1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Wireless questions about antennas and Tomato.

Discussion in 'Tomato Firmware' started by Rocky Grim, Apr 30, 2012.

  1. Rocky Grim

    Rocky Grim Networkin' Nut Member

    Wireless Setup.jpg
    This weekend I modded my WNR3500Lv2, and added external antennas. The antennas I am currently using are only 3dBi. I wanted to see if they would give me the performance I needed. I figured that since they where external instead of the internal I would see a big improvement. I did see some improvement. However, the improvement I saw was mainly on the laptop that is positioned the furthest away. There isn't much of a difference in signal to the other devices.

    I am looking to improve the signal even more by adding bigger antennas. I'm wondering if I should go with a 7dBi, or would a 9bBi be over kill for in the house? The laptop is about 30 ft away from the router. However, it has three walls to travel through. Due to the walls I am leaning towards the 9dBi because I want to make sure I have enough power. However, I really don't like the idea of having antennas sticking out of the top of router a foot and a half lol.

    What I'm worried about with 7bBi's the signal width is 30%, and with 9dBi's it drops to 25%. I have attached a picture of the way my wirless network is setup. I'm worried that the beam will not be wide enough to hit the laptop infront of the router, and my Dish Network receiver that is positions diagnally to the front left. To give you an idea of distances. Laptop 1 is positioned directly to the front of the router, and it is no more than 10ft away. The Dish Network is diagnal from the router to the front left, and it is approxemently 25ft from the router with one wall. The distance between laptop 1 and the Dish network reciever is approxmently 15ft if that.

    Also, here are the antennas I'm looking to buy. It is the same guy I purchased the wifi pigtails from.
    9dBi's http://www.ebay.com/itm/New-Dual-Ba...936?pt=LH_DefaultDomain_0&hash=item20c4a30ac8
    7dBi's http://www.ebay.com/itm/140727221513?ssPageName=STRK:MESELX:IT&_trksid=p3984.m1555.l2649/
    6dBi's http://www.ebay.com/itm/New-Dual-Ba...564?pt=LH_DefaultDomain_0&hash=item20babb971c

    He doesn't sell 5dBi's.
  2. xtacydima

    xtacydima LI Guru Member

    remember, power doesn't come from the antenna, but from the amplifier inside the router. The antenna is the gain. Yes, with most of these new routers the antenna does not reach max gain (potential is higher) and a larger antenna will help. I just don't want you to think getting a larger antenna will mean better range, there is a cut off. I think going with the 9dbi you should be fine, this helped me a lot and I also got mine from good 'ole ebay.
  3. Rocky Grim

    Rocky Grim Networkin' Nut Member

    Hmm so there is a possibility that I could see no increase in signal at all because the amplifier inside the router is maxed out? I just loaded back up the Netgear firmware as a test to see how it does. On my Dish Network Hopper DVR it has an option to check the signal strength on a scale of 0-100. With Shibby and Toastmans Tomato builds I was seeing a signal strength of around 40. With the Netgear firmware it is bouncing around in the high 50's to low 60's. It looks like the 3dBi's I had laying around are doing more than I thought. If there is still some more power left in the router where I can benefit from an even bigger antenna I am thinking I should get the results I'm after. I'm just wondering if there's a good possibility I could see no more improvement regardless of antenna size due to the routers power being maxed out?

    Well, I have just noticed something with the stock firmware. I am getting full signal to all of my devices but, I'm only getting it because is reverted to 72Mbps. Here it will do that if other networks are detected. Therefore, it looks like to me that Tomato does a better job than we think at least when it comes to 2.4Ghz networks. I was at least getting a faster connection speed with it. This same behavior is probably what we see in Tomato from time to time.
  4. Toastman

    Toastman Super Moderator Staff Member Member

    Think of it this way.

    As you increase the power from the router's transmitter, there is a point where you reach a maximum and there is no further increase in output. At that point the only way to increase performance would be to have a bigger amplifier in, or external to, the router. But lets assume we aren't going to do that and we are stuck with the router just as it was built by the manufacturer.

    For example, if our router was running 125mW and we had a signal at our receiver of -30, we could double the transmitter power to 250mW. The signal at the receiver should increase by approx. 3dB and now it's -27. At this point we've reached the limit because the P.A. in most routers uses a PA chip whose maximum output is 250mW.

    So, what can we do if we don't want the hassle and expense of adding a bigger P.A. or "booster" on to the router? We can use a different antenna.

    Notice I do not say a "better" antenna!

    No antenna is inherently "better". They can be "better" for some clients, or some directions, but maybe worse for others.

    Antennas can increase our signal at some points by taking the RF energy available from the transmitter and concentrating it in the direction of the receiver we want to reach. To do this we REDUCE the signal in OTHER directions. That is how an antenna produces GAIN.

    There is no "free lunch".

    The antennas that screw into the back of the router are very simple. Directly above and below, the signal is weakest, and the energy is concentrated into a 360 degree horizontal donut shape. We call that the antenna's "radiation pattern". The bigger, or longer, the antenna is - the flatter the donut becomes, the higher the "GAIN" in the horizontal direction, the less the signal is radiated above and below. That's all it does. No magic here.

    So this kind of antenna is useful if the clients you want to reach are all on the same floor as the router. They will reduce the signal on other floors and use it to increase the strength on that floor.

    Therefore, for houses with 2 or 3 floors, using them may make the overall coverage much worse than it was before. But with many router antennas, you can of course swivel the antenna to a horizontal position where it will transmit upwards... and experiment with one antenna horizontal or at 45 degrees and the others in different positions. But it will be difficult to find a compromise.

    So what do we gain by making the antenna longer still?

    The real gain of these antennas, due to the manufacturing tolerances and the inherent losses involved in making antennas at these frequencies, is never anywhere near the silly figures claimed on the packaging. Those are theoretical, wildly exaggerated figures. In reality, you probably get about half of what is claimed. That is a VERY small amount and unlikely to make much difference to most of us. The step up to the 9dB antenna will result in even less change as the longer the antenna gets, the bigger the internal losses. To get a theoretical 3dB gain we have to double the length each time. But you don't actually get that theoretical 3dB change but more like 1.5-2dB at best and even this small increase is harder to achieve as the antenna gets bigger. There's a law of diminishing returns due to internal losses.

    There is actually a practical limit when making this kind of antenna where the losses become so great that there is no point in making them above a certain length, because the losses cancel the theoretical gain. Any bigger, and we see no benefits at all. For this reason, the supposed "9dB" antennas are usually only slightly better than the "7dB" ones.

    Couple that with the fact that many or even MOST of the antennas on the market are fakes, and even the good ones were never individually tuned for best performance, and you will see that it's a bit like Russian Roulette.

    In other words, you may get some improvement from these screw-in antennas, but it is not the huge leap that many people in the forums seem to be expecting. And if you measure the results for a significant period, by taking at least a few hundred measurements over a period of several hours, you will usually find that it wasn't really worth the hassle and expense.

    I personally wouldn't even bother because it's pretty much a negligible improvement for the money. The best and most stable way to get better coverage is with the addition of another wired AP on the other side of the house. An AP is cheap, a 100% solution, and very easy to set up. Add-on (booster) power amplifiers and preamplifiers are very expensive, and do not work nearly as well.
  5. Rocky Grim

    Rocky Grim Networkin' Nut Member

    Thanks for the very detailed reply Toastman. As always I greatly appreciate it. I was looking forward to a reply from you because I know you know whats up when it comes to this sort of stuff. The 3dBi antennas I am using are definatly making a big difference. I took them off and the signal reduced to hardly anything at all. Some connection were lost completely. I think I am going to try increasing the power a bit. I am still just using the default power. Then, check to see if the signal strength is improving at all. Then, measure it for a few days like you said over a period of time.

    The only laptop I am really having a problem with is the laptop directly in front of the router. I have both attennas pointing straight up. The router is ontop of my fireplace mantal (which is a fake fireplace so no heat just for looks) and the laptop in question is positioned on a glass computer desk about 10ft. infront of the router. After reading your reply I think the problem with it is the laptop is positioned downward so it is below the donut shape of the signal. Should I try lowering the router to the same level as all the computers though? I might end up moving that laptop to my sons room which is across from the furthest laptop. Therefore, I see a AP in my future.
  6. Toastman

    Toastman Super Moderator Staff Member Member

    OK on that!

    I don't believe that making any movement of the router like that will normally affect the laptop that is only 10 feet away. Thinking of possible other causes ...

    Of course, the wireless in that particular laptop may be faulty, have some setting that isn't correct, or the manufacturer's chipset or driver may be to blame.

    But there is another phenomenon that could explain what is happening. You've probably tested this already, but I mention it here for others to read too.

    If a client receiver is too close to the router, it may just be getting TOO MUCH signal from the router causing it to overload and the data rate to drop. On some chipsets this happens more than others, it is due to the design of the particular chipset in use. That is why I recommend people who are testing their routers to put them around 10m away. It explains why some people find their throughput actually DROPS when they increase transmit power, while others say the opposite.

    I have seen several laptops here that have overloaded while still 3-4m away from the router. But it's unusual. Certainly 1m away is dodgy, though.


    I know that my own opinions don't necessarily help anyone with their router problems, but Wifi is actually nothing to do with routing. Since manufacturers added it onto consumer routers they have become unreliable and they are getting worse, not better. That's why for serious use I never have wireless enabled on a mission-critical router. All of the Wifi needs are handled by wired AP's placed in the best position for good coverage of the premises. Here, for example, there are 28 AP's and 2 gateways.

  7. xtacydima

    xtacydima LI Guru Member

    Have you tried only with one laptop, maybe it's a low end mini pcie card in there. I always prefer the atheros cards personally, what brand is in the laptop?
  8. Rocky Grim

    Rocky Grim Networkin' Nut Member

    The laptop in question is a Compaq CQ57-319WM. It has a Ralink RT5390 in it. I have tried using the drivers from HP/Compaq, drivers from Windows update, and now the latest directly from Ralink. I have read similar problems with this network adapter. There aren't many options that can be set for this wirless adapter. Below are the possible settings. I have tried different combinations of settings with no improvement. The signal bouces all around. Even when it is showing full signal most of the time the connection speed is very low. I'm talking like in the 20's or lower.

    Adhoc support 802.11N
    Multimedia/Gaming Enviroment
    Radio On/Off
    Roaming Sensitivity
    1. Highest
    2. High
    3. Medium
    4. Low
    5. Lowest
    6. Disable

    I have managed to get my hands on two Netgear WNDA3100v2 network adapters. I also have one Netgear WN111v2. I plugged the Netgear WN111v2 into the laptop in question, and it seams to be doing the same thing. Therefore, I believe Toastman is right about having a wireless device to close to the router. Maybe since the signal is good enough to the distant laptop I could try reflashing back to Tomato, and lowering the power some to see if this makes any difference?
  9. xtacydima

    xtacydima LI Guru Member

    I stopped reading after I saw RA Link. Based on my personal experience, this is one of the worst wireless NIC's to have. This does not mean it's the cause of your particular problem at hand, but if I may suggest you try a different brand card in the laptop and see if it gives ou better results. Intel is stable and atheros is hard to get (seems like ebay only nowadays, sometimes newegg).

    I just hope your BIOS isn't locked to an approved whitelist of cards... although there are work arounds for that as well.
  10. Rocky Grim

    Rocky Grim Networkin' Nut Member

    I disabled the Ralink. I plugged in the WN111v2 Netgear adapter instead. I downloaded the latest driver from Netgear for it. It doesn't have any special settings you can modify for it. Therefore, by default it should be ready to rock n rool for wireless N up to 300Mbps. Well, so far I am seeing full signal strength (5 bars) on the device, but it is only reporting a wireless speed of 54Mbps. Tomato reports a quality of 63 with an RSSI of -36 fo this device. ***Update I managed to get it to break a wireless speed of 200Mbps. It was like 216Mbps. What solved the problem was the WN111v2 does not like afterburner.

    For the laptop that is the furthest away ( around 30ft with 3 walls) I used a WNDA3100v2 for it. It had like a million extra settings that coul be set from the driver page. I looked the ones I didn't know up through Google. The signal strength on the device is bouncing between 4 and 5 bars. The wireless speed on this device is in the 200's. Somtimes it is as high as 243Mbps. Tomato reports a quality of 39 with an RSSI of -60.

    My Dish Network Hopper also has the WND3100v2 installed. It is closer to the router than the laptop above. It is only around 20ft with one wall. However, the fastest I have see it connect is 130Mbps. I think it could be the Dish Network reciever that is limiting the wireless speed somehow. However, if it was only able to produce speeds of under 150Mbps you would think they would distribute Lite N adapters instead at 150Mbps. Tomato reports a quality of 53 with a RSSI of -46 for this device.

    I managed to correct the problem of the furthest device to gain two additional problems with the closest two devices lol. I don't think I can win! I'm not sure what else to even they at this point. *** Now I am trying to get the WNDA3100v2 connected to the Dish Network reciever to achieve a wireless speed higher than 130Mbps.

Share This Page