satellite dish/cantenna related question

[daveoily]

One for those among you that know the theory....

I have a spare sattellite dish, and have connected it to my belkin usb 811 g stick, (I'll edit this later with exact details of what I'm using if that will help). the dish seemed to amplify the signal, so I thought covering it with bacofoil would make the signal even stronger (it's one of those dishes with loads of holes in it)....

Not so! It seems the signal is weaker if anything, so I was talking to my mate in the pub about it, and he started waffling on about wavelengths and other things, (I can't chat too much physics and maths when i'm drinking, but sometimes it can be fun to try :p) and he thinks that when recieving these signals it's better to have something with spaces in it, so I mentioned chicken wire to him and he says that'd probably be good.

So can someone give me an explanation of why this would be so, something fairly simple to understand. I'm just not getting it at the moment, another friend seems to think that the bacofoil covered dish is acting like an antenna, and trying to recieve the signal, which would explain my poor results.

Now this same friend was talked about an array, mentioning strips of bacofoil hanging off of a clothes horse, that made no sense to me, can anyone give me a little insight into what he might have been on about (he's the sort of friend that likes to know a little, but not bother with the detailed explanation, which can be annoying sometimes!)

[Greg]

Maybe you can get better help searching for antenna building?

That's outta my realm.

[daveoily]

I've had a look, but i thought someone here might be able to give me the quick version, good links to look at that i've found so far include...

http://users.skynet.be/chricat/horn/horn-javascript.html

http://users.skynet.be/chricat/BT.html

http://www.netscum.com/~clapp/wireless.html

http://www.oreillynet.com/cs/weblog/view/wlg/448

http://www.turnpoint.net/wireless/cantennahowto.html

I've yet to find anything on an array really, I guess it's because the word array can be used in so many other contexts!

[Greg]

Looks like fun, but I don't use wireless.

[daveoily]

well, it all makes good reading, if you want to increase the range, signal strength or security of your wireless network, so it may be of some use, even if i don't get the noddy answers to my original questions....

[Greg]

well, it all makes good reading, if you want to increase the range, signal strength or security of your wireless network, so it may be of some use, even if i don't get the noddy answers to my original questions....


It would be great to see you get this going and become an authority on it here at the computer help forum.

Good luck.

[daveoily]

Admins eh? Always after help :P well, I guess I already owe you guys for sorting out my wired networking question!

Like I said, I'll try and help out on at least a one question solved, one question answered type basis, but I'm not always able to make it online... work restrictions etc...(which by the way it looks like I may have found a way around, more reports to follow IF it works :P)

It might take me a while to become an authority on the subject...

I mean, I don't even know the difference between b and g in this field, still, it's a thought, I know a few odds and sods of stuff, so you never know :P

[Greg]

We're always happy to hear from ya. Keep us updated on your experiment.

[bigH2O]

One for those among you that know the theory....

I have a spare sattellite dish, and have connected it to my belkin usb 811 g stick, (I'll edit this later with exact details of what I'm using if that will help). the dish seemed to amplify the signal, so I thought covering it with bacofoil would make the signal even stronger (it's one of those dishes with loads of holes in it)....

Not so! It seems the signal is weaker if anything, so I was talking to my mate in the pub about it, and he started waffling on about wavelengths and other things, (I can't chat too much physics and maths when i'm drinking, but sometimes it can be fun to try :p) and he thinks that when recieving these signals it's better to have something with spaces in it, so I mentioned chicken wire to him and he says that'd probably be good.

So can someone give me an explanation of why this would be so, something fairly simple to understand. I'm just not getting it at the moment, another friend seems to think that the bacofoil covered dish is acting like an antenna, and trying to recieve the signal, which would explain my poor results.

Now this same friend was talked about an array, mentioning strips of bacofoil hanging off of a clothes horse, that made no sense to me, can anyone give me a little insight into what he might have been on about (he's the sort of friend that likes to know a little, but not bother with the detailed explanation, which can be annoying sometimes!)

I'm not going to get too technical with you here, but here is the basic premise of your issue. The "holes" or mesh in the dish is specifically designed to produce a 1:1 gain on the signal coming in... or no gain. The curvature, size of the dish, and size of the mesh all mathematically cooperate to turn the 1/4 wavelength signal received into 1/4 wavelength targeted at the low noise block. These antenas are designed this way because the signal coming from space is the signal that the old C & Ku band receivers expect to receive, but more so because of aiming difficulties. These dishes are designed to be rotated from bird to bird, and rarely have a fixed azimuth, so the signal has to be able to "bounce" around the dish and still provide the appropriate 1/4 wavelength signal to the LNB. You can get within 3-5 degrees of the bird you're shooting for and get the appropriate signal because of the cooperative design of the mesh.

Now, fleshing that dish out with a metalic wrap will give you a gain on the dish... BUT, it has to be perfectly smooth, and every single spot on the wrap will have to be perfectly angled at the focal point of the dish or you will get negating spots that will reduce your gain. Enough "rough" spots in your skin and you will reduce the gain back to the standard 1:1 gain the dish is designed for, and further variations in the skin will take it below 1:1.

Once you get a perfect skin, then aiming becomes critical. No longer do you have a 3-5 degree window of aiming to collect your signal... you're now down to 1 degree or less.

My advice would be to get rid of the mesh dish and find an old oval shaped satellite dish like an Echostar, DirecTV or DishNetwork solid mini dish. I don't know where you hail from... but with the terminology you are using I'm assuming you aren't in the US. I don't know what brands of satellite television you guys have where you're from, but a standard mini oval dish with a home brewed cantenna in place of the LNB should give you somewhere in the neighborhood of a 20-25dB gain on your signal. That's more gain than you'll ever need... would probably give you over 10 miles of point to point (line of sight) signal with no appreciable loss of connection speed. Aiming the antenna would be another issue though. The ten mile distance is purely theoretical, because there is no way in hell you could ever aim it that precicely, even if you could find two points ten miles apart that had line of site.

The dish is really moot though. Two appropriately designed cantennas aimed corectly at each other can achieve about 18dB gain, and give you full strength signal for over a mile line of sight. Tough to aim, yes, but if you're shooting at your neighbors house 500 yards away, it's no big problem.

[Zero Tolerance]

I agree with all of what BigH20 said except I wouldnt use 2 cantennas. I would use 1 cantenna and a 90 degree or even a 360 degree antannae. That would put you over 20dB of gain and you wouldnt have any problems other than if you have neighbors within your range that are doing the same thing.

Everone needs to remember that 2.4 wirelss only has 11 channels of operation and you have to be at least 5 channels apart to keep from interfering with each other. So that leaves only channels 1, 6, and 11 that would operate without any interferance. If someone has a wireless network on 2,3,4,5,7,8,9,or 10...then their signal is going to interfere with yours if they are within your range.

So my advice to you would be to get an application like netstumber and drive around your intended range and see what signals you find out there. It will also tell you what channel they are operating on as well. A GPS will help you with a lineplot of the area so you can detail your range. Then plan how much gain you really need so maybe you can eliminate any problems before you even get started.

Wireless problems can be a total bitch to figure out because there are so many variables.

PS...the main difference between B and G standards is the speed. B=11mbps and G=54mpbs

PSS...Moved this thread to the Networking forums where it would be more appropriate

[bigH2O]

My main reasoning behind using two cantennas instead of one cantenna and an omnidirectional antenna are three fold. One of the reasons addresses the very concerns you have regarding available channels. Direct point to point "pencil beam" between two cantennas will almost totally eliminate any chance of interference. The second reason is security. With a pinpoint line of communication, someone has to be directly in the beam to pick up your signal at all. Even with WEP security enabled, you're signal can still be intercepted with an omnidirectional antenna, and if somebody wants in bad enough... well, you know the drill. The final reason is economics. A good omnidirectional Yagi antenna that will supply the same point to point gain in signal strength as a basic cantenna is going to cost you well over $100.00. A cantenna can be built for under five bucks, and you get a meal out of it at the same time :)

Yes, aiming two of them to work together can be a bitch... but once it's done, it's done, and all of the other issues then go away.

[Zero Tolerance]

Point to point with cantenna's dont provide pencil like wavelengths. Although it is a more concentrated pattern....the Frenel Zone is pretty large when you pump the gain up higher than any recommended indoor antannae.

I just tested this theory minutes ago by connecting a cantenna to my wireless router and pointing it out a window. Took my laptop outside and could pick up full signal 54Mbps when in the line of sight of antannae. However I could go 45 degrees or more from center on either side and still get a good signal...high 20Mbps to low 30Mbps. As I got further into the Frenel Zone the signal increased in either area of center. I could maintain high 40Mbps to full 54Mbps while in line of sight.

My test equipment was as follows...

Linksys WRT54G modded with DD-WRT firmware by Brainslayer for 1st point wireless
IBM laptop with pcmcia "G wireless card" for 2nd point wireless
Acer laptop with built in wireless G running netstumbler for the test

No matter what kind of wireless hardware you use...the Frenel Zone is going to be a factor, especially over long distances. If you are only going a short distance and dont mind the time making sure you have a clear unobstructed path from point A to point B then go with 2 cantenna's.

If you are going a greater distance then I would go with a cantenna and a 90 degree directional antannae....and then lower your voltages on your router or access point so you are dont drive your signal too far past your destination.

Both of us are right in the end.....how you do it is up to you. You just now have at least 2 options to chose from. :D

[bigH2O]

You are most likely correct in your tests Zero, and I'm not denying your test results... more presenting this as a reason why you obtained the results you obtained.

A perfectly aligned N connector will make a beam no larger than the diameter of the can you are using. Of course a perfect alignement is theoretical. Even if you measured to 1/1000 of an inch, you'll be off somewhat, and the drift of your drill bit before it bites into the metal is going to throw you off a tad. Theoretically, the beam in the top picture is what you want to achieve. Realistically, you'll get one of the other two.

The more precise you can be when building the cantenna, the narrower your beam will be.

[bigH2O]

Actually my middle sketch should be more like the one attached here, but you get the drift (no pun intended :))

[Zero Tolerance]

I wasnt using a home-made cantenna. I was using a prefab one from Cisco...so I would think that the N-connector was in the right spot. But ya never know...LOL :D

[Zero Tolerance]

Heres a pic.

PS...dont tell Nopp I didnt use my gallery! LMAO :D

[bufordt]

I was the person who did all of the wireless testing for the state of PA when they were developing a wireless policy for all departments that are under the jurisdiction of the Govenors office. We used both comercial antennas, and cantenna's, and professional and home made Yagi antennas. Without a doubt, comercial Yagi antennas like the one that Zero shows in his picture are the best if you would like keep the beam as narrow as possible. The pictures from H2O are very good in showing how the placemant of the NConn can have drastic effects on signal direction. But as you know, radio signals, and that is all that WiFi signals are, will not stay straight unless you have them contained within a shielded area. They are affected by way to many outside variables. We did extensive testing on the state hospital property that showed that even with the best built commercial Yagi's we could not contain the signals. If you have ever seen the effects of a ripple created in a pond by dropping a stone, then you will know what I am talking about. Radio signals get bigger as the distance increases until the signal is so weak as to be of no consequence. But put obstacles in the way of the ripple(a Branch or stone) and see how the ripples change directions. This is also true of WiFi signals.

Cantenna to Cantenna, or even better yet, Yagi to Yagi is the best bet for keeping the signal on track. If you are going for distance only, two parabolic dishes of the mesh type are best, but the signal will go in all directions.

We had some really interesting results with some of our experiments. No matter how we positioned the antennas, or what type of antenna we used, we were never able to predict the shape or range of the signal.

[daveoily]

Wow, you guys go all out on explaining stuff don't you?

Thanks :) and thats one mean looking heavy duty type cantenna you have there Zero!