angle bars and the likes. My hand spread out is 8 inch, half of that is
4, and even 6" is easy to estimate and even your feet make sense. :) As well as is 1" pipe and its fractions thereof, 'schedule 20' & etc.I think the practical value of inches is that they are big enough to distinguish between its multiples, for instance it's easy to see the difference between a 3" and 4" tube, whereas with centimeters one doesn't easily know the 2.5 from the 3.
For the rest, fathom, Btu, acres, bushels, cubic weight etc. have no value for me. They only give me a headache.
joe
In Gaussian units both are 1.0.
You don't like picofarads?
Gaussian units are just as "metric" as SI. When you're doing theory (as opposed to plugging in numbers) simpler is always better, because it saves blunders. For instance, I grew up using exp(i omega t) as the Fourier kernel, but that leads to all sorts of obnoxious factors of 2 pi cropping up in the normalizations. In grad school I switched to using exp(i 2 pi f t) and they all went away.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
Well, when your shoes are six feet under the surface of water, you then may FATHOM the extent of your problem.
Yes, that's very convenient in calculations but it makes you almost forget that there _is_ a dielectric constant or magnetic permeability in vacuum.
I have no problem with pF, it's just a factor 10^-12, no worries, but I don't understand the connection with centimetres. :)
I liked it very much if I could check my theoretical work with a dimensional analysis. If I were formulating a formula for the distance from here to the moon I'd be quite confused if upon dimensional analysis the result was an electrical capacitance. :)
Yes, it's all a matter of what one grew up with and therefore feels comfortable with. In our math and physics classes it was i and omega, in electrical classes the i changed into j (in order not to get confused with current, I guess) and 2 pi f. I must say that if the math got heavier I really preferred to work with the i and omega, and to convert the result back to 2 pi f and j.
joe
Completely true :)
joe
But there isn't! It's just a normalization factor. In Gaussian units there's no difference between a gauss and an oersted, which sticks much closer to the physics--H is just a mathematical construct and doesn't actually exist.
If you don't insist on importing all sorts of redundant units, it makes perfect sense. A 1-cm radius sphere all by itself in vacuo has a capacitance of 1 cm. It also makes it physically obvious that capacitance scales like length.
Dimensional analysis is unaffected. (The particle physics convention that c=1 does make DA harder--when you're done, you make the results come out in the right units by multiplying by powers of c, so it's no longer an independent check.
Everybody likes positive frequencies and waves moving in the positive direction, but you can't have both. Physicists are usually more interested in spatial behaviour, and EEs in temporal, so they use opposite conventions:
Physics Plane wave: exp(i(k x - omega t)) FT kernel: exp(i(omega t - k x))
EE Plane wave: exp(j(omega t - k x)) FT kernel: exp(j(k x - omega t))
I actually use both conventions. Using i for physics and j for EE makes it easy to keep the sign conventions straight.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
On 2016-01-06 14:09, Joe Hey wrote: [...]
Epsilon-0 has units of F/m. If you make it unit-less, the dimension of capacitance ends up being a length.
Look at the expression for a flat plate capacitor: C = E0 * A/d, if E0 is dimensionless and unity, C=A/d, which has the dimension of length.
Jeroen Belleman
Yes, that's very clear. Thanks for pointing out something that I didn't even _want_ to know. :)
joe
}snip{
Hah, that is funny.
For the isolated sphere, yes. For a double plate you'd have to scale in such a way that sqrt(A)/d is constant. Dimensionally, A/d scales like length indeed.
}snip{
}snip{
I haven't been involved with antenna's or transmission lines, so after the electromagnetic waves course given by the physics department (k x - omega t, if I even remember this correctly) I haven't been exposed to the opposite direction.
joe
The Thompson-Lampard calculable capacitor
has a capacitance that depends only on it's length (typically measured with a laser interferometer)
-- Bill Sloman, Sydney
Dumb question: Do you have high vacuum capability in your lab?
E and H are in phase in a wave propagating in a lossless medium.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
Nope. Not enough space, for one thing. I'm failing the cat test pretty badly at the moment. I'd like to get a smallish nitrogen dewar one of these times, but I haven't been able to justify it yet. Maybe when I get rid of that pile of customers' instruments in my stockroom....
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
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.