7dbi antennas worse than original ones

Discussion in 'Networking Issues' started by nkwai, Dec 7, 2006.

  1. nkwai

    nkwai Network Guru Member

    Yesterday I bought the Linksys 7dbi antennas for my WRT54GL (Running DD-WRT v23 SP2 std). To my great surprise the signal was a worse than with the original ones that come with the ruter. Even after rising the output power from 28mW to 70mW, the signal was still weaker. I followed the guide about wireless signal tweaking in the article section there was no change. Can someone just tell me whats going on, because I don't understand it.
    Thank You.
  2. d__l

    d__l Network Guru Member

    When I installed them on my WRT54GS, I had similar results. It was my theory that they expanded the range of interfering signals and noise that I could pick up. When I used the unsupported signal scan feature on the router, I definitely saw more WLANs than I saw with the original equipment antennas.

    I think they would be ideal for use in a large metal warehouse where you are shielded from other signals.

    I eventually resorted to making parabolic antenna shields to focus and redirect the signal and also block the neighbor's signals.
  3. nkwai

    nkwai Network Guru Member

    I bought them because I had to extend the signal to my neighbour wich is 4 wals and cca 20-25m away, because the original ones didn't reached her.
  4. HennieM

    HennieM Network Guru Member

    Maybe try one antenna only. It may be that, with the greater gain, the antennas are interfering with each other.
  5. Esquire

    Esquire Mesquire Staff Member Member

    Or try aligning the antennae to different planes.

    I bought a Planex PCi 10dBi omnidirectional antenna recently to add to a friend's desktop adapter and I got no improvement at the same location as the original, but by moving the antenna's location away by 3 feet I was about to improve the reception quality by 100% - thank goodness it comes with a 3' extension cable (it has its own stand), so try an extension cable to relocate it if all else fails.

    I'm even thinking of getting a couple of these antennae myself :biggrin:
  6. bcbogger

    bcbogger LI Guru Member

    A little antenna theory may help understanding here..

    An antenna's job is to take the rf power and radiate it, as such it cant radiate more power than it receives.

    the simplest antenna is what is known as an isotropic antenna, it radiates in a sphere equal power in all directions.

    next is the Dipole antenna, typically it is 3db gain, or 3DBi <- the i means its referenced off of a theoretical isotropic radiator)
    it radiates as squished sphere, radiating slightly unequally.

    then we get into co-linear antenna's which is what you have.. It consists of several dipoles stacked ontop of one another 180deg out of phase.
    these take the power that would normally be radiated top and bottom and push it out the side in a donut pattern

    in this case its taking your 28mw and making it 140mw

    28mW log 10 = 14.47 DBm + 7 = 21.47 DBm ERP

    10 ^ (21.47DBm / 10) 140.28 mW ERP

    now you are not getting the extra power for free.. it comes from somewhere, and where it comes from is a change in you radiation pattern.

    If you look at the radiation pattern of that antenna im sure you will see that it radiates in a donut shape. so there is lots of power along the plane of the antenna but its radiating very little top and bottom.

    this is important because, if your receiving antenna is not on the same plane as the radiation pattern, it will suffer.

    for example, if the gain off the bottom of you antenna is down 30db (quite typical) then doing the math gives you

    14.47dbm - 30 = -15.53dbm =.028 mW ERP or in real world terms not much..

    so the long and the skinny you need to conceder the radiation pattern of the antenna.. If you antenna is up a tower and you are 10km away with another gain antenna "catching" the signal at the other end, it works great.
    If you have the same antenna up a tower and you are sitting at the base trying to get signal.. well.. you may be out of luck.

    Hope this helps..

  7. nkwai

    nkwai Network Guru Member

    Hmm...The router with the 7dbi antennas is on my monitor, the laptop that have to recive the signal is about 0,8-1,0m lower at a distance of 20-25m.

    Can you explain a little more this phrase "if the gain off the bottom of you antenna is down 30db (quite typical)" I don't understand it.

    Would it be best if I turn one antenna horizontal and laeave the other vertical?
  8. crawdaddy

    crawdaddy Network Guru Member

    not exactly, but as you slowly move the laptop away from the router, your signal may improve. the same concept is used with cell towers. If you're standing at the base of a cell tower, you very well could have NO service because the tower is radiating the signal outwards. Tou go down trhe block a little bit and voila, you have cell service. Also, the actualy orientation of your laptop's wifi antenna can also have an advesre effect on the signal. If your router antennas are vertically polarized, then for optimum reception, your laptop antenna also needs to vertically polarized. If you have an intenrnal laptop wifi card, you may never know what the polarization is. This can seriously screw you up...Hope this helps with your understanding.
  9. nkwai

    nkwai Network Guru Member

    The linksys antenna have linear vertical polarization (don't know what it means), the pcmcia card of the laptop have only this data "Antenna: Printed Antenna with Diversity System" (don't know either).
  10. crawdaddy

    crawdaddy Network Guru Member

    in my opinion, if it's a printed card, it's probably a dipole with vertical polarization
  11. nkwai

    nkwai Network Guru Member

    Sorry but this means nothing to me.
  12. bcbogger

    bcbogger LI Guru Member

    this document http://www.pacwireless.com/24GHz_Mobile_Antenna_Patterns.pdf shows an excellent example of what the e field (or vertical) radiation pattern of a 7 dbi collinear antenna looks like..

    think of this as a view of the power output as viewed as a slice looking from the side.
    So what we have in this example is an antenna that is driven with a 0db reference signal, out the sides at the 90 and -90 positions, the signal is quite high, around 0db, now if we look at the bottom or 180 deg we see that the signal is around -25db, or "down 25 db", we can see that there are even a couple of nodes that are even less +-160 deg is down 35 or so db.

    We dont know what the h plane or top view is, because it isn't shown in this example, but with a stacked collinear, you can assume a fairly even distribution.

    ok, your laptop card, is most likely using a "chiptenna" which receives in random polarity, meaning it receives both vertical and horizontal polarity equally as crappy.
    If you stand in front of your antenna, and put you laptop and its card, in the same horizontal plane, it will receive well, because it is in the prime area of your antenna's radiation pattern, as you move below the horizontal plane of your antenna, you will notice it drop off.

    so the long and the short is, that sometimes high gain antenna's dont help, if you need equal coverage you are better off with an antenna that is closer to unity gain, where the signal is radiated in a sphere rather than a donut,

    what I've sometimes done to cover a house is to put the AP in the attic, with a 60 deg corner beam pointed down though the house, this will put the most amount of signal where you need it.

    I don't know what your situation looks like, but if you want pm me you email address so you can perhaps send me a rough drawing, i could make a recommendation what antenna and where you need to place it from that.

  13. nkwai

    nkwai Network Guru Member

    Hmm...I'm starting to understand.
    My situation is like this:
    The router is in the middle, and I have a computer exactly to the right of it, and one exactly to the left of it.
    Now what I'm thinking is:
    If I put 2 beams in the middle of the antennas so they will reflect the signal of the left antenna to the left, and the signal of the right antenna to the right, it should work a lot better?
    Does anyone have a project how to build this beams?
  14. HennieM

    HennieM Network Guru Member

  15. nkwai

    nkwai Network Guru Member

    Thank You.

    Now for thos interested here is the response of Linksys:
    This is just bull**** because they are too leazy or don't know the answer.
    The G/GS/GL are all almost the same so there is no reason they won't be campatible.
  16. HennieM

    HennieM Network Guru Member

    I can agree with the "...tested..." part, but to state "...not compatible with the GL..." is probably longhand for "I don't know".
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice