Slashdot Mirror
Ask Slashdot: Why Does Wireless Gear Degrade Over Time?
acer123 writes "Lately I have replaced several home wireless routers because the signal strength has been found to be degraded. These devices, when new (2+ years ago) would cover an entire house. Over the years, the strength seems to decrease to a point where it might only cover one or two rooms. Of the three that I have replaced for friends, I have not found a common brand, age, etc. It just seems that after time, the signal strength decreases. I know that routers are cheap and easy to replace but I'm curious what actually causes this. I would have assumed that the components would either work or not work; we would either have a full signal or have no signal. I am not an electrical engineer and I can't find the answer online so I'm reaching out to you. Can someone explain how a transmitter can slowly go bad?"
and worn out. Also I think they have pretty short life spans.
As all of your neighbors add wireless routers, the noise floor goes up, and the usable signal goes down, even though the signal strength is the same.
built in failure. bow to your corporate masters and go consume.
Over 3 years I'd imagine a greater density of wifi devices all sharing the same spectrum to have appeared. Perhaps the signal level is the same, but the noise floor has increased substantially, degrading performance.
Are you sure that it's the router itself? Have you made any changes to your house in the last 2 years?
I think the problem is the wireless pollution that is spreading. Few years ago your friends were the only one with a wireless router in the surroundings. Now more and more devices are connected and this is limiting the signal propagation of your devices.
L.
Are you sure it's signal strength (i.e. dB) or network bandwidth interpreted as "bars"? What I've noticed over the years is that the more wifi devices we add, the slower the network is, which makes sense. It could be cheap components in the amplifier section that are getting hot and degrading over time, but that's just a WAG.
I'd suggest that what's actually degrading your signal are all the neighbours who are also using the 2.4GHz band. It's not just WiFi, but a whole slew of other wireless gadgets. Move to the 5GHz band whenever possible, there's a lot less congestion.
My wireless router is ~6 years old. Works great.
The one I bought my parents is 7 or 8 years old. Still works fine.
Perhaps your old routers are just dealing with more interference as more of your neighbors are buying overpowered devices that take up more and more channels. And now you've become one of those offenders as well since you've bought yourself an overpowered device that takes up too many channels.
It could be the noise floor going up near your house, or just planned obsolescence.
Maybe there's too much people using the same channel as yours.
Check with one of thoses apps :
Wifi Analyzer on android
This one is great and work on Windows + Java : http://tools.meraki.com/stumbler
I also wonder if the quality of our devices may just SEEM like they are getting worse, because of new networks that appear in the area as time goes by. It seems that more and more of our neighbors have installed wireless networks after ours was installed.
..have a tendency to degrade and fail over time.
It isn't that the wireless router / base station is degrading. The signal quality is degrading because there is more noise from other Wi-Fi access points and devices in the area. There is a finite amount of spectrum available and when you have lots of home base stations in the same area/neighborhood the interfere with each other leading to lower range for each of them. It is similar to talking in a noisy room. In an large, empty room you can hear some one from across the room, in a noisy room with many people talking you may only be able to hear someone a few feet away.
When I installed my first Wi-Fi base station in 1998 in my parents house I could walk down the block and get reception a few houses down. Today there are dozens of base stations and potentially hundreds of devices on the street. I struggle to get consistent reception within the house despite improvements in antenna design and b/g/n routers.
Obviously the magic smoke, although not released suddenly, does gradually leech out of the components leading to loss of performance over time.
"I bless every day that I continue to live, for every day is pure profit."
Could the analog components of the amplifier/filter circuits be degrading? If capacitors are leaking, etc then that would definitely make the performance decrease but maybe not enough to completely stop working.
You should consider another option: older equipment may not have firmware as good at dealing with congestion (802.11N helps with this), or maybe the new box has 5Ghz which has much less interference issues? Maybe the real degradation was the neighbors installing access points? You may also have had certain pieces of gear installed that interacted badly with your access point (some of them have really awful firmware or very loose implementations of the standard).
These are just guesses... I haven't personally had any degradation except for interference in the 2.4Ghz band. When I bought this house devices would only detect my network and maybe one other. Now seven show up. Interference isn't just a problem in apartments anymore.
Natural != (nontoxic || beneficial)
Probably capacitors degrading, transient spikes making it through the hardware, electrostatic discharge during assembly, just plain overheating as dust coats the internals, more people using the same frequencies (possibly including yourself as you add more wireless devices), and a bevy of other reasons I can't think of at the moment.
This is a hypothesis based on peripheral involvement with analog and digital RF at 0.5 and 1.5 GHz for twenty years.
AFAIK, the output stage of anything broadcasting above about 2 GHz has to be analog, with the lower frequency signal mixed into a carrier at the higher frequency. Digital synthesizers and chips which can deal with 1.5 GHz directly are still very expensive and are unlikely to be used in the consumer routers. So the final output stage is likely an analog RF transistor.
Analog transistors change characteristics with age at elevated temperature, where elevated is anything over 20C. Implanted ions diffuse with time and temperature, changing junction characteristics. The small structures required by high frequencies are more sensitive to such things.
The actual transmitter and receiver is nearly indestructible since it's mostly a copper wire wrapped around something inductive. The electronics are also operating at a power level where even though they operate at a rather toasty temperature, it is well below the danger point where silicon might degrade, and that'd likely brick it rather than have a slow decline.
Most likely, the cause is RF interference from your neighbors.
Check the power supply. Usually the electrolytic capacitors are already dry.
Not sure about wireless gear, but some devices (e.g. printers, light bulbs, fridges) are designed to break after a certain period of time, so that you would buy a new one.
The first thing that comes to mind other than they shouldn't lose strength that quick, is cheap, or faulty components and primarily capacitors. A batch of Samsung monitors are famous or infamous for capacitors going out in the power supply. Router builders could have come up with part of that batch as well. Another is low voltage, as they all use "wall warts" AFAIK. Try a different PS before pitching the router. QC and quality of the generic wall wart is not exactly for lab standard work. Normally and I have to emphasize the "normally" capacitors last on the order of 20 years...or more. In the last 15 years (give or take) the only router failures I've had were due to lightning and one that appeared to lose its brains. Most of the "stuff" here is in two buildings, one of which has the entire interior constructed of bonded barn metal so the placement of NICs or relays is important. IE the router is in the basement and the NICs are in the shop (pretty much shielded except for windows) some 130 foot distant. I have to admit that most of the time I use CAT6 instead of wireless as large backups take far too long on wireless.
I've been using wifi instead of ethernet for about 7 years now. Almost all of the NICs/APs I've used have displayed this problem with time. It's as if the equipment somehow develops creeping signal attenuation. My guess is that it's something relating to capacitors gathering a slow overcharge of some sort, causing them to block current in a growing fashion - I seem to recall this being possible from my early days of electronics studies.
Anyhoo, I fix the problem by simply switching the equipment to another channel, say, 3-4 steps away, to make sure the frequency some of the components will be switching at will be notably different. So far it has worked with all equipment I've had this problem show up on. After a while the signal attenuation develops on the new channel as well, upon which I simply switch back to the one I used before. Rinse, repeat.
Semiconductors degrade over time, the hotter the faster. At 25C normal digital semiconductors have 30-50 years lifetime. (I have observed this myself with a batch of 25 network cards.) Halve that for every 10C. Now, this applies to the individual transistor. Even if just the RF power amplifier runs hot, it degrades at roughly this rate. As this is analog, not digital, degradation does set in in an analog fashion, and may even go much faster.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
In my experience power adaptor degradation is the main culprit. Over time the adaptor will provide lower voltages and a less stable current. This translates into a lower signal output and higher noise respectably. I've seen bad adaptor turn repeaters into signal jammers - trust me, that was not an easy issue to troubleshoot...
I've never seen nor heard of such a thing. I've got WiFi gear, some of it over ten years old and I have yet to see any evidence of "signal degradation".
Because of this article, I have reexamined several pieces of equipment, some still in production, others have been on the shelf for a while, al demonstrate signals that are within original spec.
So, Mr. Electrical engineer, can you show us any other observations to support your assertion? Can you show us any evidence at all, other than your suspicious anecdotes, of signal degradation with age? Finally, please name names. It might be believable if you were using obscure gear, but if you are talking about the usual suspects in consumer WiFi, I call BS.
It could that the routers were made from the 1st generation of ROHS (mainly lead free) components. There was a issue with Intergrated Circuits where the metal traces inside the chips would migrate causing thinning of the traces (changes circuit characteristcs) or in some cases growing 'wiskers' that would cause shorts or in RF, radiate power at the wrong point.
My experience says that this seems to be in the past as manufacturers have made new fabrication techniques.
The loss in performance could be due to the solder between components (mostly between the antenna and circuit board) is degrading overtime (this happen a lot with industrial devices), also the diferents components as capacitor and resistor could be wearing out too.
1. slow burnout of emitter gear due to thermal degradation (yes, clock chips and transistors get hot, as do solder tracks and joints). Thermal runaway can occur if a solder joint fails and arcs, or overvoltage causes signal tracks to vapourise.
2. ionising radiation, particularly on unshielded components such as antenna conductors (I've seen something like this occur on an externally mounted amateur radio antenna: the sunward side of the antenna completely degraded, the result being that the only signals received (or sent) were on the shadow side).
3. component quality on consumer gear is not as stringent as it could be. Components can and do fail, and considering the number of components in a lot of consumer gear, it's a wonder any of it actually leaves the factory.
4. the noise floor of several years ago was far, far lower than it is now. The ERP of newer gear is (by design or by necessity) higher than older gear as more and more transmitters have to share the band. As a result, the signal quality taking a dive may be at least partly illusory. The equipment may actually be perfectly fine.
5. parasitic structures in semiconductor packages may be the catalyst for failure, either immediate or delayed. Such structures may be as small as a single atom of chlorine embedded in a crystal of germanium - innocuous at first (undetectable, even), but over time and use, that contamination will alter the chemistry of the semiconductor, possibly causing it to bond with the package material and rendering it useless. This might not even be an issue in high powered gear like regulators but in something like a microprocessor, it's a showstopper.
Operation Guillotine is in effect.
It might simply be that the old D-Link 802.11b router I used to have was so slow already that I wouldn't notice... but, in any case, I never noticed it with that old router, nor have I seen this issue with any of my Airport devices (I've still got an older teardrop Extreme chugging along, serving phones and such that can't handle 5GHz 802.11n).
Are you sure it's not simply a matter of more wireless-capable devices accumulating in your home over time?
#DeleteChrome
Analog components drift in value over time, and circuits with poor compensation are cheaper than circuits with good compensation.
Radios do not transmit at a single frequency, they transmit in a narrow bell curve centered around the nominal frequency. So as analog components age and the nominal frequency drifts, the frequency that the receiver is tuned to will no longer align with the peak of the transmitted energy.
On the other hand, receivers should have phase-locked loops that allow them to tune themselves to the peak of the received signal. Problem is that with several radios on the same frequency (assuming you don't have just one access point and one remote device) the access point can't optimize for the frequency spread of several devices, and might have
Wifi uses 14 channels, as a rule of thumb a signal pollutes two channels to the left and right.
Thus a strong channel 3 wifi signal can block channels 1 throug 5.
The wifi routers around me seem to change their broadcast channel every few months but are now focust on 6 and 11.
Use a wifi tool and manually set your channel to the least used one.
As a non expert this works well for me, but there must be smarter ways than this.
With adaptive rate and power control, I wouldn't suspect the output path first. I would look to the input path first. That's the delicate bit.
I should use this sig to advertise my book ISBN-13 : 978-1501515132.
It's by design. Especially if those devices are marketed for the American market. Wanna know what else is designed to fail after a certain set time?
Well, Microwave ovens, cars, especially those from one once big American car company, that recieved millions in bailout cash under this president.
In industry, it's called Planned Obsolescence, and Americans are pioneers.
Here's a write-up about it.
Capitalism at its best!
I've a Thinkpad T60 laptop and the WiFi adapter just seems to work fine, even after 6 years of use & abuse. Yes, I've replaced two WiFi routers.
Eating away at the PCB!
Obviously Opera's bees have taken residence in the walls of the house and are absorbing gamma radiation from the wifi antennas.
There's these little tiny critters called electrons, right?
And these tiny boogers, these electrons, go through these here digital circuits, also quite tiny, and their movement at that micron level eventually wears the holy hell out of those pathways they travel, causing pitted chips at the micron level --- easily observable with a high-powered electron microscope and other instruments.
I guess they don't teach science in them thar schools anymore, huh?????
And for your further edification, sir:
http://www.scribd.com/doc/110039863/A-Lawsuit-Against-Private-Equity
Flash memory fragmentation is a problem with lots of older routers (and maybe newer ones too, I'm not sure). A fragmented environment can cause all kinds of degradations from poor wireless performance right up to ethernet ports failing and the entire device failing beyond repair (I have experienced this).
There is some good info at http://www.routertech.org/firmware-faq/
RouterTech is a good site for networking info and also offers a FOSS Linux based replacement firmware for Texas Instruments AR7 based ADSL modem/routers.
"Q. I have heard about fragmented flash memory (or environment) in routers. What is this, and how do I deal with it?
A. Flashing firmwares, saving configuration settings, and doing stuff with environment variables all involve writing to the router's flash chip. Over time, the flash memory (particularly the area holding the configuration information and the router's environment variables - i.e., the first 10kb of the router's mtd3 partition) can become fragmented. If this happens, you can have all sorts of problems. The most common ones include not being able to upgrade your firmware successfully via the web interface, not being able to save your configuration settings successfully, and routers bricking themselves spontaneously. This is a major issue on routers with the Adam2 bootloader, which is seriously broken. In our experience, it is particularly problematic with DLink Adam2-based routers - but all Adam2-based routers suffer from this problem, because the bootloader cannot defragment its environment properly, except manually from the Adam2 bootloader command prompt itself. The PSP bootloader, on the other hand, does the job pretty well by itself. "
N networks seem to be very frail to interference. G might be a bit slower, but they seem to be a bit more robust.
Perhaps it's frequency drift. As components age their values change slightly. And when dealing with 2.4 Ghz and above tolerances are strict. It's just my guess.. Take it or leave it.
dissipation path in RF power amplifiers? I don't know whether common wifi routers have discrete RF power amp devices to feed their antennas, but if they do, they are probably connected by thermal grease to a heat spreader of some kind. Over years of thermal cycling, thermomechanical expansion can create voids in the thermal grease, which will increase the device-to-ambient thermal resistance. The device will run hotter (= lower efficiency, less RF power to the antenna) or fold its power back to stay within thermal limits (same result). This could also happen if there's lots of accumulated dust on the heat sink, which would increase device to ambient thermal resistance. My 2 cents.
That won't work because moset people are too stupid to change channels on thier routers, I bet 90% of all routers in residencial use are on the same channel, which ever one is selected by default in the firmware. So if YOU change channels you might be on a clear channel!
This is the correct answer. There needs to be a mod rating about 5, because this is the first person to post it.
While I don't have too much trouble with my routers (I have two in my house since one can't cover the whole house) I DO have issues with my garage door opener. Most of the time I can open it from the street, but sometimes I have to hold the transmitter right next to the door before it will open. It's almost as if a phantom jamming station goes on the air at specific hours of the day! The batteries in the transmitters are up to snuff, and one transmitter is buit into our Dodge Caravan and runs off the car's electrical system. All of the transmitters have the same problem at random times.
It may be that a capacitor on the router itself or inside the power supply went bad.
My first guess would be corrosion or some such, impeading the transmitted signal. If the connection loosens, or becomes corroded the SWR will increase because of the impeadence mis-match. The antenna is no longer ressonant so only a fraction of the full transmit power is actuallay radiated over the air.
vry 73
1. slow burnout of emitter gear due to thermal degradation (yes, clock chips and transistors get hot, as do solder tracks and joints). Thermal runaway can occur if a solder joint fails and arcs, or overvoltage causes signal tracks to vapourise. 2. ionising radiation, particularly on unshielded components such as antenna conductors (I've seen something like this occur on an externally mounted amateur radio antenna: the sunward side of the antenna completely degraded, the result being that the only signals received (or sent) were on the shadow side). 3. component quality on consumer gear is not as stringent as it could be. Components can and do fail, and considering the number of components in a lot of consumer gear, it's a wonder any of it actually leaves the factory. 4. the noise floor of several years ago was far, far lower than it is now. The ERP of newer gear is (by design or by necessity) higher than older gear as more and more transmitters have to share the band. As a result, the signal quality taking a dive may be at least partly illusory. The equipment may actually be perfectly fine. 5. parasitic structures in semiconductor packages may be the catalyst for failure, either immediate or delayed. Such structures may be as small as a single atom of chlorine embedded in a crystal of germanium - innocuous at first (undetectable, even), but over time and use, that contamination will alter the chemistry of the semiconductor, possibly causing it to bond with the package material and rendering it useless. This might not even be an issue in high powered gear like regulators but in something like a microprocessor, it's a showstopper.
No facts, please. We're all happy complaining about signals, noise, and the general decline of the human race.
Quartz Extreme and Core Image. Are there any other real reasons to spend all that money on generic hardware?
While I haven't noticed signal strength problems per se, I have noticed that routers seem to be particularly prone to creeping death syndrome, where they slowly start behaving increasingly erratically with disconnections, speed fluctuations and the like starting out at low levels, and increasing slowly over the course of many months. And I've noticed this even on wired connections...
Nobody seems to have yet mentioned tin-whiskers.
Tin atoms in solder under voltage migrate to sharp points, leading to the growth of very thin whiskers. These whiskers eventually can bridge components leading to shorts, and they also introduce noise to the electrical system as they change its reception characteristics.
The technical solution to this problem is to add more lead to the solder or to use higher temperature solders which don't include tin. Higher temperature solders are more expensive to work with, while lead has (more or less rightfully) become a boogeyman metal of late, so the problem of tin-whiskers has been increasing in recent years.
Haven't seen much mention of it but you might wear out the radio by running it at full power. You can also damage it by running something with too hot a signal. So for example if you are running it in the same room as your laptop, and it doesn't reduce its power level for transmitting, a -30 or even -45 dbm signal can cause problems even if there are no other radios around. The same goes in reverse. If you run the AP at maximum power, and the laptop is right next to it, you can damage the radio within. I worked at an ISP recently doing WISP stuff, and it became quite clear we damaged some of the point to point link radios by running them 4-6 times "hotter" than we should have. If you're on a mac, hold down the option key and click on the airport signal meter. Your target power should be around -55 to -75. Not sure where to look for that on a windows box.
I had a sucky sig.
It's built to degrade, to make sure the company has a profit stream for ever!
As your wifi ages, you pick up more and more freeloaders which slow down your net. When you replace your equipment, the freeloaders get booted off and have to rejoin. Hence the temporary speed increase. Duh!
The poor ministrations of the duties of the tech priests leads to decay. It's either that or it's been touched by Nurgle.
The answer is usually cheap electrolytic capacitors in the power supply.
I bought several D-Link DWL-3200AP "commercial" access points around 2006-ish, and by 2008 they all had failed, and all for the same reason: they used cheap, 1000-hour/85C capacitors in the power supply. They had dried out and swelled, and reduced the power output capability of the power supply for the RF section, and the voltage drop during transmit was reducing RF power output.
I replaced them with decent 10000hr/105C caps and they've been fine ever since.
Very likely, it's a power supply issue. Cheap capacitors.
If any part of the RF circuit absorbs moisture, performance will decline over time. The trouble is that it isn't obvious what is absorbing the moisture. It could be the printed circuit board itself. There's no fix for that.
If you're in a dry place like North Dakota, moisture probably isn't your problem. If you're on the Florida Keys, you can bet that your electronics will eventually have a moisture problem.
What happens is that moisture kills the Q of the RF circuits by absorbing energy. I spent a lot of time in a previous job replacing tuning coils on walkie-talkies because the paper cores on which they were wound had absorbed moisture.
If drying capacitors were the problem, all your electronics would be affected equally not just routers in particular.
Nice election-cycle troll.
With the swing to digital, the quality of analog electronics (like amplifiers, detectors, and rf modulators) seems to have declined. Analog devices used to contain adjustable coils, capacitors, and pots to match resonant frequencies in those assemblies, and the alignment procedures set them to precision far beyond the purchasable steps. Depending on the situation and the skill of the tech this might just require a plastic screwdriver, or also scopes, signal generators and/or frequency counters.
Because components drift with time, loss of output or sensitivity was common with age and curable with redoing the alignment. The symptoms he describes fit.
Like, oh, say, "smart meters".
When they started installing "smart meters" in my area because "too many" people were installing solar, and PGE were paniced that they'd have to pay the same rate they were charging instead of cutting off the payout at net zero by having differential rates so they payed less when solar was active, my bandwidth went to hell, and I had to more centrally locate my AP to avoid the interference.
I haven't been to an Applied Superconductivity Conference for a decade, but at that time people were beginning to sell racks of very narrow band receivers for cell systems with high temperature superconductors allowing a narrower bandwidth than anything one can do at room temperature. Sterling refrigerators at 80K. One was able to increase channel density about a factor of three. I don't think this technology has made it to consumer electronics yet. Or will.
Very likely, it's a power supply issue. Cheap capacitors. Seen this a lot, with different results. One of the weirdest: routers stop working after you insert another ethernet cable (the power supply was already flaky, adding another extra load + LEDs pulled the voltage low enough to stop the processor). Also, on Powerline stuff.
And they want you to be buying their $10,000 router, not a $100 Linksys one. They're artificially degraded.
The best explanation for this that I've ever heard is heat. Producers of consumer grade wireless equipment don't put large enough heat sinks (Or hear skins at all in some cases) into their routers. I had a roomate who moved out and suddenly I didn't need to buy a new router every six months. Check that any file sharing software is set to connect to a limited number of peers (No more than 30 at a time.), saving the hardware from rewriting it's routing tables constantly and damaging the memory.
But it's clearly fat electrons stuck in the wiring.
To force you to buy a new one. Duh
Yours is the second such suggestion I've seen on Slashdot in the past month.
But, as I said to the last one, I have yet to see any real world evidence of tin whisker problems. I have yet to hear from anyone I know that has seen or had problems from them.
Indeed, the only place I have even seen mention of tin whiskers is Slashdot. Is it really a problem?
and nobody asked:
âoeDid you measure the signal quality?â
There are lots of apps that show you various signal parameters for most platforms (Linux, iOS, Windows). Even if you donâ(TM)t know what the signal parameters mean, the magnitude of difference between an old and new router can tell you something.
I think, if there is no difference, it is perception based; otherwise an overworked flux capacitor.
Go get some Apple AirPort Expresses. :)
Note: I'm an Apple fanboy and heavily invested.
I've tried DLink, Linksys, Cisco [which works, but on the $$$ corporate level], a few others, and Zyxel. Zyxel came close -- but the configuration has to be specific [repeater talk to SSID w/ specific MAC id]. The default quick setup could leave the sub-routers chattering amongst themselves... But I digress.
The AirPort's at $99 pay for themselves in setup alone. And frankly, they "just work". Unlike all the others the AirPort DOES PROPERLY PASS ALONG MULTI-CAST THROUGHOUT THE NETWORK. All the other products sub-routers ... dropped multi-cast. No more AirPrint, AirVideo, etc... Yeah -- there's a ton of iOS devices along with Mac's involved on my networks now. :)
They dynamically can be setup as a sub-sub-repeater. Wander the network rather seamlessly. I've just recently gone through this headache and with the AirPort's they will *OWN* the area I want to cover -- add AirPort's as needed to have signal strength / coverage. Just did a 6,000sq/ft house -- all three floors, my home, and the office at 18,000 sq/ft plus yard coverage [as the bay doors are opened :-].
Amazing product.
[Planned obsolescent products include] cars, especially those from one once big American car company, that recieved millions in bailout cash under this president.
In other words, not Ford. Unlike GM and Chrysler, Ford didn't take a dime of Bush's bridging loans, even though it did benefit a bit from herd immunity.
I've been using my WRT54GS since 2005 or so.
I've replaced the adapter twice so that's data point towards the bad power brick theory.
As lots of people have said, the RF noise floor in the area will increase over time as more "wireless" (especially 2.4GHz, and some 5.8GHz) gear is deployed (especially in the 2.4GHz ISM band, not just "wireless" computer networking, but lots of other stuff). Since the ability to recognise/decode your desired signal is significantly affected by signal to noise ratio, over time the RF environment gets worse.
New equipment generally comes with more sensitive receivers, which can work with a lower signal to noise ratio (over the couple of years I was following it closely, the improvement was something like half an order of magnitude less signal required -- relative to the same noise floor). Newer equipment often comes with new standards too, which also help improve reception (either changing to a less crowded band, eg 802.11A in 5.8 GHz instead of the crowded 2.4 GHz; or by better antennas that "localise" the signal -- eg, 802.11N deployed with multiple antennas and complex decoders to mix what it learns from each together for a better overall decode). So, in general, newly designed equipment is going to work better in the same environment than older equipment using the same standards as everyone else nearby.
It's also technically possible that there is increased resistance in, eg, the connectors between the transmitter/receiver and the radio antenna. This is mostly a risk in outdoor installations (where weather can get at it, particularly water -- usually in the form of water in the air, rather than rain; if you're getting direct rain on bare connectors you've already lost!). Outdoor installations near the sea are particularly at risk. Good weather sealing is critical in out door installations for this reason.
Usually one of the best things you can do to improve your reception and that of your neighbors (ie, encourage them to do it too) is to position your wireless APs as close as possible to where you want to use them (and if you have a large house, use more than one rather than trying to find a super "covers the whole house at once" solution; if they're got the same access credentials and are on the same LAN, modern computers should switch around as needed). RF power drops off rapidly the further you are away from the source, so getting closer to the source gives you a much better connection for the same amount of power. If you have the option to change antennas (at least older consumer wireless gear had that option, as does a lot of business gear), then picking an antenna design that focuses the power you're allowed to use into the area you want to use it will also help significantly (for the same reason). Eg, using a sector antenna to just focus everything into one corner. It'll help if your neighbours do that too. (And if you're talking to them, and they seem to have a clue, try to arrange not to be on overlapping channels.)
Ewen
I have a seven-year-old Linksys that still broadcasts through my whole house and to my back yard.
See my journal for slashdot ID's by year. Mine created in 2005. http://slashdot.org/journal/289875/slashdot-ids-by-year
It has been 7 years, it just would not die!
Monster Cables(tm) are much better for carrying electric audio signals, because they are outrageously expensive.
You need to buy cans of Monster Air to spray around your house.
You can find the product on their web site, and you really will be able to hear the difference on your wireless connections.
Probably.
Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
I've been using wifi instead of ethernet for about 7 years now.
Bingo, I would wager that most households use wireless only now, since wireless only devices are becoming so popular. I just bought a house...not one inch of ethernet in the place. I don't know about you guys...but that would drive me crazy to make all my desktops wireless!
It's by design, of course, to get you to buy more routers. Duh!
I have seen gear do this. The newer stuff is better as is enterprise gear. Get a Cisco 1231 AP which is rated to operate in 122 degree F envronments (case is solid di-cast alumminum, you can stand on this thing). Been using one for several years without this problem. You can get these on ebay for about the same price as a consumer WIFI router. Of course you will still need a sepate router as this is only an AP.
The first stage preamp(recive side) can get damaged by signal overload. Don't put your wifi device within 2 meters of your router/ap. The input of your laptop/computer can also be trashed by signal overload. This you will not find in any manuals for wifi equipment. Old HAMS will understand this. I am a little lazy and tend to test units too close together. Shame on me. My father used to work for NASA and rebuilding "front ends" of recievers was one of his primary jobs. Changing small indicator lights in the consoles was the other. There was a switch on the bottom of the console that turned on all the lights to see witch ones were burn out. Anyway i degress.... Analog/RF/Transistors ROCK!! Long live analog!
I know mobile phones and towers (probably) aren't like the low power 2.GHz networking gear this article's asking about, and the degradation I'm talking about is on the scale of seconds, not years.
But why does the signal strength indicated by the bars on my mobile phone often rise and fall across the entire range a lot of the time? What is "wavering"? Come to think of it, this does also seem to sometimes happen with the WiFi signal strength I see on my mobile devices, even within a few dozen feet line of sight of the AP with few or no other APs or 2.4GHz signals.
--
make install -not war
Those black psu that cost $3 are faulty. Get a new one
Liberty freedom are no1, not dicks in suits.
They'll swap brand parts for something their brother, cousin, whatever makes for a fraction of the cost, bugger the value tolerances or the longevity of the substituted components. Then in that tiny, unobtrusive case, the crappy parts cook, degrade and eventually blow up if you haven't replaced the whole device in the meantime.
Remember the problems with the first batch of Raspberry Pi's? The manufacturing samples came back all to spec and with the correct components and the contractor was given the go-ahead to produce the first batch. First thing they did was to switch the ethernet socket with magnetics for a cheaper one without magnetics. Passed the static check on the production line fine, arrived back in the UK and failed when actually required to work in a live situation. Luckily, the Foundation had access to some non-destructive investigation equipment and x-ray examination of the ethernet socket revealed the lack of those pesky little ferrite cores.
If that happens to a small outfit with its finger on the piulse, then I suppose the scope for sharp practice is greater with large companies and their long production runs and more remote quality testing.
Let me share some of my experience. I have been a volunteer for about 10 years in two Wi-Fi networks, and I have seen a lot of radios fail (http://www.zgwireless.net/ and http://www.vallisaurea.net/). Imagine a network of about 500 and 100 radios (wi-fi cards) connected to cables, and antennas. There are three main reasons for failure:
1) Static electricity can, and does damage circuitry. This is especially the case with "open dipole" antennas, such as common "stick" antenna, found on routers. To put it simply, one spark between your finger and router antenna can damage the radio inside and cause lower or no signal at all. I've seen it happen over and over.
2) Bad parts. Let's face it, parts inside routers are dirt cheap. We are talking about antennas, cables or semiconductors inside, and they do fail.
3) Noise increase. Wireless keyboards, mouse, remote controller, other Wi-Fi equipment, etc. It happens all the time, to everybody, but in that case replacing it with new does not help much or nothing at all.
The FCC's technical enforcement is a joke. I lodged an interference complaint many years ago, including documentation, flagging the power company (I have a commercial FCC license, so I'm competent in this regard). They sent me back a nice form letter and all my documentation, saying they don't investigate interference complaints.
I have done a factory reset and that helped too
I had the same problem over the same time period. In my case it turned out to be a cheap cable into the ADSL router whose degradation caused it all. One new cable and the problem went away.
I said - don't look Ethel!..., but it was too late..., she'd already looked.
These cheap routers are charged with an initial stock of electrons, and the processes that use them gradually erode them, wearing them to tiny nubs of their original brilliance. Much like memory leakage, this process can be controlled, but not eliminated. In addition, this process is irreversible and contributes to the soon coming heat death of the Universe...
I can haz hyperbole?
"Bingo, I would wager that most households use wireless only now, since wireless only devices are becoming so popular. I just bought a house...not one inch of ethernet in the place. I don't know about you guys...but that would drive me crazy to make all my desktops wireless!"
Same here. My house was built in the 1950s. Guess they used mostly wifi back then, too.
Frequency drift due to discrete components changing over time -- capacitor, resisters, Etc. CHEAP stuff.
Lead free solder degrades over time. Cracks and whiskers develop. A simple reflow might solve the issue. I have the same issue with cell. When it becomes an issue, I disassemble them and run them through the reflow oven at work.
It is similar to when you are given a new, fast PC at work, and your home PC seems to slow down...almost overnight! Try out your iPhone 3gs that you once thought was incredibly fast, especially compared with the Edge network you previously used. Now that you have 4G or LTE, it seems unbearably slow. Is it possible that you are getting faster wi-fi service elsewhere and your expectations of "normal" responsiveness have changed when the speed has actually stayed the same? Just a thought....
Heat. Its that simple. (BTW, before I got a BSc in CS, I got an ASc in Electronics Engineering). I have a broadcom wireless router (a famous WRT-54GL). Now the WRT-54GL is a fine router. I've had DDWRT on it for years, and have de-bricked it once. Even before I put on the 3rd party firmware, I noticed that it would crash about every 6 months and lose all old data. Unplug, look at what's wrong, plug in, and all would be fine, but I didn't know why it died. Finally I clued in: Broadcom put in a fine chip, but that bad boy gets good and hot (not quite hot enough to put a blister on your finger if you touch it extremely briefly, but close). My solution was to stick a heat sink on top (about 2" by 3"). Its shallow enough so that the case still fits on. I stuck it on with heat sink grease. It gets warm, but I haven't had the router die at all in over 24 months (and it would fail regularly on a 4-6 month basis). Along with hot chips, capacitors will dry out (especially electrolytic capacitors). When 'dried out' the case may turn brown or the end might start to pop out. If you are 'game', try opening up the case, and with dry fingers, just try touching components. Better, if you have a thermal gun or thermal sensor, use that. Anything over 30C is something that would be better if it were cool. In the old days, radio amateurs 'hams' would run high power transmitter tubes up to 'cherry hot', and it would be ok, but with solid state transmitters (mostly all in a chip), cool is better.
Six years ago, I took some tupperware cereal containers, drilled two holes in the bottom and pushed a WRT54G router's twin antennae through each hole. I caulked both holes. I inverted the containers, took an old broom stick and jammed it up into the contaners and screwed each router up high outdoors on buildings spaced 400' apart. I configured all three routers with DD-WRT in bridged/AP mode. I have never taken them down. This is in Vermont. They work great and cover some 18 acres. What more can I say?
*** Don't be dull.***
How the hell did the parent post get modded funny?
It said using channel 2 would REDUCE interference.
The channels in the bands used by WiFi (and other, narrower, systems) are 5 MHz wide. The WiFi signal is 25 MHz wide. So it uses five channels, starting with the one selected and going up. So to avoid overlap and interference, the selected channel number on two WiFi systems must differ by (at least) 5. In the U.S. the channel sets typically used for WiFi are channels 1-5 (select "1"), channels 6-10 (select "6"), and channels 11-15 (select "11").
Selecting channel 2 means you're using the top 4/5s of the first set and the bottom fifth of the second set. You will interfere with, and be degraded by, both those systems tuned to 1 AND those tuned to 6.
So "tune to channel 2 to reduce interference"? Ho ho ho!
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Switch your AP and wireless cards to 802.11A or 802.11N. Stay in the 5Ghz band. You won't have contention with any of your muggle neighbors.
Compare:
http://www.netstumbler.org/resources/image/5686
to
http://eetimes.com/ContentEETimes/Images/Design/TestMeasureDL/2011-11-06_crh_agilent_win_802.11ac_fig2.gif
I have Ch 44 to myself.
No points today and this is the best post in this thread I've read so far. Mods! Attack!
Weaselmancer
rediculous.
You bought a DLINK, which you shouldn't have done. And yeah, capacitors dry out... that just happens naturally.
So long as we're talking failure modes, don't forget dendrites. Nasty SOB's.
The problem: No lead-tin solder allowed in consumer gear. So they use stuff that doesn't have lead.
No problem, per se, with that, except that stuff that has tin (but no lead) tends to grow little, tiny, molecular-wide spikes that point from positive voltages to negative voltages. Time passes by, they grow longer, and Zap. If it's a high enough voltage (>10V with sufficient current behind it) they blow into vapor, but not before creating a spike on the power supply rails. If the voltage is lower, the blame things make contact and, at least to start with, a high resistance path. As time goes on the resistance gets lower and you get malfunctioning gear.
Take the board with this kind of a problem on it, wash with alcohol and all those tiny little dendrites break right off and wash away. And everything starts working again.
Space system guys dream of dendrites when they want nightmares. Business guys dream of them when they want repeat business. Manufacturing process engineers sweat this stuff, hard, because the effects are slow to chug in (years are typical) and it's all soldering process, process process.
Before replacing non-busted caps, try washing the board in electronics solvent with a bristle brush, then power it up again. If it works, you've found the problem.
Yes, I do play a hardware engineer in real life.
Electronics , specially cheap electronics suffer a great deal from metal migration in the junctions. The hotter the devices the more migration takes place and degradation in the signal is inevitable. Simple. Case closed.
Very important to have an accurate reference to keep rf frequencies on target. Problem is that crystals drift over time. Very slightly, a few parts per million and you're screwed.
http://www.bbc.co.uk/news/magazine-19760006
Or one line ASCII arts like http://i.imgur.com/3Y1VW.png from http://www.reddit.com/r/futurama/comments/11fy4o/our_neighbors_wifi_name/ ... I will have to do that for my future SSIDs. ;)
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
I'm glad this question was asked, because I've experienced the same thing in many situations. (When I worked doing on-site PC service, I had a lot of service calls related to wi-fi issues, and often found routers in a degraded state. One was so bad, you could literally only see the SSID and make a connection if you had a laptop or mobile device within 2-3 feet of it!)
The explanation about bad capacitors sounds about right though, since that's been the bane of most other modern electronics over the last decade or so. If not capacitors, I imagine other inferior quality parts soldered onto the circuit board that fail with heat and time.
As cheap as some of the wireless routers have become though, I suppose they really are pretty disposable. I've got a Tenda branded Chinese wireless N router here that's actually working pretty well for me with a 50mbit cable modem connection from Comcast attached to it. I have a 3 story townhouse apt. and the signal is still pretty good on the 3rd. floor with this down in the corner of the basement. The router's cost? Absolutely free last year with a coupon Micro Center mailed out. But I can buy them there all day long for about $20 each.
Years ago, the wifi was fast - WAY faster than the 2G signal you otherwise had in the living room. Today, it's WAY slower than than your phone, your 4G phone.
I currently have a D-Link DIR655, I've probably had it for four years, but within the last two it has been increasingly dropping my connection where I keep my laptop. There are only about 5 different wireless networks within the vicinity and only one on the 80211.n band I was using, the rest were on g; I tried changing channels and power but that only worked for a little while. Eventually I changed it to 80211.g only and it has been working great ever since.
In Google we trust.
It may be that the wireless device hasn't degraded. The RF environment may have just become noisier, thereby reducing the S/N ratio of your links. In fact, I would be very surprised if this hasn't happened regardless of any degradation in the condition of the wireless device.
I believe that everything is simply related to heat and microerosion caused by heat.
Semiconductors are made of silicium and dopants. At higher temperatures the atoms that make up the lattice are more "free" to move, as if they where ions in a fluid.
The effect is slow because the material is pretty solid, but, with the thermal stress plus the eletric fields, those small dopants move in the structure, decreasing the semiconductor characteristics of the device.
Being an analog device (The PA that amplifies the signal) their failure mode is by decreasing gain, until they become pretty much deaf.
Thats my hypothesis...
Heat is the silent killer. Apply a steady heat long enough and you damage the physical structure of the chips, their surface mounts and pins.
Tiny fluctuations in electricity, volts, amps and frequency wreak havoc with electronic circuits at the micro chip level since small changes in the over and under power against the standard are quite destructive at the atomic level of semiconductors.
But see, I don't consider that funny. Misinformation needs to be corrected - not just laughed at. Rate the answer down if it is wrong, not just up and funny :/
William George
"Bingo, I would wager that most households use wireless only now, since wireless only devices are becoming so popular. I just bought a house...not one inch of ethernet in the place. I don't know about you guys...but that would drive me crazy to make all my desktops wireless!"
Same here. My house was built in the 1950s. Guess they used mostly wifi back then, too.
They did use WiFi, of a sort. Except it wasn't branded. They called it face-to-face communications.
Have a nice time.
Old Bathroom cabinets (1950s) have a slit to dispose of razor blades by throwing them down the wall.
While this was o.k. to do in the 1950s, the inside walls of your house slowly filling up with razor blades will degrade the wireless signal. To make matters worse, blades were specifically designed to be lambda/4 of 2.4Ghz. Two touching razor blades that are sufficiently dirty and corroded will create a simple crystal receiver and rebroadcast your wireless signal with a random frequency shift. (The original intention of course was to disrupt Russian radar.) /p
To make matters worse, having two or more bathrooms might create a phase shifting device that could concentrate your and your neighbors' microwave radiation at random spots in your house or yard. My only advice is, never to stay at one spot too long, as this has been linked to spontaneous human combustion.
Having worked on some of the 802.11 algorithms, I can safely predict that all equipment with a crystal oscillator will eventually degrade.
802.11g was speced for a differential frequency deviation of 40ppm. Nominally thats 20 ppm allocated to AP and 20 to the remote.
In an effort to cut costs as much as possible (because most of you shop only on price, right?), the crystals used do not have good aging properties. It's not unusual to see crystals used that have as much as a 5-10ppm/yr agiing spec.
Unluckly customers will see their AP and remote device diverge, and within 3-4 years the frequencies are out of range of the algorithm's ability to compensate.
My access point lives in the U.S.A., you insensitive clod!
Although WiFi is digital protocols, the TX/RX is still done in analog. On 1 side you have the Transmitter that pump 1 W, and at the other you have receiver that must receiver that must "hear" a signal that's as small as 0.000001 W. Since they used the same antenna to do both sending & receiving, there had to be some switch and protection for the receiving side as to not damaged by the transmitter side. This switch (they usually called PIN diode) is unfortunately "degraded" over time. As this is still analog component.
If this switch fail completely you wouldn't have any signal. But if it degraded over time, what you'll noticed is your received got "overloaded" by the much bigger transmitter signal leaking into your receiver side or the resistance got higher (actually they call this insertion loss got higher) so the receiving signal actually get less into your receiver (we called this deaf receiver).
news.thomasnet.com/fullstory/PIN-Diode-Switch-suits-Wimax-and-Wi-Fi-applications-822264
www.eetimes.com/electronics-blogs/other/4371601/Receiver-protection--a-game-changer
Other problem is the limited number of channel available on WiFi
en.wikipedia.org/wiki/List_of_WLAN_channels
In a "busy" environment (with a lot of AP turned on) that can became problem. That although WiFi employs a "collision avoidance" transmit mechanism, there's a limit on how it can go with limited channel resource.
http://revolutionwifi.blogspot.com/2011/03/understanding-wi-fi-carrier-sense.html
So in the busy environment, you probably would have a symptom of lack of speed and/or lack of signal caused your router are struggle to get space in order to transmit/received into their client/s.
---
What i do currently is:
- Use WiFi that's supported by DDwrt.
- Buy those router and put DDwrt on it.
- Set WiFi channel to AUTO.
- Set power as i need (i don't want my AP to bleeds its signal too far away than necessary)
- Set to reboot the router every week (at Sunday 2 AM in the morning time)
The reboot is required so the WiFi would set to look if another AP is turned ON and occupies the same channel with my router. DDwrt itself is pretty much stable & have a lot more functionality than stock firmware.
CAUTION:
Know what you're doing. You could brick your router (doing the above). See if the above worth your gamble. I'm not responsible for your bricked router. You've been warned.
It might be a high swr standing wave ratio caused by metal objects nearby. This can gradually burn out the "finals" that is the power transistors in the transmitting circuit. Something similar could be happening to the router.
www.Migrainesoft.com - Computer giving you a headache? We can fix that!
Most consumer used capacitors age over time, this could be related, but I do not design routers.
The problem may not be the wireless router deteriorating - it may be the other unlicensed devices deteriorating and leaking more RF.
Replacing the 802.11 devices with more contemporary ones designed to cope with noise will help, but perhaps the problem isn't the original routers deteriorating - maybe the household microwave oven is leaking more
If your children ever found out how lame you are, they'd murder you in your sleep
heat causes semiconductor devices to work less well over time. The design must provide for appropriate heat dissipation. A design in which the devices are running at 110 or 120C on a consistent basis will begin to slowly fade. I'm talking about channel temperatures. your box may be "warm" but that could mean that the devices are running hotter than they should for long term reliability.
also a possibility, and my personal favorite. the clock oscillator. poor, even mediocre, clock oscillators have crummy aging characetistics. take that design which is running too hot : the crystals age faster than normal and eventually your device's frequency is too far off. as it begins to drift out, the reception will suffer.
repeat after me: modern electronics for the consumer market are designed for cost. it is expected that you will replace it in 2-3 years. when something lasts you 5 or 6, it's not by design. so if they can save $1 because they don't use the necessary heat sinking or board area to dissipate the heat, and everything is running at 90 or 100C, that's ok. ship it !
Absolute statements are never true
It's an exploit of a loophole in /. moderation. Funny mods provide no 'karma'. The basic idea is mod the post up funny, and back down overrated. This can repeat without end with enough moderators, and the poster goes to full negative karma for just one post.
I can think of a few very common occasions where you will in fact be wrong.
Increasing the transmit power on the base station will increase the amount of packets that reach the guests without corruption, even if the ACK (assuming you are using TCP and not UDP or some other non IP network protocol) might fail some time. This may increase actual bandwidth and very possible the rate at which the two devices "sync". Very often, transmission to guests will be rather asynchronous, so if your transmission type is adaptive to buffer bloat, a few retransmits might actually get your data in quicker than a lower data ratio.
Several streaming protocols use UDP and as long as the client sends regular signals it's still connected data will remain coming in. Some even have an automatic bandwidth throttling mechanism built in and will allow higher quality streams this way. Granted, not very common for laptops, but smart phones and tablets that use 3G and wireless tend to use the same protocols for both networks and will profit from this.
Also, who's saying you are only increasing the base station transmit power? My laptop has increased transmit power in the AC profile. Once it goes to battery mode, it's on a more conservative setting. Works just fine, I get double the data rate on the ground floor and upstairs I still get usable signal, compared to intermittent losing connectivity altogether.
I was promised a flying car. Where is my flying car?
Whats being described sounds common to the transmitter power.
Just guessing - but I would say that the RF power transistors in the PA
are slowly losing efficiency.
Could not find any burn in data for GaAs power transistors, but its a possibility.
www.effectiveelectrons.com "chips that work" Analog, RF, Mixed Signal
Low quality capacitors are known to go to crap rather quickly. As the capacitors die the filter circuits will "spread" the range of frequencies that will be accepted as valid signal. Timers dependent on capacitive charge will also get "slower" (e.g. take longer to reach the charging threshold and rigger the time event).
The sad truth is that modern electronics, particularly those built on the cheap or with poorly sourced components, just go stale on the shelf, and go stale faster when in use.
Innocent people shouldn't be forced to pay for inferior software development.
--"Code Complete" Microsoft Press
Here is the correct information for the PGE smart meters:
The meter is radiating on two bands:
NAN Communications
- Frequency: 915-928 MHz
- Spread spectrum technology: FHSS
- Channels: 43
- Receiver sensitivity: -97 dBm for 1% PER
- Modulation: Binary FSK
- Transmitter output: 30 dBm
HAN Communications
- Frequency: 2.4 GHz ISM Band
- Spreading technology: Direct Sequence
- PHY/MAC: 802.15.4
- Transmitter output: 20 to 23 dBm (200 mW)
- Receiver sensitivity: -97 dBm for 1% PER
- Power, Transmit: 1.6 W (1.8 W max.)
The transmissions on the HAN band are nearly continuous, since it does station connection heartbeat broadcasts at 11 second intervals (same as anything talking to a 2.,4GHz AP with SSID enabled).
This is specifically why PGE has had to remove the "smart meters" on homes with Arlec Model PC600 "Automatic Light", which operate on ultrasonics in the 2,5GHz band, and which are capable of being triggered by the 2.4GHz "smart meter" broadcasts. Basically the damn security lights turn on every 11 seconds; time out after 5 or so, go dark for a bit, and then come back on. All night long.
I had a DD-WRT router a couple years ago, which allowed you to change the wireless transmit power. Somehow pushing more power through the antennas caused the unit to degrade faster, although I got excellent range for the first 30 days it worked.
Almost all cheap and even some expensive router's are built to fail. Cheap components and no cooling end up causing weak BGA joints and dryedup cap's causing it to fail. It the same with new TV's they last a few years then they die because of bad caps or overheated voltage regulators. look at old tv's they go forever. well until the tube burns out. I have a old console TV still works.
Two words: Planned obsolescence
Change the wireless channel on the WiFi router. Try channels one by one until you find one with good signal coverage. Use a less popular channel where you won't run into interference from neighbors and your signal strength problems will be solved. As easy as 1, 2, 3, ... 4, 5, 6, 7, 8, 9, ... you get the idea.
What most DSS cards have at the output is an array of transistors. Those transistors do degrade, specially when used with cheap antennas with high SWR.
SWR is the reflected energy that is not irradiated, and rises the voltage at the transistor output. Think of it like thowing balls to a wall with holes. Some balls bounce back.
Over the years I've owned modem/routers by Netgear, Linksys, Cisco, Swann, TP-Link, Thomson, Corega, Origo, D-Link, probably some others too. I've used them with different ISP in different countries, various physical locations, in the presence and absence of other local networks, with stock or custom firmware (DD-WRT, Tomato and OpenWRT) and I've come to one conclusion: they're all crap.
For the most part they work ok when new, but I have never had a router that stayed working for more than a couple of years. Usually the first symptom is wifi dropouts, then dropping ADSL, then router crashes. Eventually they just stop responding at all. It's not interference since getting a new router always fixes it all (for a bit anyway). I really don't understand how they would 'wear out' (being solid state); perhaps poor thermal design? I'm not a demanding user - they need to connect my ADSL to my ethernet and wifi quickly and reliably and provide NAT, DHCP, basic firewalling, operation as a DHCP relay or bridged access point if it's not a modem; I can happily do without media servers, port triggers, QoS, static routes, print serving etc - if it can't provide my most basic requirement of simply maintaining a connection, all that's just excess junk.
I've had the worst record with Netgear; their stuff just likes to die. I've not tried Apple, Buffalo or Draytek. Most of the open firmware is generally better than stock, but instability and risk of 'bricking' is stupidly high, and we wouldn't need them if stock firmware was decent in the first place.
It all seems a bit like my experience with Terry Pratchett books - I have a bad experience with one, so I look for suggestions, get recommended something, try that, and and am disappointed yet again. I'm certain there must be a market for wifi kit that isn't crap; it seems to be up for grabs.
I recently had the wifi on a Billion 7800N just up and die on me totally. It turned out that when they installed the wifi card (wifi was a separate miniPCI card) they used some sort of glue to make sure it didn't come loose (either that or the glue was there to show if someone tampered with the card...) Anyway, apparently this glue absorbs moisture over time and becomes both brittle and conductive as a result. My wifi card had some glue bridging a few pins of an IC - I removed all glue off the card and it now works fine again. From my research, the glue normally "fails" after roughly two years. (For the record, my failure was not a radio problem, the daughter board totally stopped responding to the main board.)
I'm not suggesting that in all cases this is why the wifi performance degrades, but it is at least possible that some cases (where this magic built in obsolescence glue is used) that it could be a contributing factor. Just a thought.
That is incorrect. Although there are eleven "channels", the frequency on most of them is so close together that that they still interfere with each other. In reality, there are only three: 1, 6 and 11. By setting it to a value in between, you are in fact bleeding into (and getting interference from) the two that are on either side of you; it is bad for you and everyone around you. You should stick to 1, 6 and 11. This is pretty limited but in reality most people never change the channel, which is usually set to 1 or sometimes 6 at the factory. So, for the time being the geek in the know can get good signal on 11, even in crowed urban areas.
http://en.wikipedia.org/wiki/Capacitor_plague
Experience: Like TFA poster, with the exception of my two Apple Routers. While I balked at paying double price for my first one, I gladly paid for the second. The first lasted five years and is puttering away at the inlaws, and the new one (dual band) keeps ten devices running without a blip. The four cheap routers before those all "went". The only real difference I can see is the Apple is a massive heat sink and the others had none or very little.
Waxes and oils that are particulates in the typical home settles everywhere, even on antennae. Waxes and oils, grease etc reflect RF signals. Old wi-fi antennas should be cleaned regularly and you'll find that the router will come up to spec, if it's not any of the other causes mentioned.
Don't be apathetic. Procrastinate!
Thanks for the detailed answer.
There really should be a "+1 so good I was satisfied enough to stop reading this thread at this point" moderation option that includes an extra karma bonus for the poster.
Hmmm, no kiddin'..... Okay, so I've replaced my WRT54GL router 3 times since I started using one. Exact replacement, same placement and connection, same 'ol same 'ol. I got six of 'em gratis, they fell off a truck, or so I'm told. Worx for me. All undamaged, however, I tested each of 'em, shared a couple, and put the others away. I'm glad now that I did that, since both of the first two slowly 'died', indicated by diminishing range. The router(s) remain connected in the same exact spot, on the second floor of the house. The house and furnishings and their orientation changes very little. Antennas both stand upright. There's no new devices in the home, other than a few that come and go periodically with friends. Friends have noted some loss of received signal strength here over time. So, the physical and RF environment where the router lives is virtually(pardon punnage) unchanging. Only the results have changed. The router's not exposed to extremes of heat or cold, and the system as well as the router is always on and connected, very rarely rebooted or powered down, so there's no thermal/power cycles to stress it. I've had no reason/need to upgrade the router firmware, they're completely stock. (I'm gonna to try doin' this, however.) The PC/router/external drives/etc. all connect to a large well-regulated and -filtered UPS, so the system sees no significant variations in power, or transients such as those caused by lightning. No one smokes in our house, and there's no other significant airborne contaminants or vapors, the home's old enough for most outgassing to have ended and we've brought in nothing to change that, no new sheetrock, carpets, or furniture. There's no significant amounts of humidity, or lack of it, and the HVAC system has HEPA filtration, the filter media changed a lilbit mo' often than recommended, just 'cuz. I'm an Extra-class radio amateur, so I have some experience and knowledge in the field, and access to test equipment occasionally. This house is far enough from our neighbors and any other RF sources that an RF spectrum analyzer shows minimal extraneous noise, apart from the normal low-level terrestrial background things and cosmic sources. No cordless phones in the home, the 'ol lady and I use only cellys. So, I've had two of these routers lose approximately 25-40 percent of their range before I replaced them with another. Replacing them with another exactly like it, almost sequential serials, likely made the same day(Friday? Monday? Heh.), brings the range back to optimal. The ones I've removed from service show no obvious internal defects or damage, no indications of heating, no significant dust or films of anything airborne. Connections of cables/plugs/antennae are firm and have remained so over the lifetime of the unit. I haven't done any deeper or more scientific testing or autopsy, I thought this was just my comeuppance for having acquired them so cheaply, and had spare(s) in the closet, so... As Dad always used to say, nuttin's free, kid. So, yeah, add me to the growing list of folks that've seen routers fadin' slowly into obscurity. They still work, of course, but lost a fair chunk of range over time. One's down to roughly 25-33% less range, the other maybe 33%-50% weaker, all other parameters being equal. I haven't put any of 'em on any sort of test equipment to measure more exactly, sorry 'bout that, y'all. I hadn't really worried over it that much til tonight, readin' this in /.. I can't account for why they're failing, no more than why there are Republikkkans. Is what it is what it is. I'm considering now what I can do with the weaker routers in here to improve/enhance the network, and now I'm also watchin' this thread. Ain't /. great? ;)
Peace.
Olphart at play. Ruck FepubliKKKans. Welcome to the Worldwide Idiocracy, y'all.
Cats!
I'm in yer router, stealin' yer megahertz.
Electronic components wear out and degrade with age, same as everything else. Sometimes it's heat, sometimes crappy construction, sometimes just poor design. It's the automotive equivalent of saying "I bought new tires for my car three years ago and now they are worn out, why is that?"
The transmitter doesn't 'go bad' unless it just stops working altogether. 802.11 wireless is a shared aggregate medium. There are 11 channels (like a TV channel) in the US (there are 13 in all but they are restricted for use). the most common things effecting wireless transmission are:
It's important to understand that if you are using 802.11a-g you have up to 52mbit worth of bandwidth, for 802.11n it's 300mbit. This is shared in the channel you're on.
1: Multiple radios same channel. So if you're router is on Channel 1 (software select), and 4 other routers are also on channel 1, then you are sharing that single block of 54mbit of radio space between all 5 radios. 54mbit isn't much in the grand scheme, and if you're all streaming, forget it. Also, there is a 2 band overlap on channels, so if you set your router to channel 3, then it could be effected by channels 1,2 4 and 5. Cisco recommends setting radios to channels 1 6 and 11 for this reason if you are installing multiple radios.
2: too many devices. See #1.
3: Your router doesn't support multiplexing. Some (not all) routers have more than one antenna. Better routers can use one antenna to output data, while the other is used to input data. This helps reduce errors in the signal.
4: You have a noisy environment. Microwaves. Wireless TV repeaters, cordless phones. Anything that sits in the 2.5 and 5 ghz range of radio WILL interfere with 802.11a-n. This can be isolated by acquiring a spectrum analyzer. "Wi Spy" makes one that is pretty inexpensive for the 2.5ghz range that is really useful for this purpose. The 5ghz one is a bit more. Actually, the Wi-Spy will help you resolve most of the issues I've listed here.
5. 802.11 doesn't do well with concrete and/or lots of metal. If you live in a bunker (like where I work) your signal will fall off faster than normal with distance.
Really, the take away of this is: By a >good router to start with, Cheap ones will work if you know you're all alone in the spectrum, but if you live in say, and apartment the better radios will get you better performance. If you continue to have issues, then a spectrum analyzer will help you figure out exactly what you need to do. I'd also recommend making sure you get a router that's compatible with DD-WRT simply because the imbedded software that comes from vendors doesn't always provide you with the required diag or tuning parameters you need to get to the bottom of this.
Most people on slashdot are fixated on one answer. It could be any of them.
1. Bad hardware including AC adapters, poor caps, bad cooling, etc.
2. Interference from neighbor's devices, this doesn't just have to be wifi access points. Had a neighbor who was into building tesla coils. It totally screwed up my wifi when he was playing with it.
3. Poor drivers or wifi devices in your computer. Had a linksys card die on me. It wouldn't stay connected more than a few minutes at a time (pc card)
4. All your neighbors caught up to your wifi router's capabilities. Cheap 802.11n devices are available now.
I moved into a new house and everyone in the neighborhood had linksys routers operating on 802.11g. There was so much interference all around me that I couldn't get a connection upstairs and two concurrent computers could never stay connected. I upgraded to an 802.11n apple airport extreme. It can operate at 2.4 and 5Ghz and now I'm the dick in the neighborhood that overwhelms everyone else's wifi signal because I spent more than $50 on a router. Two years later, I only see one device that's on 802.11n besides mine.
At this point, I just gave up on wifi for serious tasks and use powerline networking. It's much faster although I've noticed some CFLs cause interference with it so I've had to go to LED light bulbs.
MidnightBSD: The BSD for Everyone
They have obviously been deliberately designed/programmed by the manufacturer to degrade gradually so that consumers are forced to buy new products regularly. This has happened in many industries before.
They did use WiFi, of a sort. Except it wasn't branded. They called it face-to-face communications.
You're confusing their NFC with their WiFi. Their WiFi was called "yelling down the hall".
You have a good point, but decent quality replacements won't always fit in small cases, as the low ESR types are often a bit bigger.
Right now, I have a similar problem on my hands. An old Siemens Gigaset 515se that tends to hang up after ten minutes of serious DSL traffic. I've already re-flashed the firmware and started to exchange capacitors on suspicion, but for two of them I see a problem with putting anything but the standard types in:
The old capacitors are 470microfarad / 35V, but the case and the neighboring parts allow at best a part with 12mm diameter/18mm length. Any matching low ESR type I could find is at least 20 mm long.
So I guess I'll try replacing the existing caps with new "standard" caps. If that does not help, I see one trusty old DSL router heading for the scrap :-(
C - the footgun of programming languages
Astonished that "arms race" comments are coming out ahead of "put a UPS on it", so I read down a few screens' worth of replies. Glad I saw your response, now I can explain to folks WHY step #1 is keep your electronics on a UPS.
If I try to run any of mine at full Tx power and full traffic load in the summer, they overheat and need to cool off before they will work again. I bet that OP has increased his network load, like by watching videos across it.
I have had a wall wart partially fail too. I repowered that one with a way over spec brick I had lying around.
My solution is to remove the (router) case. I can run it at full power without it overheating so it doesn't crap out after watching half a show.
I've only seen fat caps or desoldered caps on motherboards. Never seen a router desolder a cap.
This seems to have degraded into a general hardware failure thread. for my $.02, I'd like to remind readers that maybe 25% of small consumer products that fail (like routers, switches, etc) can be traced to a failed wall wart.
Competition Good, Monopoly Bad.
This is the reason that some mixed signal chips have both analog and digital power supplies -- to allow the board designer a small chance of keeping the analog signals clean by filtering the digital switching noise before it gets to the analog input.
The problem is compounded when you have multiple switching chips.
Of course, as other posters have pointed out, the problem certainly could be in the power supply itself, in which case swapping the power supply certainly could be a fix.
I have three Apple Airport base stations. One is the very original which is now 13 years old. The next is an Airport Express which is 8 years old. The last is an Airport Extreme which is 5 years old. All are still running just fine, excellent range and interacting well.
I did replace the power supply capacitors on the original Airport Base Station when it was four years old but aside from that no issues.
All are sitting behind top quality surge suppressors for both phone/ADSL and power lines. We get a lot of lightning so these are important. Perhaps if someone had one plugged directly into the wall it would be getting surge hits that are damaging it gradually although usually that sort of thing does a lot of damage rather than gradual decline.
Do you live near the ocean by any chance?
In Reason We Trust
What is a good solution for a WiFi access point or router that works well and is easily obtainable. It seems that everything on the market these days is consumer grade crap.
You could always try upgrading the firmware before replacing the router. A lot of people never think to do that and it can make a big difference. A new router would always have way newer firmware.
Much of it can be attributed to semiconductor reliability. When you operate things hot, they age faster - failure mechanisms accelerate. This is the opposite of why you refrigerate food, which doesn't kill a single bacteria present but slows down their metabolism enough to extend the effective shelf-life of the product. Microwave semiconductors used in wireless (from cellular to Wifi) are typically operated at high temperature to achieve performance. So these parts simply age quicker.
Added to this is the fact that microwave semiconductor device manufacturing involves materials other than silicon quite often (the highest performance devices are rarely silicon). Non-silicon semiconductor manufacturing is roughly 15-30 years behind silicon in reliability testing and design; some key manufacturers do no reliability testing at all so have no actual process control for reliability (I won't name names but I know this from visiting them and discussing it directly) - they simply do "end-of-line" functional test and hope for the best.
It is possible that the caps are getting older, and thus draining power from the antenna. Just guessing...
Capacitors are often the culprit. The cheap ones are infamous for losing their accuracy.
This is not a new or recent problem. It has been a known problem since solid state devices replaced tubes in RF transmitter circuitry. It occurs at a higher rate with devices that operate at higher frequencies, are highly integrated and that operate continuously.
How do you know the signal is degrading vs a dozen other things that could be different? Walk around the house with your cell phone and map out the actual signal strength. Take notes. Check again next year, same weather.
Either you are buying astoundingly poor routers or something else is changing. This isn't something that just happens -- especially after only two years! My parents still use a nearly 10 year old belkin router that still covers the entire house and yard. The router in my old apartment was four years old and never had an issue. My current router is about two years old and still blankets the entire apartment building.
I've never in my life heard of a router breaking or degrading from normal use. Maybe you get one with defects here and there, as with anything, but they're not that complex. There's absolutely no reason a router shouldn't work just as well in a decade as the day it was purchased.
Want a useful experiment? Take one of the old routers and install OpenWRT on it. See if it works better.
Bruce Perens.
Jebus! Doesn't anyone teach basic electronics in high school anymore? Tuned circuits can and do degrade over time. Your cell phone is a tiny radio and guess what?, it's happening there too. Apparently Electrical and Computer Engineers don't matter anymore, ya'll just keep larnin' yer C# and Java.... that'll save the world while AT&T T-Carrier and CDMA/TDMA/GSM go down the toilet.
Radio transmitters have "finals" that can get degraded over time by manufacturers allowing their radios to over power the finals. This is often done to compensate for the poor quality, inefficient antennas attached to the radios. (Overcoming loss)
I've seen devices made in the last ~3 years falter in this manner while my 8+ year old WRT-54G works fine. My investigation has found: +++Wall warts suck -- BAD. Replacing them by connecting them to my own power supply yielded greater stability (fewer crashes) and greater range. +++Checking the RF modules find that many newer RF modules are designed for low cost. +++ +++Low cost RF modules were designed with lower Max Power limits (full legal power is full power of the module) +++ +++Low cost RF modules were found to use components with wider variance (i.e., tolerance values of resistors, etc.) +++ +++Low cost RF modules tend to just barely meet specification working on the ragged edge right out of the box. +++ +++Low cost RF modules tend to have very miniaturized antennae or even antennae only provided by PCB tracing, with no provision for external antennae My next Wi-Fi I will take great pains to ensure is designed for long life. I may even build my own.
OP's post is not very useful. At best we can idly speculate.
1) What (specifically) makes the OP think it's actually a drop in transmission power? Is it ONLY the fact that replacing the unit makes it work "better"?
2) What has the OP done to diagnose/debug the old router? Has he/she (at the very least) tried different channels?
I've been using an ancient WRT54G for close to a decade, and have noticed no drop in performance. (Yes, the blue guy) I can cover a significant portion of my city block, at least 1-2 houses down the street either direction. I live in a somewhat dense neighborhood, so every year or two I have to change channels when a neighbor's network collides with mine until it works well again.
I have no problem with your religion until you decide it's reason to deprive others of the truth.
I am replacing 'consumer' network equipment with 'commercial' level equipment. I'll know in 5+ years if it helps.
Why would DVD players or TVs go out? Most of us upgrade because we want 'new toys' before the old ones ware out. But infrastructure items we expect to last forever.
Where I live we have power spikes and sags and some short term outages (5 sec to 10 minute). I put APC Pro UPSes on all elecronics I want to keep (BackupsPro UPSes provide true sine wave, Backups UPSes provide 'modified' or 'stepped' sign wave. Cheaper UPSes all seem to be 'modified' or 'stepped' power, only more expensive ones do true sine wave.
Anyway the electronics on UPSes don't seem to fail as often. My old Linksys wireless router has worked for 7+ years now, but is showing signs of degrading. I just can't find anything with the same power.
Some of my replacement equipment is now the Ubiquiti series ( ubnt.com ) of routers, access points, etc. Their price isn't bad, and they have better remote configuration options (a big thing for me) than consumer level stuff (like Linksys, et al)
... "When you pry the source from my cold dead hands."
Hmmm. Either the devices are degrading, or you are comparing devices with different standards, or the environment is changing.
I would suspect routers using 802.11g. These seem to be (in my experience) temperamental. 802.11b is fairly robust, but way outdated these days. 802.11n seems to perform very well (at least for me.) If you, and your friends, were using 802.11g, you could get that behavior by having the environment change on you. Having the noise floor increase, for example. If you were accidentally mixing 11g, 11b, and 11n, the performance would vary wildly and unpredictably, from an end user standpoint. Personally, I would dump 11b and 11g and go with 11n as much as possible.
When buying a new router, Google the model to find out the supported standards. If it is newer than 11n (standards march on), make sure it is CDMA based and not TDMA based (IMHO). Google should be able to answer that question.
The other possibility is that the routers were operated in an enclosed space without ventilation. It is unlikely that all would have this problem, but make sure they are ventilated to avoid overheating.
I've looked at the top threads here and I've noticed that most people blame the noise floor rising as WiFi devices become more capable, and a few people talking about how to configure the routers or flash the firmware to get into less-occupied frequencies. A few people have described how 5 GHz spectrum is more vacant, but doesn't travel as far, how there are only three channels that don't overlap in the US, and how traffic slows down when new routers have to step down to old protocols.
I don't hear much about the gains that can be made by freeing up one's dependence on WiFi within the home. If your computer is next to the router, running an Ethernet cable to it is a no-brainer, and certainly too obvious for this thread. But video streaming gets more common every year, so maybe these homes have reached the point where it's worthwhile to run Ethernet to the room with the TV in it. And I'm sure there are kids out there who think nothing of running BitTorrent on a WiFi device. Is it worthwhile to move devices to Ethernet, or is that just a lay assumption? Or maybe also too obvious for this thread?
Moreover, I've never heard anyone talk about consumer-friendly methods to block WiFi signals from outside the home. Whenever I ask my network engineer friends about this, they act like I'd have to build a Faraday cage around my entire network with specialized copper mesh. No, I'd just like to do my level best to discourage signals from passing through my exterior walls, so that my traffic doesn't have to compete with weak packets from across the street. My router is close to the north wall, so is there something small I can hang behind the router to attenuate any of the signal through that wall? What about replacing my chain link fence with concrete?
Besides what they mentioned about bad power capacitors and bad power supply "block", it could also be a "drifted" frequency crystal. They probably don't have a adjustment for that, on cheap boards, and you would need instruments anyway.
The newer boards might be able to adjust frequency automatically, but it probably still degrades if it drifts too far.
1: SWR caused by mismatched antennas (and influenced by the area around the antennas) causes the RF front-end a lot more stress and can lead to breakdown. The front-end on these wifi cards is about as cheap as it gets.
2: Bad or cheap components. As others have pointed out, bad caps can cause major problems and do plague modern electronics. Your RF bias circuits could swing all over the place if the caps are going bad. Or the heatsinking in the power supply. There are lots of potential failure points in circuit components.
3: SNR: Some other folks pointed out that as the wifi band gets crowded, it's harder to get a good signal. When you buy a new router, perhaps it has more intelligent rate algorithms and interference mitigation techniques?
Really though, ask yourself why *anything* electronic needs replacement. There are usually a number of believable answers.
PS: As someone who works a lot on wifi drivers, antennas, and systems, I use an RT-N56U at home and it works quite well. Ugly but great.
So what you're saying is, "as the Borg adapt, change frequencies of your phasers".
From your noisy power company breaking down the capacitors in the power supply, you get gradually weaker signal until nothing.
I had a 'dead' WRT54G that finally stayed off after weeks of spotty connection. Switched to a 12v tap from the server it lived next to, and it was good as new again.
Same with a trashed Belkin el cheapo: new power = new life.
If the transmitter isn't exactly matched with the antenna, it will cause a standing wave. Sometimes noted as a SWR or Standing Wave Ratio. If it's too high it's going to wear on the electronics, eventually killing them. Back in the old days with CB type radios it was very simple to fix. Wave was I think 10' if I remember right. So the closer you got it to exactly 10', the better. Some radios would allow you to do it with a meter. With these routers you're in the millimeter range - good luck getting that right. I have an old WTG54 and it still works well. The Netgear I bought didn't last a month. They're cheap enough, just buy a new one.
As all of your neighbors add wireless routers, the noise floor goes up, and the usable signal goes down, even though the signal strength is the same.
Yup
Spot on, Ano_Coward !
Cheers, Ozogg
http://dsc.discovery.com/gear-gadgets/boost-your-wifi-signal-using-only-a-beer-can.html
Free speech was meant to be free for all... how can anyone grow up in a nanny state ?