Is there a way to ID where the antenna is on a usb wifi adapter? I apparently misunderstood a web page I found, and in these pictures you can see that I’ve soldered my antenna to probably just some housing. But where is the antenna connector then? Is this information more obvious on another brand of adapter? Any ideas of what I can do? Thanks.
I suspect you have soldered to the aerial shield - on a small device like that, they have probably used a planar fractal aerial for size.
On the back side of the board, it looks like there is a pad for a coax style connector - the bullseye feature top right. The outer ring will be a shield, the inner spot the signal. However, I am guessing and YMMV.
Si
There are dirt cheap usb wifi adapters that have an SMA connector, so you can connect any antenna you like.
I wonder if that is a patch antenna. The bullseye feature si_blakely mentioned could be where the feed wire connects to drive it. If it is a patch antenna, the fields are strongest at the edges of the square, and connecting where the OP did wouldn’t work very well.
Duh. I thank you for the reminder. It would be kind of neat to buy the even cheaper ones and modify them, but you’re right, it’s kind of being too precious about the whole thing.
Ok, since you guys have been so helpful, I have further questions. I’ve been working from the pictures of a homebrew patch antenna on this page:
http://www.lincomatic.com/wireless/wifipatchantenna.html
The diagram of the antenna with dimensions is here: Inside the Comtelco 7.5dBi
I got tired of the design with the little metal tabs going to the n-type connector, and I modified it. I was figuring that the keys were the size of the patch on top and the distance between it and the ground. Is that right? Also in these pictures, I realize this is a kludge, and that cardboard is probably not transparent to microwaves and should have been avoided as the spacer. I won’t do that next time.
Pics of kluge patch antenna:
The wikipedia page referenced by ZenBeam shows just a wire going from the patch to the connector. It also calls the bottom plane “ground” My impression is that there is supposed to be no electrical signal going through the bottom plane, and that’s actually the point of the connector. Is that right?
Come to think of it, couldn’t I avoid the connector entirely and pass the wire around the ground/reflector and just out the box? I think the point of the connector is to keep the signal insulated while going through the other metal part.
On the other hand, Figure 2 at the bottom of Inside the Comtelco 7.5dBi seems to show the patch soldered to the ground/bottom plane. Which is right?
What you’ve got there is a Cameo WLG-1502 (rev B3) which lots of companies slap a label on and call theirs. The datasheet linked above says “Antenna - Internal printed antenna”.
You can view the pictures on file with the FCC for the earlier rev B1 board which is very similar. Go to the FCC web page here and enter NHP in the first box and WLG1502C in the second, then click Search. You should see one result. Click on “Detail”. You should end up on a page with 12 exhibits for that ID. Click “Internal photos”, then scroll down to page 3 of the PDF. You’ll see the components under the RF shield that you mistakenly soldered to.
The actual antenna can be seen in that same picture - it is the curved loop etched on the PC board at the top.
That’s some might fine wizarding there, Terry. The antenna loop seems to be covered with a plastic film. You think if I scrape off some of that, I could solder on a wire, or will I burn through or something?
Thanks!
In the FCC photos, it just looks like green PCB resist. If you scrape it, you’ll probably get a solderable copper surface. As long as you have a reasonably low-power soldering iron and don’t leave it on the copper for an extended period, you should be able to solder to it. Generally you’ll smell burning circuit board before the copper delaminates from the underlying board.
However… Bigger isn’t necessarily better. That antenna is tuned for the frequency range used by WiFi. Random lengths of wire will probably just detune it and end up reducing the range. You might want to look at this white paper on 2.4GHz antenna design for some guidelines.
Normally, if you’re going to use an external antenna, you’d use a “rubber ducky” type antenna and something like an MCP01 (PDF datasheet) bidirectional amplifier chip. However, you need a signal to tell the amp when to switch from receive to transmit, which may or may not be available in a convenient manner on the board you have. Plus, by the time you do all this you could just buy an adapter with an external antenna. OTOH, if you’re doing this just for the fun of experimenting, go for it…
Thanks for the input. Did you read my reply at post 6? I am in fact trying to make a dedicated wifi antenna. Do you have any insight to my questions there?
So I have this signal coming from my antenna, and I need to get it into a male rp-sma connector. Do I need shielded coaxial cable? I had this (possibly laughable) idea that I could just take any old wire off the receiving plane of my kludge patch antenna and put it through the rp-sma connector. Would that small signal be degraded too much without being shielded, and is coax really the only practical solution for attaching to the connector?
I admit, I’m trying to keep the cost of this way down.