maemo.org - Talk - Wanted: N900

maemo.org - Talk (https://talk.maemo.org/index.php)

- Buy & Sell (https://talk.maemo.org/forumdisplay.php?f=27)

- - Wanted: N900 (https://talk.maemo.org/showthread.php?t=90589)

Re: Wanted: N900

It would surely help, but then, having all antennas close to each other isn't very good, due to unpredictable results of so-called radiation coupling. It may, or may not be severe, depending on random conditions - hard/impossible to predict, so they're put on different sides of device - basically, as far from each other-, as possible/feasible - for a purpose. Note, that even N950 put them to both sides (thus complicating body design, due to two plastic parts above antenna, instead of one).

Coupling mentioned above would happen no matter how many antennas are actively used at the same time. But, there is other possible problem - active interference, when using radios simultaneously. Thankfully, we're rarely using both 3G and WiFi transmission at the same time (3G is closer frequency to WiFi, while GSM/2G is far away, much lower), but to minimize it, users would need to get habit of switching to 2G-only, while on WiFi. I do it intuitively for battery saving purposes, but for some it may be PITA to remember.

Now, it is possible that you may just scratch all of above, as you would never encounter any problems in real life. sadly, no other way to test, than doing prototype and test-driving it (and adjust in case of failure, then re-do prototype, then test drive again... and again...). Despite advancements on antenna design simulators, such complicated array of antennas and feeds so close together - with half of feeds hardware properties unknown/closed - is simply uncomputable.

That's why guys designing mobiles are doing it first, together with whole board project. We meet existing design, trying to put it in different "box", (and/or using antenna replacements) changing layout, etc - it's *much* more complicated than some would think.

Funny enough, the design that should work in *all* cases, would be to add stick-type antenna, like in those early days mobiles:

http://pesatalk.com/wp-content/uploa...e_timeline.png

Putting antenna outside "hull", far away from noises, interferences, heat, etc, clear 99,99% of problems.

/Estel

Re: Wanted: N900

Yes, those were the days, Motorola v60, Ericsson R320, Treo 650. With the antenna sticking out. Easy to locate the upside of the phone while taking the phone out of the pocket. No death grip :)

How about keeping the antennae at the same place and building the metal case around it?
The antennae are on the outside so it really should not matter right?

Re: Wanted: N900

Then, we're back at point one ;)

Quote:

Originally Posted by Estel (Post 1355674)

In case of GSM radio, it causes - sadly - quite a lot of directionality (for other types, like WiFi, too, but not as much, due to lower operating range as a whole, and higher frequency). This is, probably, reason why N950 sucks so much @ signal strength compared to N900 - even despite clever helical antennas design
(...)
This result in super signal when aiming device at "correct" angle, and 80% worse, when changing it just a 40 degrees (both numbers made up, to give and idea of what I'm talking about).

Consider, that iPad is designed to work with WiFi, not GSM/3G - especially the former will get affected.

Cover- which must/should be connected with N900 ground plane - will cancel signals when N900 would face base station by side. Of course, in real life, where signals bounce xyz times from everything, it won't cause 0 signal - it just be much worse than it should (aka N950).

Note, that N950 was designed from the beginning to be metallic, and they had *everything* they might need to make it work right - still, it fail to work as it should. Who knows, if it isn't main reason, why N9 is polycarbonate. (or any other funny-named thing, which means "plastic")

/Estel