Hawking 15dbi Corner Antenna - What am I doing wrong?

So that's why CompUSA was selling that antenna for $10.

Wow . . . that doubles the pain. CompUSA got $49.95 from me for the POS.

jm

I was hoping to make wireless work because the building contains only category 3 wire, and the architecture makes further wiring/rewiring impractical. I've used cat3 before for internet-only networks, and I can do it again here by robbing some spare pairs. It's just a jack-leg way of doing it.

Can you believe the cabling folks used ALL category 3 wire . . . in a large prominent church built less than two years ago??

jm

CAT5e costs me $58/1000ft while CAT3 is listed at $75/1000ft. In other words, since CAT5e is much more common, it's cheaper. Sigh.

CAT3 can be used for 10mbit/sec 10BaseT-HDX (half duplex) networking quite effectively. Just don't try it at 100BaseTX. It may require some trickery, such as using a managed switch or ancient 10baseT hub to force everyone down to 10baseT speeds, but it will work. I've used much worse cabling than CAT3 for 10baseT and survived. Same with 25 pair telco bundles.

If you run out of pairs, you might look into phone line networking as in:

If you have RG-6/u CATV coax cable runs, there are ways to make that work also:

I just received one of the Hawking HAI6SIP omni antennas and gave it a try on a Linksys WMP11 pci card. It worked about as well as the stock rubber duck antenna. There's a long thin cable on the Hawking antenna which just about negates the 6 dB antenna gain, so you end up about where you start. But you can move the Hawking antenna around more than the rubber duck so that might help. I might try shortening the Hawking cable when I get the urge to destroy.

Like Clarence says, the Windsurfer reflectors are good. I was able to double my range with one on a Netgear MR814v2 router.

Bruce

Wow... long paragraph.

The setup:

Okay. This would be a good place to test antenna improvements. Marginal signal can be improved and observed.

Does this location work at all with the stock antenna? Do you see that the

6dB omni on the router makes a difference at this location?

If there is no signal, we don't know yet if we need 6dB of improvement, or

100dB of improvement. It might not be possible to work at that location, due to some unknown factors.

So the antenna helps, just not enough.

What does "Swapping" mean? Which various trials did you do? If you went from no signal to some signal, you don't know how much improvement you have.

Did you try the 15dB antenna in the middle location? Did it improve the signal?

Did you try the 6dB and then the 15dB at the router, measuring the change at the middle location?

The 15dB directional antenna might need to be pointed somewhere other than what you expect. Did you try rotating and angling in all three dimensions? In my layout, There are two exterior walls and about 50 feet between the router and my laptop. A directional antenna on the laptop doesn't get its best signal pointed directly at the router. I can point it toward a hallway, through a window that seems out of line, or at a wall toward a window in the other room. Straight through two walls is worse.

Rotating the antenna on its horizontal axis also makes a difference. I get better signal with the antenna horizontal than I do vertical.

It would be good to know if the 15dB directional antenna is better than the

6dB omni. If it isn't return it and get another.

A directional antenna at each end would be a good thing.

I have used the "Hawking HAI6SDA Directional 6dBi 2.4GHz Antenna"

with good success.

I have used the freeantennas.com Windsurfer EZ-12 with good success.

I can use my windsurfer equipped Netgear router through three or four exterior walls to an outside location 100 feet from the router to a laptop that has no additional antenna.

If there was some volunteer work involved, yes. I see well intentioned folks who donate things that would be better left in their garage.

3-4 walls! What are they made from, ferro-concrete or with some other metallic material in them that is acting as a shield. It seems as if you are quite lucky even getting sufficient signal into the first room.

It would seem as if the router should be located somewhere in the three offices where it will not be shielded to the same extent. It may be easier to ethernet the three offices back to a point where there is an acceptable signal but then perhaps the complete link should be ethernet and not wireless anyway. There are some situations where wireless is really not practical and you may well have one of them despite good antennas etc.

Peter

CAT 3 may not be the latest and greatest but it will work just fine for an office network unless they are moving giant files around.

Very perceptive. In fact, "some volunteer work" *was* done in this case. The guy is a long-time professional, and this was his last job prior to retirement. In his defense, the church decision makers never imaged a true data network, and the wiring job is first-rate - very clean and efficient. My beef is simply that he didn't use cat5/5e anywhere.

jm

Sorry about that. Thanks for muddling through it and posting such a thoughtful reply.

The first location is working with the stock antenna. Yes, the 6dB omni does make a difference of the stock antenna on the router. The signal here drops by about 1-2dB when the corner antenna is on the router. However, as will be clarified below, I have not exhausted my options for positioning. I really only tried the corner in one position. I guess I assumed the direction for this antenna to be obvious because of its design.

I made a mistake in my original post. Of the two "bad" office locations, one (the farthest away) is using a Dlink wireless pci NIC. It, when used with the corner antenna, will achieve approx 25% connection on the Dlink utility, but it's not consistent, and the signal will drop as people move in/out of the office and through the hallway by the door. I'm confident it can be made to work. However, the other office, which is one wall closer to the router, is using a Hawking pci NIC. It will connect at the "bad" rating (on the Hawking utility), but I cannot pull an IP address. It was late, and I did not measure the signal level at that location.

Correct.

By "swapping" I mean that I tried the putting the corner antenna on the router and the omni on the clients and v/v. At least as far as my own implementations, the overall performace decreased anytime the corner antenna was on the router. In other words, it *seems* clear that the 6dB omni is the better choice to be used with the router.

The 15dB (corner) antenna performed about the same as the stock pencil antenna *when attached to the router*. However, it seems to be a slight improvement over the same antenna when attached to the client.

I did not measure the difference between the two antennas in the bad offices. I was running out of time and simply trying to establish connection.

This is all GREAT input, thank you. You have given me a lot of ideas. So much so that I'm going to go out there today (Saturday) and try to work this thing out. I will report my experience. In the meantime, any other thoughts you have are welcome. I really appreciate your help. I've been in the telecom/data business for 7 years, but I am a noob in the wireless stuff. I desperately need to flatten out the learning curve. Thank you again.

jm

That was me. The tiny coax is either RG174 (1.2dB/meter) or RG316 (1.1dB/meter). As I recall, they usually came with 3 meters (about

10ft) of either coax, which is about 3dB coax loss, plus about 0.5dB connector loss. That would give the 5dBi antenna a net gain of about 2dB, which is about the same as the stock rubber ducky antennas. There's no improvement over a rubber ducky antenna, but as you mentioned, there is an improvement in the location of the antenna, which makes a *BIG* difference.

The old and rare stuff always costs more. I suspect the electrician had some lying around from ten years ago. That made it free for the church, a good deal.

And, as you note, cat3 was fine for 10BaseT, which would be just as fast as a poor quality 54g connection, and better than an 11b connection. Much faster to the office that doesn't work at all.

I just returned from the church after 2 hours of trial and error. I moved my two antennas around, changed positions, etc. The most striking revelation is that these Hawking units add very, very little to the equation. With the 6dB antenna on the router and the 15dB corner antenna on the client, I was able to achieve a continuous 26%-32% signal (using the Dlink utility) at the farthest office. This was after repositioning the router/antenna in the equipment room many times. That strength of signal is just barely sufficient for internet usage, and certainly unacceptable as a long-term solution. For comparison, I replaced the 6dB antenna with the stock antenna on the router - leaving the 15dB corner antenna on the client. I was hugely disappointed (in one sense) to see that the signal dropped only a couple of percentage points (on average), not enough to effect the link in any significant way. In other words, the "+6dB" antenna is virtually no improvement over the stock antenna. In fact, it seems that any performance increase has more to do with being able to move the 6dB antenna around slightly by virtue of its 15" cable and angle adjustment. I don't think the antenna itself adds one bit.

I believe the same can be said for the corner antenna. I'd like to know how Hawking arrives at this "+15dB" rating. I simply don't see it - or anything close to it - and I'm in a real-life situation trying every possible position and combination I can come up with.

Can these free antennas, such as the EZ-12, be used to enhance either of my Hawking antennas. I really think I'm close to making this work, if I could just get another consistent 10%-15% signal at the clients. I've got a customer who's been using a similar Dlink setup (sans Hawking units), who never gets better than 40% on the Dlink utility, but they're very happy with things. The links don't drop, and the speed is sufficient.

If I can't achieve 40%+ in these two offices, I'm likely going to steal two pair off a phone extension and hang an AP in one of the offices. It's probably what I should do anyway.

jm

"JM" wrote in message news:8wG1f.1461$ snipped-for-privacy@tornado.ohiordc.rr.com...

You're correct, and I've decided to use one of the home runs to hang an AP in the area of the 3 offices. I would feel much better having a hard line to that part of the building, and cat3 will do just fine, as they really only need internet access anyway. So the weak link won't be the cable. The rated 10Mps for cat3, or anything near it, will easily accomodate their roadrunner broadband speeds.

jm

Someone suggested that the skinny cable lost about as much signal as the antenna gained. I think that is true of the mag mount 5dBi antenna that I have for my laptop. There isn't any more signal, according to Netstumbler, but there are a couple of places where it works, and without it doesn't.

Here, the cable shouldn't be so much of a factor. 15dB is a lot. I saw some other respondents say the 15dB Hawking was junk. I've only used one product from Hawking, the 6dB directional, and it helped a lot, but then it was from a PCI card where just moving the antenna out from behind the computer might have made a difference. I didn't spend much time with it. I think it was $10 after rebates. I plugged it in, and it was good.

No. All of the freeantennas.com templates are designed to fit onto a stock rubber duckie dipole (except one).

You could put the Windsurfer or two onto the router. A pair of them should give better gain than one 6dBi external antenna. Did you remove both rubber duckies when you added the Hawking antenna? I get very poor results with one duck and one reflector on an SMC router.

You could put windsurfers onto the DWL-520 stock antennas. The location behind the PC is still a problem, though. As I said, I had success with the "Hawking HAI6SDA Directional 6dBi 2.4GHz Antenna" The link I gave earlier didn't work. Sorry, I didn't check it earlier.

is the proper link, today, for the 6dBI directional.

Since all your gear is 11b, there would be no loss of performance in running 10BaseT on the cat 3 that you have, verses a perfect 802.11b link.

What is the feed, anyway? If it's DSL, move the DSL modem into the office with the three computers. Same with cable. Why is the router in the room that it is in? Maybe relocate all the goodies near the three PCs and run

10BaseT back to the other room if you need it.

The DI-614+ I'm using is the rev b or c, whichever is the single antenna model.

Well, that's sort of another issue, too. Originally I had not planned to use the 614, but rather a DI-524, which is 802.11g. However, as I stated in a post above, that thing is terrible. It is much, much weaker (as a transmiter) than the 614. Rooms where I get 30% with the 614 will drop to

10% using the 524. Huge disappointment. I just assumed that Dlink would retain their transmit power as the products evolved.

It's roadrunner cable modem internet, and it was terminated in the same small equipment room where the telephone and alarm systems are. That's where all the pipes and conduit enter/leave the building. Unfortunately, the cable modem has to stay where it is.

However, after today's trip I've decided to use two pair of the cat3 that terminates into one unused phone jack. I'm going to put a data jack in and hang the router there.

jm

I think so. None of that other equipment appreciates having good access to the cable modem. That feed is only ~3Mbps anyway. Using existing cat3 to run from there back to where the computers are sounds like a good deal.