alkats

Hello, Group!

I would like to ask anyone with enough experience a question regarding a simple seven element colinear antenna shown below.

As I was browsing the internet in search of that "ultimate" antenna design, I found very few pages with the description of this antenna. All others were much more elaborate and more time consuming to make.

So, I decided to repeat the shown below antenna -- the result is shown on the attached photographs. It took me ten minutes to make:

bent wire on a wooden chuck with pliers,
solder wire it to the end of RG-58 and
mount BNC jack to the other end.

I was amazed with its performance! The antenna was 80%~85% as effective compared to the tested gamma feed yagi with 14dB gain!

Since I don't have any equipment to test this antenna except view the chart on the netstubler, I humbly ask if someone could analyze this antenna and kindly give an intelligent estimate to its parameters.

Here is the ASCII sketch of it in case photos would not upload:


tip 83mm - 5x90mm middle sections - base 61mm - 150mm sleeve soldered at the base
each /\ is an open loop 10.8mm outside diameter

----/\-----/\-----/\------/\-------/\-------/\---[============150mm brass sleeve

Any opinion would be greatly appreciated -- thank you. In one of my posts below I added external links to show pictures in bigger size.

Attached Images

	7 elements colinear.gif (6.8 KB, 47 views)
	7el_colin_tip_01.jpg (33.0 KB, 43 views)
	7el_colin_middle_02.jpg (26.1 KB, 42 views)
	7el_colin_base_03.jpg (43.2 KB, 46 views)

streaker69

RG58 is NOT rated for 2.4ghz.

BNC connectors are not recommended for 2.4ghz.

I've seen a blind man's helper monkey solder better.

What card did you use to come up with your 80-85% improvement figures?

Are you related to this guy?

building antennas

__________________
Treat your gun like your genitals, only whip it out when it's absolutely necessary.

theprez98

Recommendation

Also, as stated in the previous post, BNC connectors are not recommended due to high losses.

RG-58 loss is 1.2 dB per meter at 2.4 GHz.

__________________

alkats

I used D-Link DWL-AG520 PCI bus card.

Agreed, soldering could be better but I was aiming at electrical connection not on a museum piece.

I though BNC stands for "bayonet" and TNC stands for "thread" connector, and TNC is widely used for 50 cable and wifi applications.

I used following method to estimate the performance:
I have 11 elements gamma match tuned yagi with tested 14 db gain and it showed a green bar at approx 28~30 mark with netstumbler.

When I attached seven element colinear the signal was approximately at 22~24 mark at the same location. If I disconnect antenna and cove the pigtail with the palm of my hand the signal would drop to 1~2 mark on the same screen.

That is all and thank you for an opinion.

theprez98

Here's your first problem. The card is NOT supported, the SNR info is NOT accurate, thus your number estimates are wrong.
BNC stands for bayonet Neill-Concelman. The BNC is not recommended because of high losses. TNC is a threaded variant. Its performance is considered better than the BNC at microwave frequencies and thus why it is widely used.
See above why your numbers are not valid.

__________________

streaker69

If you're taking your measurements with Netstumbler and that card, then your measurements by no means accurate.

You are correct about BNC and TNC, but BNC while it's spec's say will work with 2.4ghz it is not recommended. TNC connectors should be used.

While you may not have been trying for a 'museum' piece, good soldering does count when you're working in the 2.4ghz range. As does accuracy of the bends.

__________________
Treat your gun like your genitals, only whip it out when it's absolutely necessary.

alkats

Thanks, so what would be the estimate gain/performance for this antenna?
When I used D-Link supplied wifi manager, the yagi showed 54MB at 60% signal strength. This colinear showed 46MB at 40% signal strength.
That is a much of a high-tech equipment I got.

wrzwaldo

Taking the soldering and wavy lengths of wire into consideration I'd give it no more than a 7 in gain.

Barry

You thing there would actually be a gain?? That first big blob looks like it's shorting the shield and center conductor.

__________________
Never do anything you don't want to explain to the paramedics.

alkats

Well, I did not expect such sarcastic reaction at all. I was hoping for an intellegent engineering opinion. Regardless whatever is said above the antenna does work admirably (for me at least), the signal quality and strength are very good and I am pulling in two more SSID from around the place where I live. I uploaded full images (1024x768) to photobucket.com and it could be seen that there is no short at the base. I pulled RG-58 through 5/8" diameter by 150 mm long brass tube, brought it to the base of the antenna and folded over the brass tube some of the braid shield. Then I looped the shield in place with a turn of the same wire and soldered it around in one spot very close to the base. That is why it appears to have a short.

http://i136.photobucket.com/albums/q...olinear_01.gif

http://i136.photobucket.com/albums/q...olinear_02.gif

http://i136.photobucket.com/albums/q...olinear_03.jpg

http://i136.photobucket.com/albums/q...olinear_04.jpg

http://i136.photobucket.com/albums/q...olinear_05.jpg

http://i136.photobucket.com/albums/q...olinear_06.jpg

I hope this cutting and pasting works...
Cheers amigos...

Scruge

Although you are correct that TNC and BNC are pretty much the same, the real difference is their fixator. The BNC is spring loaded using a bayonet/cam lock, that's held under compression typically with a wave washer or elastomer . Which opens the door to several problems, vibration and moisture/humidity. The TNCs are better suited in most apps when properly torqued. Both have pretty much the same usable freq range of up through 9ghz+ in a static application. However the nickel plating isn't best suited for freqs above 8ghz, so most companies will usually offer a silver plated version. Depending on manufacturer the insertion losses for both are .2 - .35db

The problem with RG-58 Mil17/28 isn't that it can't be used or that it isn't capable of handling 2.4ghz effectively. It's the fact that there are so many unqualified companies producing crap variations. If you can purchased from a reputable source in compliance to Mil spec, you shouldn't have any more problems with it verses any other cable in it's size range. A Mil spec RG58 cable will have losses in the <1db / meter, typ .75 - .85db @ 2.4ghz.
Even Times Microware doesn't list their RG-58 equivalent LMR-195 as being M17/28 qualified, they actually list a different product to meet Mil Specs. Which probably explains why their LMR-195 has better specs than RG58, because they are most likely over sizing something to get an edge. We used to call this "Specmanship".

You might want to record your signal readings as reported by NS in dbm without the antenna and then record them again with the antenna attached. The difference between the two readings should give you a clue as to your gain.

__________________
KNSGEM
A wifi boundary plotter for Google Earth
Click Me

wrzwaldo

With that bent up POS you have produced?

My previous comments stand as "intellegent" (intelligent) engineering opinion.

Barry

Okay, I can see it better now. I still wouldn't use RG-58. I'd also straighten out as much of the straight parts as possible. The bends will throw the signal slightly out of wack. To be honest I bet you could see the same two extra signals with just a piece of RG-58 and no antenna.

__________________
Never do anything you don't want to explain to the paramedics.

alkats

I am still in pursuit of building a simple vertical colinear antenna.

I stumbled on the assembly instructions of AEROLIX colinear antenna. Studying dimensions and comparing to other dimensions to the info available on the net, I came up with my version of colinear antenna. The idea was to make a single wire continuous design without use of tubes, soldering, etc.

I found a dry wall screw, held one end of wire with pliers and wound the coil after I marked the required length of 60 mm. I figured out that extra 1 mm will be added as I bend the wire. Because the screw has constant pitch all coils are predictable and easy to measure. Coils are 10 mm long, 5.5 mm outer diameter and have about 4.5 turns. After I made four coils as a test, I made a few sections of the colinear antenna.

I would like to know what people think. Please be considerate for my language and spelling because English is not my native language. Here are photos of what I made:

http://i136.photobucket.com/albums/q...AEROLIX_02.jpg

http://i136.photobucket.com/albums/q...AEROLIX_01.jpg

Please let me know if this is good way to build vertical colinear -- thank you.

wrzwaldo

Nope, that wire needs to be a bit more bent-up.