Explanation of iphone 4G's antenna problem?

[endervalentine]

First off, let's keep any fanboism or apple bashing out ... that's what off topic is for Second, I'm not sure if this is 'highly technical' but I'm hoping the answer would be ...

my background is MS in EE but it's mostly on device physics so I'm not familiar with all this RF stuff ... but curious as to what the problem is.

First off, very basic, how does the antenna work? Second, how does the antenna gain depend on the signal strength, and lastly can someone explain why there is a problem?

I understand that if you somehow bridge the two sides by holding it in your left hand, then you'll lose some signal strength ... is it because you're body is shorting part of the signal to ground by doing that?

TIA!

 

[silverpig]

Very basically I think it works like this:

An RF signal is an oscillation of the EM field which travels (in this case) through the air. The purpose of the antenna is to provide some conductor with some electrons which will "capture" this EM oscillation, making the electrons in the conductor oscillate. The better the antenna is designed, the larger the signal to noise ratio in the conductor. In this sense, the RF wave acts like a little signal generator or AC voltage source to provide some current.

I don't know if antenna GAIN depends on signal strength per se, as my understanding of gain is as a multiplicative effect.

Your guess as to why you lose signal strength is as good as mine - you short something out so now this voltage difference you are trying to measure is almost zero because you've grounded something.

 

[futuristicmonkey]

If you have any sort of conductor in an electric field you'll develop a potential difference between the ends of the wire (whether or not it is null because you have a loop or some other random orientation is beside the point here). Imagine a straight piece of wire, for clarity's sake. This forms a dipole antenna, the kind you're used to seeing on cars for the radio.

Now instead of having a static field, think about what would happen if the electric field were to oscillate. (Further, why would a dipole antenna be a half-wavelength long to be most efficient? Think sine waves.) This would induce a corresponding voltage (potential difference) between the ends of the wires.

Note that the electric field, magnetic field and propagation vectors are all mutually-orthogonal in the air. Without really getting into it, think of it this way, the electric field and the magnetic field both have to be there (its electromagnetic waves for a reason) so it doesn't make sense to ask which is there first, or which is causing the other.

In this specific case, a dipole antenna, the electric field is "doing the work". For loop antennas, the magnetic field vector needs to be orthogonal to the face of the loops of wire. These types of systems form air-cored, long-distance transformers. You might think that an air-cored transformer wouldn't be very useful (in power systems at least, it sure wouldn't) however the high frequency nature of radio frequencies (measured in MHz/GHz) allows them to work.

You are correct in your last comment about shorting out the system with your body. Depending on the frequencies, your body's loss-tangent (google it) may translate into a relatively high conductance. That's likely exactly what's happening.

The bit about an antenna's "gain" is misleading. The gain of an antenna is the ratio of the power it radiates divided by the amount of power it would radiate if it radiated at an isotropic level for the extent of its beamwidth. I know that last part's as clear as mud, but think of it thusly. Imagine a sphere of diameter 1 meter with a 1-Watt transmitter placed at the centre and the power through the sphere was the same everywhere. Now find the total power going through the surface of the sphere where the antenna would be radiating and that is the denominator. This is called intrinsic gain or dBi on datasheets.