The great antenna test thread

Since 4.1 has given us back the ability to use Field Test Mode, we can all check the numbers for ourselves. Up until now, I've only seen two sites that were able to do testing with these kinds of results instead of just the 0-5 bars. That isn't much. One reported up to a 24 decibel reduction in signal and the other said 20.

To activate field test mode, on your phone keypad enter:

3001#12345# and press the call button. You'll get an almost blank screen - it has a refresh button which doesn't seem to do anything. The numbers seem to change without doing anything with that button. In a few seconds, your antenna signal bars should turn into negative numbers (bigger negative numbers are worse signal). Here are how the numbers correspond to the bars on the iphones since they adjusted the formula in 4.0.1:

-51 to -76 = 5 bars
-76 to -87 = 4 bars
-87 to -98 = 3 bars
-98 to -107 = 2 bars
-107 to -121 = 1 bar

I believe only the iphone 4 can go that low and still make calls. Older phones can go to -113.

I do not know how quickly the numbers change when the signal changes. I know the bars have a built-in delay so this may as well. Some testing should clarify this.

Now maybe we can get a lot of tests, like this:

1. No case, iphone 4 setting on a table or desk to minimize interference. Write down the number.

2. Pick up the phone and hold it "wrong" (cover both sides of the plastic separator on the left side). Write down the number you get.

3. Put on the case and repeat the above tests.

Keep in mind that the numbers can suddenly change anyway. I don't know all the situations, but, for example, your phone could switch to a different tower. I also read someone say that the kind of network AT&T uses will have sudden fluctuations. I know I've seen the bars change without me doing anything with the phone. Using the field test mode today, I've seen the numbers suddenly jump 10 or more, though most of the time it tends to stay about the same.

So, when testing, multiple tests would be wise to separate the odd ones from what is normal. If you run, say, 5 tests like the first two on the list and usually lose 20 db, but one time you only lose 10 or lose 30 instead, that is probably a change that is in the network and should be discarded.

I'd like to see a lot of people testing to give a better idea of how much of a problem this is and how effective different cases are.