Quote:
Originally Posted by kovidgoyal
@pricew: Actually, you don't have to shield planes from all devices or even test each plane with every device. What you need is to establish a set of standards about the electromagnetic radiation permissible from electronic devices. In the US, the FCC already does this. The problem is that the FCC standards are incompatible with similar (though more rigorous) standards setup by the FAA.
Basically, what's really needed is for the FCC and FAA to co-operate.
And you really don't need to worry about random circuit boards becoming antennae. It's very simple to shield circuitry from EM radiation. Just put it behind metal. The problem is antennae, which cannot be put behind metal for the obvious reason that they need to receive EM radiation to work.
And if I understand the problem, as described in the reports with the GPS systems correctly, the problem isn't that a cell phone signal can cause a momentary loss in GPS tracking. This would be harmless, since software can easily compensate for it. The problem is that the cell phone signals have been shown to actually cause the GPS system to lose satellite lock, which then takes a significant amount of time to re-establish. I've never seen that happen with car/hand held GPS systems.
And finally, you don't need to upgrade the whole fleet all at once. You need to upgrade it a few planes at a time and remove the restrictions only on the upgraded planes.
@HarryT: Actually if you read the reports TimS linked to, even a single cell phone can cause problems. Undoubtedly more phones will make it more likely that a problem occurs.
|
The problem, the FCC regulates and has tests for Class A and Class B devices. What they are concerned with is if you are leaking a signal powerful enough to interrupt critical systems outside of your home or business. They are not concerned if you interfere with your TV, as long as you don't interfere with the neighbors TV.
I could probably dig up the testing (have been peripherally involved in the past) for consumer devices, but the whole thing is a very complex set of government docs I don't want to fully understand.
What is true though, I have seen the EM reports on products, one connector (say USB for arguments sake) from manufacturer A will leak EM horribly, a connector that looks to the naked eye as identical from manufacturer B will not leak.
Other issues, say you have SPI eeprom on your board, your main system chip supports 2 chip selects, so the vendor routes to 2 spots. Early production loaded both of the pads (ie the traces are terminated with a part), later the manufacturer went to one part. Now, the board has stubs on the circuit board that depending on length, routing, etc can be an antenna.
Now, the vendor has been say using a 10MHz clock on the SPI device, everything works well, but boot is slower than they like, so they do a firmware update that fixes a few things, and boosts the SPI clock to say 50Mhz, boot gets faster, but you have just introduced a faster clock to your potential antenna, no telling if that causes a problem. How do you test, generally, that is only accessed at boot, but if you change a permanent setting, it will access it.
It is easy to go on and on about theoretical problems, the issue is, none of us know unless we are actively working on it. Some places may have ruled that the risk of issues is low enough to allow use, others have not.
In the US it is currently a felony to refuse to follow the instructions of a flight crew, are you willing to risk arrest to read your kindle? Please lobby the industry/government if you don't like the rule, but don't risk others because you think you understand the art of EM interference.
--Carl