Posted: Fri Jul 28, 2006 4:03 Post subject: xmit power & high gain antennas
Hi guys, I have a question regarding the xmit power and high gain antennas.
I bought a couple of 9dBi high gain antennas & installed them on my WRT v.3. Should I decrease the xmit power by 8mW to offset the antennas or increase the xmit power by 8mW to offset the antennas? I figure 9dBi @50 ohms is roughly 8mW. I know it's a noob question as I don't know anything about radio physics, but can someone point me to the right direction. Thanks
I don't know much about this either, but naturally increasing the antenna size is going to improve your signal more than increase milliwattage. As it was explained to me, additional milliwattage actually can distort the signal, and cause increased interference, sort of like a really blazingly loud speaker. My only suggestion is you intend on gaining additional range is to read what the wiki suggests: Most Common Problems
Right now what I did to increase my signal is I recycled my BEFW11S4 v4 signal booster which I bought seperately, it increases the 802.11b signals, so everything works fine now. I made the recommended changed under the link above, as well as set my mW output to 28 (which is the default setting). _________________ WRT54G v 5, running DD-WRT v23 SP2 build 7/15/06 (the folder). Currently boosting range using an 802.11b Linksys signal booster, piggy back.
Start with leaving the power at default level and see if just switching the new antennas improove things. Then, increase the power to 50mW and then 80mW and compare. If it gets worse, drop the power to 15mW and 10mW and compare once again. Don't forget to properly align the antennas so that they are parallel with each other since the angle of vertical polarisation decreases as the gain increases.
Hmm.. i don't know how to explain it better but let me put it this way : if the routers are at the sam height, align all the antennas vertically :
router 1 = ||, router 2 = ||.
If one of the routers is higher then the other one, make sure the antennas are tilted so that their sides are parallel ex: router 1 = //, router 2 = //.
This is much more important when dealing with higher gain omni antennas like 9dbi - 15 dbi.
Joined: 17 Jul 2006 Posts: 96 Location: Austin, Texas USA
Posted: Fri Jul 28, 2006 12:32 Post subject:
This is how I did it:
Distance is about 800' between host and client routers. Host has a 10 dbi directional panel antenna and client has a 8 dbi omni directional antenna. Both antennas are outdoors with clear line of sight. Client needed to be omni directional in order for multiple laptops to access it from different locations surrounding the omni.
After installation brought up two IE windows and tweaked power settings until max signal was reached. Then went through the channels until the highest signal reached was found. Happened to be channel one. That was it. Power setting was set to 120mw. Anything higher than 120mw made no difference. Anything much lower than 120mw dropped the signal strength.
Joined: 22 Jul 2006 Posts: 102 Location: somewhere, where it snows
Posted: Fri Jul 28, 2006 18:57 Post subject:
here is the link to PDF file that contents all the info about sizes and construction.
well .. texts are little bit in russian - if it will be some troubles for understanding details - i'll make english translation and publish it somewhere - may be even here.
I think you misunderstood my question. But thanks for answering.
I'm not transmitting over town. I was using default settings but my wireless clients constantly got "Low" signal strength. That's why I bought the 9dBi antennas. I understand that making the antennas are not a hard task, but the ones in the store look a lot better.
Anyway, so by having the high gain antennas, I wouldn't need to increase the xmit power? I left xmit power at 28mW and saw a little boost in the signal strength, increased it by 8, no difference, decrease it by 8, no difference. What is the best choice? Thanks
Ouch. Hope the AIR-ANT2506 I just won for under $10 isn't built like that. _________________ D-link DI-524 -> DD-WRT v23sp1 WRT54GS v2 client bridge -> DD-WRT v23sp1 WRT54GS v1 AP and Apple AirPort Graphite base station bridge
Let's remember, adding 3dB of antenna is doubling the send and receive power. This is the equiv of going from 100 mWatts to 200mWatts.
9dB is 3 times the power - 100mW to 800mW.
Most antennas that I have seen can take a lot of input...certainly 1 Watt. Turning up the WRT is not going to harm the antenna. But, turning up the WRT is going to be asymetrical - not as much receive as gain, so unless you are communicating with another WRT or similar device, you will be inviting problems; as was pointed out, one of the worst is that the long distance link will cause the whole wireless system to slow down.
There is also the problem of distortion increasing somewhere above 100mW to a point where there is not only diminising returns, but also diminishing results.
Of course, we in the EU don't think of these things, as we are only allowed 100mW power output legally anyway, 20dB. And none of us would do anything illegal. Period.
Finally, don't forget about your cables. LM 400 will have little drop over 10 or meters. LM 195 will eat much of your 9dB antenna's positive influence.
i believe that increasing power inside the building may lead to strong multipath signal,
and with having good SNR you will get no speed .
also, if antenna is at open air, it needs to be set, that the signal from all the clients MUST be the same.
otherwise, the closer and stronger client will "blind" the router with signal, RF part autoconfiguring to lower sensitivity and far and weak clients goes offline.
Your understanding of multipath is incorrect. If you increase the transmitter's radiated power by more antenna gain or more TX output, the direct and reflected paths all increase the same amount. Think of a disco ball in a mirrored wall room as an example. Shine a flashligtht towards it. Some light goes past it to the far wall; some reflects to hit that wall. Brighter flashlight, everything gets brighter.
The closer-stronger client idea is also incorrect. WiFi 802.11 is carrier sense multiple access (CSMA). Listen before transmitting a data frame to try to aovid two-at-once collision. So the receiver gets frames from many clients each at a different signal strength and quality. THe RF part doesn't adjust to lower sensitivity as you say.