RT2571WF - External Antenna Hack

[amnya]

Hey guys, i have this old USB CNet Wireless Adapter which i am trying to connect an external antenna onto, as this weekend project

I am having a problem locating where to exactly iron the two ends of the antenna onto, and whether i need to remove a certain resistor for things to work. Below i have attached three screenshots of my chipset in hopes that one of the geniuses around here can figure it out for me

Cheers, and thanks in advance.

[Simon]

what you propse can be tricky and turn out to be worse than before. These things are generally very specificly designed, this is not a shortwave radio you know

[neoone]

yep, I agree with what Simon said. But isn't that an antenna connector at the right top corner of the second picture ? You could find a proper pigtail for that. You would also probably need to get rid of the middle resistor/capacitor (those 3 vertical, where one is not there).

[joelby]

I thought that it looked like a connector jack, too. If so, you shouldn't need to make any modifications at all when attaching a pigtail - they're designed for this sort of thing.

[amnya]

Thanks for your reply guys, i did what this site says, as in, CUT the brass board so its disconnected, and then soldered everything almost like in the photos, but i did not notice ANY signal enhancement. Could it be coz i did not remove that middle resistor? As i think my 8db antenna would want some kind of more power or sth

Thanks

[Fraser]

I have modified PCMCIA Wi-Fi cards to take external aerials and can confirm what Simon said about things being a tad tricky at 2.4GHz !

The built in aerial is carefully designed to match the transceiver chip and must be removed from circuit if an external aerial is to be used. You would otherwise have two aerials working in parallel = impedance mismatch. The fact that you have an RF test point on the PCB is helpful. This is unlikely to be a true external aerial connector, but more like a coaxial test point and is used in the production testing stages with an automated prober. You can disconnect the internal aerial and use the solder pad of the test connector for your coax cable connection.

It is always best to draw the circuit of the transceiver output stage feeding the aerial so as to understand what components are in play. You can have important filters and matching components in such stages that must remain in circuit.

With regard to external aerials.... you MUST use decent quality coaxial cable to connect between the transceiver and the aerial. RG58U would not be my choice at 2.4GHz ! Decent PTFE miniature coax with silver plated conductors is my coax of choice for such applications. The coax feed should be kept as short as practical as every additional cm is additional signal loss, which can be significant on the receive path. Look up the coaxial cable loss figures at 2.4GHz or there about and you will see how much loss you will have for a given length. It is imperative that all connections are of good quality and tails are kept as short as practical.

External aerials can be advantageous over the small internal types but remember, the internal aerial is a well designed and, for it’s size, pretty efficient. Replacing it with a poorly designed aerial on a long coaxial feed will likely result in appalling performance inferior to the internal aerial. When considering external aerials I recommend you limit your high quality silver plated coaxial cable to less than 1m  and use a quality aerial that gives true gain and not the stuff claimed on cheap OMNIDIRECTIONAL  far east Wi-Fi antennas ! Many such antennas claim 8dBi or 12dBi gain yet when I have tested them they are very poor performers. When dissected they contain very cheaply produced stacked dipoles made from poor quality FR4 PCB material and tin plated tracks. Not all are bad but be warned that cheap ones do tend to be poorer performers.
A quality brand name antenna/aerial is recommended for serious work. Think Cisco, Racal or one of the really well known Wi-Fi brands and you will be fine. These will cost you a lot of money though !

One area where you can get VFM with regard to antennas is Yagi’s or patches. These seem to perform well even when sourced from the Far East via e*ay. The amount of gain that such a directional antenna produces tends to overcome it’s inherent inefficiencies !  Cheap ones rarely provide the gain they claim but are normally still pretty good. i.e a claimed 18dBi Yagi will often produce 12dBi useful gain on the antenna test range. They can be very cheap as well.  The down side of such antennas is that they are very directional.

Consider this....

I have had excellent results placing an unmodified USB Wi-Fi 'stick' at the focal point of a parabolic dish. = High performance directional gain !

P.S. That 'Brass' track you mentioned....it's actually GOLD plated copper = high performance at microwave frequencies  

Have fun  

[amnya]

Wow, thanks Aurora for the wonderful info Thats way more than what i expected, seriously thanks

For your COAX advice, I will not even use a coax cable, i will simply solder the the SMA female connector directly on the chip. I soldered the center of my antenna to a soldering point that is connected to the center of that "small connector" on the top right of the 2nd screenshot. Hope that was the right spot.

Today i will remove that middle resistor as neoone mentioned. (Please tell me if that is a correct move to do).

1st Image: Antenna to be used

2nd Image is the result of what i am doing, please advise!

3rd Image is an SMA connector close to the one i am going to solder.

Please verify as you all seem to be very experienced

[neoone]

what I meant by removing this resistor is that from the photo it seams to physically connect PCB antenna with the connector therefor with signal track from the WiFi module. Confirm that first.

[amnya]

what I meant by removing this resistor is that from the photo it seams to physically connect PCB antenna with the connector therefor with signal track from the WiFi module. Confirm that first.

I can't quite understand what you mean Can you please clarify? Should i or should not i remove that resistor? Coz from what understand since i am willing to connect an external antenna, i should disable the signal track on the chip. And yes, looking at the chip, is looks like after this resistor we have the track

I will try to remove that resistor, though i already physically cut the track as shows in the photo, but i will now also remove the resistor wish someone can verify my previous post

[amnya]

Just did the stuff in the post, did not change the signal at all! neither better nor worse, its like am totally not at the right location/point.

Reverted back to stock settings until we get some feedback

[Fraser]

Amnya,

Please do not take offence but you appear to be 'dabbling' in an area in which you lack experience. Neither I nor other members of this group can analyse what is going on with your Wi-Fi transceiver remotely. I have explained that you should draw out the schematic for the aerial/antenna section from where it leaves the transceiver. If we have that information we may be able to assist with some useful comment.

I have to ask why you believe your whip antenna will provide much increased performance ? If you were to take that type of antenna apart you would find a pretty poor design not much more efficient than a simple 2 element dipole at 2.4GHz. Have you tried the antenna on another Wi-Fi unit that is designed to feed an external antenna ? This would give you some idea of the performance that you can expect from the antenna when fitted to your USB stick.

P.S. that component that you described as 'Resistor' and were about to remove looks distinctly like a chip capacitor.

If you can't identify the components when drawing the schematic, at least draw small boxes and give them some form of identitiy so that we can tell you which ones are likely to be important to your modification. I can try to work out the circuit from your pictures but that takes time and I have little of that spare at the moment. If you get really stuck, I will have a go for you.

[Fraser]

OK,

I took pity on your plight and drew the schematic on my notepad.

The RF path to/from the transceiver is clearly visible on your picture named 'Back.JPG'. It is in the lower left corner of the PCB ... a surface mount capacitor feeding a curved PCB track that in turn is connected to the RF test socket we have discussed. The output of the test socket connects to two capcitors, one is coupling to 0V and the other is the feed to the internal antenna. There is also a small black SMD resistor across the input to the internal antenna. Note. The RF test socket looks to contain an internal switch that disconnects the internal antenna when a probe is inserted into it.

The internal antenna should be removed from circuit by removing the capacitor that feeds it, and the capacitor to 0V. The transceiver input/output will then be connected only to the RF test socket. Connect you external antenna to the curved PCB trace (already mentioned) on the bottom of the PCB and connect the ground side of the antenna to the adjacent groundplane copper.

If this gives no significant improvement in performance, ensure you have your external antenna in the same polarisation as the antenna on the Wi-Fi access point. If there is no improvement, you either have a damaged external antenna or it is just cheap junk and should be replaced with something better.

Hope this helps.   

[neoone]

thumbs up for your effort Aurora That's basically what I meant with removing those elements.

[Mechatrommer]

if me. i will cut the track as you shown above but not taking out any component, the track on the left will be connected to ground (already did) and just connect the sma connector to there. ie ground female to ground track, and center pin to the right side cutted track.