WiFi AP Gets Antenna Augmentation

Feeling bad that his access point was being made fun of by models with beefier external antennas, [Customer Service] decided to do something about it. After cracking open the Asus wl-330ge he found it would be quite easy to add a connector. This access point has two internal antennas that are quite small and use a spring connection to the signal and ground pads on the PCB. Those pads are fairly large and separated, making it easy to solder the connections. Scavenging an antenna connector from an older device, [Customer Service] soldered it in place and drilled a mounting hole in the plastic case. After flashing DD-WRT firmware he’s now got everything he wants from the little guy.

29 thoughts on “WiFi AP Gets Antenna Augmentation

  1. @goldscott and Face

    Mike didn’t mention it but, before the mod [Larry] only picked up about 3 hotspots in his neighborhood, and with the antenna he gets 12. While its not a /real/ measure of strength. It does go to show there is significant improvement with just a 2Dbi antenna.

  2. Nice!

    I have a Mini PCI E wireless card I plan on dropping into an MSI Wind Barebones Desktop using a pair of antennas intended for a laptop. (heck, it’s netbook hardware as it is! ;) )
    I plan on securing them discretely somewhere.
    I’d much rather have big antennas though. I’ll have to see what I can do about finding a pair used somewhere.

    Good stuff!

  3. I recently purchased an Asus wireless access point/router from eBay for the purpose of using it for mobility. I wanted something that I could use while in a car, boat, rv, camping, etc. Due to it’s size. (its smaller then a deck of cards) I could use it about anywhere but first I had to make some changes.

    For starters I flashed the Asus wl-330ge using the open source (linux based) firware called dd-wrt so it would be alot more powerful and be able to be used as an access point, client bridge, client, repeater, repeater bridge, etc.

    Once this was complete I was so impressed with this little wifi router that I decided to take it one step further. I wanted to mod it to use an external antenna.

    Once opening it up out of the plastic case I got a couple of suprises. First, I found that not only did it have 2 internal antennas but it also had another connection near antenna 1 that could also be used to attach an external antenna (like the wires used on laptop wifi cards built in) so a possibility of 3 antennas in total.

    I took an old wifi adapter and de-soldered the external wire and connector and soldered it to the number 2 antenna of the asus pcb. In the pictures I have taken of the unit modified you can see that I attached (soldered) 2 external antenna connectors to it however in the end I only used just one external antenna as I did not notice alot of difference using 2 of them.

    Just using the 2 internal antennas I was only able to see 2 access points near my home. With the one new external antenna mod I am now able to see and connect to over 12! (this using only a standard 2 Dbi antenna) When you add this to the fact that I can now use the advanced control panel provided by the dd-wrt firware, you can see what a great addition this would be for anyone that travels.

    The Asus wl-330ge requires only 5volts DC @ 1 amp to run which also makes it perfect for use with a small solar panel.

    Here are pics I did during the modding. I hope this helps some of you that have this tiny little wireless access point and are considering modding it.


    Larry Lawrence

  4. There is a big difference between *seeing* more access points, and actually being able to use them. He speaks of all this theoretical extra range that this antenna *could* provide, but I see absolutely NO data supporting this theory. I wouldn’t even call it a theory, I’d call it a poor assumption.

    Also, how does that obscenely horrible solder job affect the impedance of the connection?

    How about you guys actually look at something before you call it a ‘hack’ and post it…?

  5. I dont think its the same connector as used on laptops. I had one similar on an old cardbus adapter that nothing would fit into. It was under the cover so not made for end user use.

    After asking around a lot it turns out it was some test connector that was not made for constant plugging and unplugging. I just hardwired in a cable where it was connected to.

  6. @Jake How about instead of complaining and trolling you support the fact that; indeed by seeing more connections, this would mean that the device is receiving more than it did previously, also the soldering job was even called horrible by the modder (Can’t you read?) Go troll somewhere else, or buy a bike and learn to ride it. It’s fun, easy and you don’t even have to compile. People like you, are the reason no one wants to post anything worth a crap. Your negative criticism kills communities like this.

  7. I wonder, does the router automatically determine which antennas are hooked up and so not transmit through the other pads? It might be nice to solder a chip antenna onto one of the other antenna pads. If it is smart enough to fall back to that and not burn out the finals on the main antenna connection you could unscrew your external antenna when you want it to be small and don’t need range.

  8. To improve it I would desolder what is attached now and clean up the connections. From the way it looks now there is too much of the center wire exposed. You are losing about half or more of your output power right there at the solder joint. The more wire you have exposed the more you are losing. In essence that bit of wire from the solder joint to where it enters the wire shield is your new antenna.

  9. @Michael Wheeler

    Yeah, it would be much cleaner and probably work better if he used an adapter. Pasternack has a UMCX to RPSMA bulkhead cable assembly for around $30, which is compatible with UFL.

  10. I modded a Linksys wusb11 with no connector at all. My main need is access to the internet and allow positioning away from the PC, in window. The neighborhood is generous with open access points. This older model is 2.4 GHz only and has only one antenna to deal with. Other units TX on one RX on other, or use 2.4 and 5.6 GHz dual band. All of those newer models complicate hooking one high gain directional antenna. The wire came form one of those wind box mini’s, antenna included, wifi card extra. I could have used the original, but test strips and tinnings later I went with the freebie wire. More than a foot but I only used about quarter wave, 1 inch plus directly soldered,tiny size and length! The usb unit is soldered with short ground wires ,IE bug legs to the out side of the cookie tin lid. From tiny pads thru cookie tin lid, to a double diamond bent copper wire. Works great, aim-able. About a 3 fold increase in spots fair or good. It can suffer from rained out access though. Can’t complain though.

  11. I’ve been doing this since 2005 to low cost routers and routers with fixed or internal antennas, nothing new.

    I’ve even added external antennas & connectors to pcmcia cards.

    I have some pictures of this at my not-so-updated website. (www.patulawifi.com)

  12. @T-Rev

    Your assumption that “seeing” more access points = Greater usable range of the adapter (as a fact, lol) is a laughable assumption, at best.

    People like me? People like me want to see things posted that *are* worth half a crap, or more. Try teaching/TA’ing in an electrical engineering program. You’ll quickly see that a good percentage of the kids in the program are there not to become engineers, but simply to acquire a degree that says “engineer” on it. Instead of encouraging kids to build a “hack” full of 100% assumptions and 0% theory/data, why don’t we actually help them learn something? THAT is the proper way to teach a kid to better themselves as an electronic hobbyist or engineer. If you aren’t willing to do it right, then don’t do it at all, or at least don’t tell people about your abomination.

    I just feel like the point of this site is solely to get traffic, and to do that, they often link to projects which are full of assumptions and bad (or no data). Oh, and then they link to things like the mounting bracket for a cheesy power drill motor, and call it… a hack!


  13. @Jake you are an idiot who doesn’t recognise data when he sees it. The only way you would suddenly recognise a larger number of accesspoint is if more of them are visible to the receiver. All else being equal that would equate to a larger signal to noise ratio at the receiver. Given the radiating patterns of some internal antennae an improvement would also be easily backed up by RF theory.

    No one claimed that this is how people should do things. This is a hack. Be happy that he used a soldering iron and not superglue or it wouldn’t be called a hack. Yeah if he didn’t solder and probably used that connector instead he may even pick up more access points, but I think it’s quite clear that many people here know that.

    Since you’re so hell bent on teaching kids what have we learnt? We now know that unless you have a $200000 network analyser no hack involving an antenna is worth doing. …. Ok no what have we really learnt? That with $5 of parts and a bit of playing around we can suddenly see access points we weren’t able to see previously, we’ve learnt that external antennas may have a better radiating pattern in some circumstances, and we have learnt, completely backed up by facts that you sir are a tosser.

    On behalf of all of us here who enjoy $5 hacks just for shits and giggles, piss off.

  14. @Jake

    Your theories and data that you crave seeing are what makes electronics (such as a dongle USB device or in your case this router) “holy-as-Swiss-cheese.”

    (The reason we’re able to hack them in the first place)
    Just to prove my point Read below, not that it’s relevant.

    You trust data right?

    So does Microsoft.

    (Look at how DATA told M$ that the newer versions of Windows were not crackable!)

    (Within 15 minutes I bypassed WPA in Windows Seven running Ring0)

    People can have all the “solid proof-data” in the world and still be off 150%

    I do not have a degree, and that being said I have been reverse engineering commercial protections in software such as Armadillo/ASProtect/VMProtect since 6th grade.

    I’ve been cracking Dongle protection(s) since the age of 14. I could link you to a few reads on HASP and some papers I wrote on various different dongles.

    (People I know who are good in there respective area never took a class on anything, like myself.)

    Just as a side note: If I was in a classroom with your arrogant ass, I’d probably stab you in the eye with a soldering iron.

    Everyone in your class is probably just getting the degree and not learning a damn thing because they get discouraged when you barrage them with an endless amount of negative criticism.
    (I’d be in a hurry to get the hell out too.)

    Your previous statements above pretty much cement all of our “theories” on that. No need for more “data” there.

    By the way!
    How is learning to ride that bike going? Or do you need some solid “proof” on that before you do that to? ;D

    For an example on solid proof for the fail see this:


    ^ Windows Fail.

  15. @T-Rev

    Once again, you claim that *seeing* more access points means that your usable range has increased, and that’s just not the case unless you can provide real data.

    It sounds like you need to get out in the real world and see how things work. If you walk in to a meeting, and say “YES, IT WORKS! I COULD SEE MORE STUFF WITH IT!”, and have absolutely no data to back up this wild claim, you’ll be out of a job. You can’t *assume* that just because the device can *see* more access points, it can also *connect to* and *reliably communicate* with them. It sounds like you are smart enough to understand things like SNR, BER, MER, etc, so I suggest that you admit the fact that there is absolutely no data here that shows the device’s usable range has been increased. That is my only point, and it’s blatantly obvious that I am right. If that’s arrogance, then damn, the engineering world is full of a bunch of reeeeeeallly arrogant people! They *LOVE* data!

    Also, yes, yes, you and every other 14 year old are making keygens and cracks, big deal. Would you like a cookie? Do you want a shirt that says “I’m a HACKER!”? Lol. Good for you, kid. Now go do something meaningful with those skills.

  16. I gave this mod a go but didn’t really notice any improvement. I crimped a short piece of coax cable onto an RP-SMA male connector so that I could swap out the antenna. Unlike the mod in the photos, I cut the core shorter and extended the ground to the GND contact point, with the hope that this would provide better shielding. I don’t know much about radio-electronics, and I wondered afterwards if this made sense. I also though about tying the ground and core contacts together, but again I’m not sure if this makes any sense.

    I used a multimeter to check DC resistance on the contacts and everything was fine, but maybe I messed up the AC impedance somehow?

    I also tried to figure out what the CON1 connector is, and I’d say it’s not a U-FL connector as some had suggested. You can see in the photos that it has a conical shaped socket and no male pin in the centre like a U-FL socket would.

    Anyone have any ideas how I could improve the performance? Anyone else try this and have any luck?

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.