Need Wireless Router Antenna Mod Help

[lucky777]

I think I might've royally screwed my router up. Here's what I was trying to do:
http://www.instructables.com/id/STN4BF1F6S8S10K/

Here's the results:

I didn't get the results I was expecting. The router dropped 2 bars from stock. I've already trimmed off all of the stock wires to add the copper wire (circled in black).
Is there anything I could do to make the signal better?
Should I remove the resistor? (circled in red)
What's that metallic thing for and should I remove it? (circled in blue).

Notes:
The copper wire is from a stripped coaxial cable, should I be using something different?
Is there anything I need to insulate? I already stripped off all the stock wire insulation, is there a substitute insulation I could use (would aluminum foil work)?

The router is a AT&T 6800G.

 

[lucky777]

Bump.

 

[Enigma]

not sure if this is the issue or not just looking at your picture none of the loops in the coil are touching right? Another thing you might try is re winding the coil and checking the length of the wire..... Yet another thing are you sure that the swr of that modded ant. matches the radio that you have it attached too? these are some things that might help.

Edit: Another Question is this a 5ghz radio or a 2.4 ghz radio........ ant design for a 5ghz is different if this is the case.... One suggestion that might help is try making a signal amplifier out of tin foil or a metal dish this would help increase the signal from the ant. granted it would turn the omni ant. into a directional ant and you would have to point the "dish" where you need the signal to go.

 

[TJohnny]

The antenna I see in the picture cannot work, it is too short. Also it looks pretty bad, you should mind geometry more, microwaves are very sensitive to it. The router you are trying to destroy is an 802.11 b/g band (2.4GHz) and is compatible with the antenna design I saw following your link, even if I should suggest you not to follow an antenna construction plan so approximate next time. Ther is not a single performance data and if that thing I saw could work or not we will probably never know.

If you are to experiment with 802.11 antennas you need instruments and deontology. First rule is to remove as little as possible from the router (all of its parts cost and are there not without a reason) preferably not even the antenna. The reflectors Enigma is suggesting you can reach astonishing results wth unmodified antennas. I achieve a 17dB signal-to-noise-ratio increase over basic performance (from -72dB to -55dB) by putting a metal sigar box 1.6 cm behind the double antenna of my laptop (pictures to come, my camera was recently stolen @$#!). That sigar box allows me to reach a network with a speedtest country rank of "faster than 96% of connections" which is not too bad considering it cost nothing and not even damaging the laptop.

Now for the instruments. Network Stumbler is a an old program with a lot of defects, but also one you cannot do without. It shows the inensity of the wireless signal with a resolution of 1 sec/1 dB.

Unfortunately there is much more to know about a .11 connection than the raw intensity of the signal, but strangely enough there are not many programs monitoring, for example, the ratio between good and bad packets received and sent, a parameter we can see as the quality of the up/down link. I am sure it is my fault but I could not find a single windows freeware monitor like this. I had to write it myself, a horrible piece of software put together on a boring sunday but if you need it try and pm me (no guarantees I will answer at all). If you are a Delphi programmer convince me you will not just steal my code and I will send you the full source.

Best thing you can do now is remove the bad antenna, solder a coax cable to it and terminate it with a connector. Then build your new improved antenna and plug it in. This way you do not need to damage your router more. Enjoy!

TJ