I Cant Sleep, So Heres My Idea (wifi)


Jan 30, 2006
Visit site
With the recent Status update for the Pandora, and the setbacks we have with the lcd's i started thinking about what this means for production of other parts of the pandora.

The wifi antennae is now being developed at the best they can without paying a professional to maximize wifi signal and its design. If you dont agree with what im saying, its probably because i am extremely simplistic and cautious at heart, but i was thinking about the support from the community behind this project. My thoughts as i tried to sleep then wondered onto the dev fund that was opened when pandoras were ordered, one thing led to another and i thought about a new fund that could be opened (if enough support was shown), since we may now have to the time to carefully piece everything together.

this fund would be to pay someone with extensive knowledge of wifi to maximize wifi design, i would think that there would be enough people willing to chip in the extra few dollars, euros and whatever else they may be paying in to get something that is sure to be a success. PLEASE, i mean no disrespect to the dev's working on the wifi, all intentions are just to help maximize efficiency.

The design may be far in enough now that it wouldnt be worth it, or this may be seen as a stupid idea, in either case feel free to close and delete this thread. My only real intention was to avoid one of my biggest pandora scares which is the first batch pandoras wifi signal is functional, but improved for the 2nd batch and my friends laugh at me for not having waited (scary thought). This one piece of hardware just seemed to stand out as a avoidable possible problem, falling into the better safe than sorry category.
Okay I'm not going to be very helpful. It's quite easily to build your own external wifi antenna especially if you have some quality shielded coaxial cables, I just soldered my own directly onto my router.

But you need something more sturdy like those little a stick, you would have to draw the current from the wifi section wherever it is inside the pandora and drill the case and put in the female 'sma' adaptor or something like that, which you just don't have space for since from what I know the Pandora will be completely packed. So I'm not really sure there is anyway to do this unless the designer specifically design the case for the purpose.

An easy way would simply buy an expensive usb wifi adaptor with an antenna, but you would need to port the driver for it. Personally I'm not worried about it since wifi signal are so abundance these day in the city.
This is interesting actually...

of course the antenna design inside the case is very important...most laptops have the antenna inside the plastic LCD enclosure so I guess that's a start...

...so I had a quick look around, and there is lots of information available about this on the net (it helps if you have access to ieee) A summary of my brief research:

...it appears that the best thing is to place an planar Inverted F Antenna (INF/IFA) as high up as possible inside the LCD case, mounted in the verticle direction (i.e. flat when the pandora is closed) with a large grounding plane around it and the antenna placed at the edge of this grounding plane. It is important to place the antenna as high up as possible for best reception...even if it is printed on a PCB it should definently be inside the LCD case. Exact positioning gets a bit complicated although there are some studies available. The type of casing used makes a difference, however I believe the pandora uses some sort of ABS which is the best I think (compared to carbon fibre or aluminium, I think they are the other common materials for laptop cases). Cable length is a serious issue as some estimates put losses in standard sized laptops (~15inches) due to long and thin cables from antenna to wireless chipset at around 3db which is massive, and high quality microwave cable is very expensive. The best thing to do is probably to use a USB wifi card and have the chipset inside the LCD enclosure as well, then just have USB signals going to the motherboard. This ensures virtually zero attenuation between antenna and chipset. Fortunately losses in the Pandora will be minimal due to its small size.

After that the precise size and shape of the antenna need to be designed, which is essentially dependent on the centre frequency and bandwidth you want resonance in. So this will depend on basically whether the antenna will be used for 802.11b, g or bluetooth or some combination, and can also be tuned to specific 802.11 channels. There are many studies also available on this.

Some useful references for above information and more detailed design information:

IFA basics: http://www.supernec.com/ifa.htm

Detailed study on positioning of IFAs in laptops, much useful info: http://www.research.ibm.com/journal/rd/472/liu.html

Another basic laptop antenna positon review, not so useful (no real conclusions but some results): http://skycross.com/technology/Whitepapers...r_Notebooks.pdf

Design of IFAs: http://focus.tij.co.jp/jp/lit/an/swru120b/swru120b.pdf

Design of PCB IFAs:

Hmmm...better get back to my real job now

Manufacturing Update
Apparently the last piece of the puzzle, the last item that needs to be finalized before we'll have a complete Pandora, is the wifi antenna. Good antenna design is the main distinguishing factor between good wifi reception and crappy reception. Having an antenna professionally designed would have cost something like $100K, so the design team had to go a different route. The testing has taken longer than expected, but the result will be excellent wifi range.

that was the source of the whole concern, an external addon wifi antennae is a great idea, but will be unnecessary if the signal is already good enough, time will tell.