IS there a distinction between the two antenna connectiond on the WRT54 wireless routers. I had heard that one is for TX and the other for RX but my router seems to work equally well with either antenna disconnected. I cant tell a difference in performance unless I take them both off.
On Tue, 22 Jul 2008 11:41:51 -0700 (PDT), jimmi...@gmail.com wrote: >IS there a distinction between the two antenna connectiond on the >WRT54 wireless routers.
No. The antennas area identical. However, the two ports are slightly different. One is designated as "main" while the other is "aux". The router will normally favor the "main" antenna and only switch to the "aux" if the diversity receive algorithm determins that the "aux" antenna has a better signal.
Unfortunately, I can't recall which is antenna is main or aux, or for that matter, which antenna is considered "left" or "right". Sigh. Looking at the inside photos at: <http://www.linksysinfo.org/forums/showthread.php?t=47124> I would *GUESS*(tm) that the antenna port with the long coax cable running across the board would be the "aux".
>I had heard that one is for TX and the other >for RX but my router seems to work equally well with either antenna >disconnected. I cant tell a difference in performance unless I take >them both off.
Some routers will only transmit on one antenna (i.e 2-wire). Routers that have two antennas invariably use diversity receive and sorta "scan" between the two antennas in receive. I vaguely recall that the WRT54 transmits on the same antenna that it receives.
However, you should be able to tell the difference with only one antenna attached. If you remove the antenna from the "main" port, the stupid firmware will continue to favor the "main" port, even though there's no antenna or signal. If you ping the router, you'll probably find that the first ping will be lost using just the "aux" antenna. That's not the case using only the main antenna.
If you insist on using only one antennna (a good idea if you're attaching external antennas), some alternative firmwares have a means of disarming the diversity switch system and just using one antenna.
> On Tue, 22 Jul 2008 11:41:51 -0700 (PDT), jimmi...@gmail.com wrote: > >IS there a distinction between the two antenna connectiond on the > >WRT54 wireless routers.
> No. The antennas area identical. However, the two ports are slightly > different. One is designated as "main" while the other is "aux". The > router will normally favor the "main" antenna and only switch to the > "aux" if the diversity receive algorithm determins that the "aux" > antenna has a better signal.
> Unfortunately, I can't recall which is antenna is main or aux, or for > that matter, which antenna is considered "left" or "right". Sigh. > Looking at the inside photos at: > <http://www.linksysinfo.org/forums/showthread.php?t=47124> > I would *GUESS*(tm) that the antenna port with the long coax cable > running across the board would be the "aux".
> >I had heard that one is for TX and the other > >for RX but my router seems to work equally well with either antenna > >disconnected. I cant tell a difference in performance unless I take > >them both off.
> Some routers will only transmit on one antenna (i.e 2-wire). Routers > that have two antennas invariably use diversity receive and sorta > "scan" between the two antennas in receive. I vaguely recall that the > WRT54 transmits on the same antenna that it receives.
> However, you should be able to tell the difference with only one > antenna attached. If you remove the antenna from the "main" port, the > stupid firmware will continue to favor the "main" port, even though > there's no antenna or signal. If you ping the router, you'll probably > find that the first ping will be lost using just the "aux" antenna. > That's not the case using only the main antenna.
> If you insist on using only one antennna (a good idea if you're > attaching external antennas), some alternative firmwares have a means > of disarming the diversity switch system and just using one antenna.
Thanks Jeff, I am guessing that the traces on the board may form a type of diplexer. If this is the case leaving one antenna disconnected and the port unterminated would defeat the operation of the diplexer. In this case one antenna connected to one port would work as well as connecting it to the other. This would explain my experience of having the unit work equally well with 1 antenna on either port. Having 1 or 2 antennas woud probably make little difference if ther is not a multipath issue
On Tue, 22 Jul 2008 15:27:51 -0700 (PDT), jimmi...@gmail.com wrote: >Thanks Jeff, I am guessing that the traces on the board may form a >type of diplexer.
Nope. It's a diversity switch. If you look carefully, there's a 6 lead chip sitting at the junction of the two i/o ports. The chip does the switching.
>If this is the case leaving one antenna disconnected >and the port unterminated would defeat the operation of the diplexer. >In this case one antenna connected to one port would work as well as >connecting it to the other.
Not exactly. The port with the long coax cable (aux) has a bit of extra loss in the cable and exhibits perhaps 0.5dB more loss. Not worth worrying about unless you're trying to squeeze every last dB out of the system.
>This would explain my experience of having >the unit work equally well with 1 antenna on either port. Having 1 or >2 antennas woud probably make little difference if ther is not a >multipath issue
Read the Cisco article. The problem is when you have two different types of antennas. It's quite possible that the wrong antenna will hear the signal first. The diversity algorithm doesn't continuously test for the best signal. If it hears something on the FIRST antenna that hears a signal, it could easily just sit there forever. The only justification for switching antennas is if there is a signal loss, or a rather high bit error rate. If you're planning to use just one antenna, you should disable the diversity switch in firmware.
> On Tue, 22 Jul 2008 15:27:51 -0700 (PDT), jimmi...@gmail.com wrote: > >Thanks Jeff, I am guessing that the traces on the board may form a > >type of diplexer.
> Nope. It's a diversity switch. If you look carefully, there's a 6 > lead chip sitting at the junction of the two i/o ports. The chip does > the switching.
> >If this is the case leaving one antenna disconnected > >and the port unterminated would defeat the operation of the diplexer. > >In this case one antenna connected to one port would work as well as > >connecting it to the other.
> Not exactly. The port with the long coax cable (aux) has a bit of > extra loss in the cable and exhibits perhaps 0.5dB more loss. Not > worth worrying about unless you're trying to squeeze every last dB out > of the system.
> >This would explain my experience of having > >the unit work equally well with 1 antenna on either port. Having 1 or > >2 antennas woud probably make little difference if ther is not a > >multipath issue
> Read the Cisco article. The problem is when you have two different > types of antennas. It's quite possible that the wrong antenna will > hear the signal first. The diversity algorithm doesn't continuously > test for the best signal. If it hears something on the FIRST antenna > that hears a signal, it could easily just sit there forever. The only > justification for switching antennas is if there is a signal loss, or > a rather high bit error rate. If you're planning to use just one > antenna, you should disable the diversity switch in firmware.
Interesing, this has got me wondering how hard it would be to alter the software to use the the switch as a T/R switch. My thought is to use one antenna for Rx with an antenna mounted LNA. I think I need to check in with the guys over at DD-WRT.
On Sat, 26 Jul 2008 11:02:17 -0700 (PDT), jimmi...@gmail.com wrote: >Interesing, this has got me wondering how hard it would be to alter >the software to use the the switch as a T/R switch.
Fairly easy. If you lift the metal shield from the RF section, and follow the RF path inside, the first part it hits is a 2.4GHz ceramic bandpass filter. The next part is the T/R PIN diode switch, which should be the same chip as the diversity switch. A bit of creative wiring should be able to make it do what you want. No sofware required.
>My thought is to >use one antenna for Rx with an antenna mounted LNA. I think I need to >check in with the guys over at DD-WRT.
The diversity algorithm is part of the firmware for the Broadcom chipset. I think (not sure) that the interface is supplied as a pre-compiled linkable library and cannot easily be tweaked. That's also why you don't much in the way of MAC layer (RF) statistics from Broadcom. There was some effort made towards reverse engineering the interface library, but I have no clue what's happening.
> On Tue, 22 Jul 2008 11:41:51 -0700 (PDT), jimmi...@gmail.com wrote: > >IS there a distinction between the two antenna connectiond on the > >WRT54 wireless routers.
> No. The antennas area identical. However, the two ports are slightly > different. One is designated as "main" while the other is "aux". The > router will normally favor the "main" antenna and only switch to the > "aux" if the diversity receive algorithm determins that the "aux" > antenna has a better signal.
> Unfortunately, I can't recall which is antenna is main or aux, or for > that matter, which antenna is considered "left" or "right". Sigh. > Looking at the inside photos at: > <http://www.linksysinfo.org/forums/showthread.php?t=47124> > I would *GUESS*(tm) that the antenna port with the long coax cable > running across the board would be the "aux".
> >I had heard that one is for TX and the other > >for RX but my router seems to work equally well with either antenna > >disconnected. I cant tell a difference in performance unless I take > >them both off.
> Some routers will only transmit on one antenna (i.e 2-wire). Routers > that have two antennas invariably use diversity receive and sorta > "scan" between the two antennas in receive. I vaguely recall that the > WRT54 transmits on the same antenna that it receives.
> However, you should be able to tell the difference with only one > antenna attached. If you remove the antenna from the "main" port, the > stupid firmware will continue to favor the "main" port, even though > there's no antenna or signal. If you ping the router, you'll probably > find that the first ping will be lost using just the "aux" antenna. > That's not the case using only the main antenna.
> If you insist on using only one antennna (a good idea if you're > attaching external antennas), some alternative firmwares have a means > of disarming the diversity switch system and just using one antenna.
Hi, I discovered DD-WRT software does enable one antenna as RX and the other as TX. This is user selectable. I dont know if the origonal Linksys software allowed this or not but I doubt it. This may come in handy because I want tot try using a 13cm LNA with mine to see what happens.
