Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

I'm sharing a house with some friends and I'm in the bedroom right up in the attic. The router is right down on the ground floor and I get connectivity issues.

I've found that if I adjust the two antennas on my network card I can sometimes get a much better signal.

Is there an optimum angle to get the best connection or is it academic? Should they be pointing towards the router? Should they be spread into a V shape or stick out parallel to each other?

share|improve this question
    
It is data dependent –  RedGrittyBrick Oct 3 '11 at 11:03
add comment

6 Answers 6

up vote 2 down vote accepted

It should be set so the waves go in the direction you want them to go. One would think that the waves emanate from the "pole" because that is what the pictures always show |)))) but in reality the pole is only one connection point for the creation of the airborn electrical wave. The "ground plane" is the other connection point. |_))) So the waves emanate out from the pole and the ground plane. (( _\

Google images of ground planes

Fixed: When you see all the VHF and UHF recievers being used in theatres and DJs and all, for some reason unknown to me , they always use the \ / shape, I assume that helps recieve as the signals bounce around so when "line of sight" gets blocked, it still gets there. ))) \__/ ((( Or maybe it works with the ground plane in some way?

Almost always when you're trying to get omni-directionality (((|))) they always go with straight up. Whenever testing any of this stuff, I have found that the most consistant results for omni-directional and movement is straight up. If you have a back ground plane that changes it.

I don't know. I thought I had an epiphany, when I look at a Tesla or large visable electrical output, radio waves in a sence are very tiny powered higher frequencey similarities. I could see the tesla. Tesla pic of the idea

For more epiphanies, it took me days before the light bulb went off that Wi-Fi adapters are not just receivers, they are a transmitter/reciever like the router is.

share|improve this answer
    
Well the router is almost directly below the network adapter so should the antennas be vertical or horizontal? –  Nick Brunt Oct 3 '11 at 0:52
1  
The least signal goes in the direction the antenna is pointing, so if you want signal to go down, you want the antenna to be horizontal. See this page of HP ProCurve documentation. –  David Schwartz Oct 3 '11 at 1:23
    
Edited: Analise the ground plane, if the ground plane is a metal computer, then think about how between the antennas and the computer, you can point the electrical air arc that departs from both as a team. The parabola. A desktop computer with antenna out back, much of the signal is blocked by . . . the ground plane. the signal is shooting out back. –  Psycogeek Oct 3 '11 at 1:23
add comment

The optimal angle is the one where you get best reception at all necessary points in your house.

Experiment!

The problem with theories and specific suggestions is that they can in no way take into account the design of your house, where the appliances are, where you'll be using the signal, and a host of other factors.

Therefore, the best suggestion is to determine the places where you're most likely to use the signal (outside on the patio or porch, in the living room or den, the kitchen, etc) and then try several different antenna arrangements until you find that arrangement that supports the best signal in the various areas of use.

share|improve this answer
1  
I've found this to be largely the case. Regardless of what theoretically is the best from a physics standpoint, the reality is that houses have so many potential sources of interference that you just need to experiment to figure out what will work best. –  nhinkle Oct 3 '11 at 23:03
add comment

Yes, it makes a difference.

There is a "cone of silence", the tip of which comes to the ends of your antennas. I'm pretty sure you get maximum power transfer 90 degrees off of the cone of silence.

In other words, you want the antennas on both sides to be running parallel to each other. It's gets more interesting if you have materials around that are reflective to the wavelengths you're dealing with, then you can end up with a strong path through some intermediate surface.

Oh silly wave geometry, how you like to crop up all the time.

share|improve this answer
add comment

You want the antennas on your card and the router to be roughly parallel to each other.

Thus, if both are on the same floor they would both be vertical.

Router !))) )) ) ) ! You

And if they were on different floors they'd be horizontal.

share|improve this answer
    
I'm trying to think of their optimal positions as being tangential to spheres that emanate from each of the antennae. Would this be correct? –  marklark Sep 9 '12 at 14:33
add comment

It really depends on the design of the antennas. My best guess is that you've got a pair of omnidirectional dipoles (probably the most common externally-visible antenna type on consumer indoor Wi-Fi gear). Omnidirectional antennas are actually only omnidirectional in 360 degrees of a 2D plane, not a 3D sphere (evenly-distributed spherical coverage is called isotropic). Dipole antennas do not have as good coverage out the "top" and "bottom" of the pole. Their coverage is optimized to be roughly equal out all 360 degrees around the "sides" of the pole. So for best coverage, make sure the side of the pole is perpendicular to the direction of the other device.

One way to visualize this is to take a small flat piece of cardboard, cut it into a circle about the size of a DVD, punch a hole in the middle of it and slide it over the dipole. Maybe use tape to keep it perpendicular to the sides of the antenna. Now imagine that the plane of the cardboard extends indefinitely in all directions, and use that to visualize how to point your antennas.

share|improve this answer
add comment

The antenna position doesn't matter but they must be set at right angles to each other.

share|improve this answer
1  
So in other words... the antenna position does matter? Can you explain why they must be set at right angles? –  nhinkle Oct 3 '11 at 23:02
1  
This is incorrect. If you're only trying to cover a single story building, and you've got omnidirectional dipoles, you want them both parallel to each other, perpendicular to the floor. –  Spiff Oct 3 '11 at 23:10
    
I agree with Spiff. +1 –  Doc Oct 3 '11 at 23:12
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.