Wireless antenna question

Not really, that's only a 3 dB increase. It takes a SIX dB increase to double your range. A 3 dB increase will help a little big, but not all that significantly.

Besides not to get "higher" signal levels, signal level fluctutas very quickly, but doesn't drop to low signal levels when compared previous

No comment more? There's no router and adapter positioning change but also signalling level is still same maybe worse maybe same, but definetely no signal level increasement after upgrading from 2dbi to

Well, welcome to the real world of antenna design. I'll hit the highlights.

1. Your theoretical gain increase is allegedly 3dB. That's good for a 1.4 times increase in range. You probably won't notice it much.
2. The "5 bars" signal indicators are very granular. I was going to measure the signal levels that correspond to each bar in a probably futile attempt to achieve some semblance of calibration. It's about as futile as doing that on a cell phone. I have one phone that shows 3-5 bars for everything except a total loss of signal, but garbles calls regularly. I also have a PDA phone, that shows zero or 1 bar regularly, but works just fine at those levels.

In short, the number of bars don't mean much and certainly not for comparing different client radios or antennas. I'll make a very bad guess that the 5 bars cover the dynamic range of the receiver. That's about an 80dB range, which makes one bar worth 80/5 = 16dB. You're not going to see a 3dB change when each bar is worth 16dB.

1. The bars may not even measure signal strength as RF level. Some chipsets use the bit error rate as an indication of signal strength. It's suppose to be an indication of signal quality, but they just take the baseline noise level, add the signal quality, and call it the signal strength. Saves on hardware and CPU cycles, I guess(tm).
2. Measuring antenna performance indoors is also futile. There are more reflections than signal. Same with going through walls, corridors, and close to the ground. What you're actually measuring is the RF reflective quality of the room, not the direct signal. I was running antenna tests in my palatial lab, the swept gain results were varying +/- 2dB by just me moving around about 1 meter from the antenna.

Those lovely symmetrical gain plots you see on the antenna manufacturers web piles are nothing even close to reality. The gain pattern is affected by literally everything in the area. Give a single antenna to several test labs, and you'll get ridiculously different patterns and result. Your two antenna have a max gain of 2 and 5dBi only on a simulator.

1. Antennas are also very frequency sensitive. This is normally not a problem with low gain 2 or 5dBi antennas, which have a fairly flat VSWR curve. However, the higher gain antennas (24dBi) often barely cover the 83.5Mhz of the Wi-Fi band and are often quite different on low and high channels.

However, the low gain antennas offer another complication. They are very crudely constructed. Look inside and they are mostly coax cable and brass tubing. The driven element is commonly just the exposed center of the coax. When 1Mhz is only 0.052 mm, it's fairly easy for production variations to cause problems. I've tested a box of allegedly identical

1. I suggest you give up on your quest to find the ultimate omni antenna. They have their place but directional antennas and reflectors work better if you can tolerate the directionality. I'm also a big fan of homemade AMOS/Franklin sector antennas:

Hi Mr. Liebermann, Thanks for the tips and experiences, actually i've got it to get rid of wireless drops although the connection is seemed as "connected" in Windows taskbar or my Broadcom's utility, but it didn't help, still having drops if i move notebook away from router, the shown speed is about 24mbps or 11mbps and it drops although seemed as "connected".

However i haven't changed adapter or router, only changed antenna from

Ok, now we know what you're trying to accomplish. I'll assume a Windoze XP laptop and some sort of internal MiniPCI card. I'll pry the details and the router model out of you later.

Try this simple test. Run continuous ping from your laptop. Don't move it during the test. Also try it with various antennas. For Windoze it's ping 192.168.1.1 -t What you should see is delays of about 2-5 msec. Less if you have a decent router. The actual value is not terribly important. The variations between ping times are an indication of how reliable your connection is running. If you get larger variations in ping times, you might have a interference problem, sick router, old firmware, buggy driver, or as I found in another similar thread this week, it's time to reboot your computer.

If you're really ambitious, you can use IPerf to determine your level of impairment. Setup a 2nd computah, with a direct CAT5 connection to the router. Run the server part or Iperf as: iperf -s On your wireless laptop, run: iperf -c ip_address_of_server and see what manner of speed you get. It should be about half the wireless connection speed. Anything less or highly variable means you've got a problem.

Whether a better antenna will solve the problem is dependent on the exact nature of the problem. Go find the cause. Then we can discuss the solution.

See:

for the thread on analyzing ping results. If you get something awful, like the initial message shows, you have a problem, somewhere.

Try putting a reflector on either of the antennas. That will help more than going from 2 to 5 .