I have a Titanium Powerbook (667 Mhz model) on which I'm running Debian Linux. Like TiBook owners everywhere I was disappointed with the limited wireless range I experienced. The page is about my attempts to improve the range.
Before reading futher be aware that I did not have much success. I created this page not as a howto but because, though I a number of success stories involving the simple fixes mentioned here, I did not read about the failures that I'm sure exist.
I did, however, eventually try a wireless pc card which worked much better. Comparisons can be seen at the bottom of this page.
For a while I had my AirPort base station attached to a hundred foot ethernet cable so I could drag it around if I actually wanted to use my TiBook in a different room. That's not exactly a commercial for the wireless lifestyle.
I eventually saw a little improvement by hanging my base station high up on a hallway wall. I could at least then get a signal in my bedroom and other nearby rooms without having to physically move the base station. But from 30, and sometimes even 20 feet away the signal could repeatedly drop off.
Then one day I stumbled onto a couple threads about this problem including a post mentioning a solution. After a call to Apple tech support, webdiva shared the following:
I was next instructed to eject the battery and look at the right side wall of the battery compartment where I was supposed to find an approximately 5 cm long plastic strip. (My particular powerbook doesn't have this visible -- instead, there is a plate with the serial number, etc...) Still, he told me to firmly press the side wall of the powerbook against the frame, just slowly and firmly pressing along its length several times for about 10 seconds. Next, he told me to replace the battery and start her back up.And suddenly range is great. You can read the threads here and here.
I was optimistic and I decided to try to quantify what sort of improvement could occur. I did a before and after comparison using wavemon. The results I saw were initially very impressive: my signal-to-noise ratio doubled.
Unfortunately, the improvement did not last. It did not take long for my signal to drop back to previous levels. Repeating the "magic pinch" did not seem to have an affect. The signal would also be completely lost at times, as it had before.
Next I tried reseting the PMU, another fix aparently proposed by Apple tech support. This had no noticeable affect.
Finally, I removed the back cover of my powerbook. The antenna did not appear out of place anywhere. I lifted up the optical drive and tried a little of the magic pressing on that side. I did some of the usual "don't know what else to do" fiddling, un- and reattaching the antenna from/to the airport card and pressing the battery side window again. The result of all these desperate attempts? A slight increase, but not as dramatic as the one first observed. And once again, the improvement did not last.
My results have not been entirely consistent and may even be the result of random fluctuations. But it does appear that, for my powerbook at least, none of these solutions really worked.
So far I've only tried simple, not too invasive fixes. The next step might be to try to attach a more effective antenna. But it might be simpler to just give up on the built in airport card, as many other TiBook owner's have, and buy an Orinoco or other card instead. Either way, I'm disappointed that Apple would release such an underperforming product.
For these tests my TiBook was sitting on a table about 30 feet (and in a different room) from my base station. From this location the signal strength was not only weak, it would also completely drop off at times and varied widely depending on where my arms were, if I was touching the laptop, etc. I took my measurements using an attached mouse so that I could keep a little distance between myself and the laptop. I always generated some network activity before taking each reading to make sure the reported bitrate remained constant. No appliances (microwaves, heaters, cordless phones, tvs, etc) were in use when I took the readings.
Here are the results (all values as reported by wavemon):
| initial readings | after pressing (battery side only) | three days later | after PMU reset | after pressing both sides | three days later | |
|---|---|---|---|---|---|---|
| signal level | -85 dBm | -78 dBm | -87 dBm | -89 dBm | -83 dBm | -89 dBm |
| noise level | -94 dBm | -98 dBm | -96 dBm | -96 dBm | -96 dBm | -95 dBm |
| signal-to-noise ratio | +9 dB | +20 dB | +9 dB | +7 dB | +13 dB | +6 dB |
| bitrate | 1 Mbit/s | 11 Mbit/s | 5.5 Mbit/s | 1 Mbit/s | 11 Mbit/s | 1 Mbit/s |
Many of these variations are minor enough that they may even be caused by randmon interference or whatnot. From this location (and similar locations) the signal often varies widely, maxing out at around +15 and ocassionally dropping to 0. The max value I get with the laptop right next to the base station is about +55.
Update, 30 March 2003
I finally borrowed a wireless card from another laptop to see what sort of performance I could get. It's a Netgear MA401RA card that works with the orinoco_ca driver under Linux. Here are the results (again, as reported by wavemon):
| original location | next to airport base station | |||
|---|---|---|---|---|
| internal airport | Netgear MA401 | internal airport | Netgear MA401 | |
| signal level | -89 dBm | -52 dBm | -36 dBm | -251 dBm |
| noise level | -96 dBm | -137 dBm | -96 dBm | -138 dBm |
| signal-to-noise ratio | +7 dB | +85 dB | +60 dB | +113 dB |
| bitrate | 11 Mbit/s | 11 Mbit/s | 11 Mbit/s | 11 Mbit/s |
As you can see, at the location of the original tests, the signal of the PCMCIA card was more than 10 times stronger than that of the internal airport card.
