I really dislike the commentary on Slashdot. It is worse than reddit in a way. Well, not really, but it is very annoying that an interesting question can be raised, and then seven thousand geekoids feel that it is very important for the world to read their own stupid little joke about the question. If someone provides actual information or rephrases the question usefully or anything like that, then it is lost in the sea of irrelevant yammering that is Slashdot.
So, yesterday or so, acer123 posted this:
“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 we learn that it is actually the expansion of the universe, we hear some of the usual algorithm jokes, and so on…Well, maybe there is some good discussion in there, but it is hard to find.
Anyway, what is the cause of wireless degradation, with the first sub-question clearly being: “Does it happen?”
Should we submit this one to the Mythbusters? Is there a reason to blow up a wireless router? Probably not…
Have you read the breakthrough novel of the year? When you are done with that, try:
or Paperback
*Please note:
Links to books and other items on this page and elsewhere on Greg Ladens' blog may send you to Amazon, where I am a registered affiliate. As an Amazon Associate I earn from qualifying purchases, which helps to fund this site.
Not a single answer, but here’s an entire thread discussing this issue over at StackExchange: http://serverfault.com/questions/20182/why-would-wireless-routers-become-less-reliable-over-time
One of the simplest suggestions was that electrolytic capacitors in the router may dry out over time. This would be gradual process so it wouldn’t cause a complete failure.
Mythbusters coverage would be pretty slow teevee for the timescales under discussion.
A few thousand years ago, I worked in radio and the tubes powering the broadcast antennas did degrade but I cannot recall anything like a linear curve. Cooling was a far more immediate and direct need.
Skipping, also a radio phenomenon, could have an analogous wifi equivalent if the routers are moved relative to large conducting surfaces.
I would like to skip the /. responses and see what others know of this.
Hi im paul.
I would say first of all that i have been in ham radio computer design and programer, and can work miracle’s with just about any electronic device for years.
and i will tell you without no uncertain doubt that a device like a router can degrade its signal over time.
i have worked with radio’s as old as 50 years old and if they have power running through them and it is functional , it will produce the same signal as it did 50 years ago without degrade. reason for this is that with exception of tube radio’s , resistors or diodes or capacitors can blow out and fail causing a unit to stop working but not simple degrade.
cause there is no ware and no moving parts.
so if there is a router that is producing a weeker signal over time than its designed to do so.
sounds crazy but if you think about it . its like a computer, a computer can not miss CALCULATE or preform a operation wrongly, if it fail its because of it was wrongly programed. computers dont make mistakes humans do..
end of story
My first thought is that many wireless access points either start with a default channel or choose a channel to operate on when first powered up based upon no other strong signals using that channel (frequency range) which can be detected at the time. Many will stick with that channel and never look again until a reboot.
It could be as simple as neighbors setting up access points which use the same channel (or nearby channel), swamping those frequencies as the person moves a mobile device around geographically in their own house. There are many Android apps which can be used to look for channel interference and the signal strength of nearby channels in use.
Does this guy live in a single family house or in an apartment/condo complex? If the latter, he could be seeing interference from his neighbors’ devices. That’s less of an issue for single family residences, but still possible for a moderately dense neighborhood. I can detect my neighbors’ wireless networks in my house, with nearest neighbors being typically ~100 feet away in the neighborhood.
If there is an uncompensated thermal drift in his wireless devices, that could also be the culprit. The effect would be that the carrier frequency of the devices has shifted off the nominal band frequency. I don’t know if this is an actual problem for actual router designs, but I could see some circumstances where this could come into play.
Also, renovations could be a factor (this would be a subset of MikeMa’s large conducting surfaces scenario). Depending on the construction material, adding or removing walls could definitely change the propagation characteristics. If he’s in a multi-unit complex, this one could combine with interfering devices owned by his neighbors.
Interference and lack of a strategy to switch is being discussed over there. So, one recommendation might be to reset (turn off, unplug, replug, restart?) your router to see if that helps.
Power supply may be an issue. One reader noted that the power supply on his router … the brick… was putting out 3 instead of 9 volts. Of course, he did not check that voltage when he installed it, so that may have always been the case.
Makes me want to start testing voltage of my power bricks.
I’ve got an ancient wireless router my son picked up second hand years ago. It still works, and I don’t notice any degradation. Just one data point.
A radio transceiver does contain parts whose performance may degrade. For instance the choke coils could become permanently magnetised if they’re kept in the same position for a long time. I’ve no reason to believe this would cause performance to degrade in practice.
A Siemens router I had for about 7 years degraded to the point it had to be replaced. I suspect in that case that it was because it was kept on top of a cupboard (ceiling height) in the kitchen and just got too dirty from accumulated cooking grease
I’ve heard the same myth, from a not-too-technical neighbor, relating that he had heard from a salesman that wireless routers are engineered to fail after a few.
People are so suggestible… Knowledge is such hard work.
I too suspect that most cases of significant degradation are due to interference.
In some environments (like my apartment complex), changing channels doesn’t really help much due to just how many devices are around. Also, unless I’m mistaken, the 802.11 channels actually overlap (IIRC you can get 3 non-overlapping channels at most).
Also, the spectrum used for consumer wifi is the same spectrum that electronic devices are allowed to emit noise into. So it isn’t just other wifi devices that could be causing the problem… and we all know how electronic gadgets have been propagating.
As for actual degradation. It can happen I suppose. Crappy capacitors, ferrite chokes getting magnetized (which I hand’t though of, thanks Tony), and a few other things could happen. I strongly suspect that these very rarely cause noticeable degradation though. General flakiness, sure, but degradation, not so much.
I am a CCNA certified Network Engineer by profession. When dealing with large companies, enterprise routers are used. I’m talking about loud, heavy, large metal boxes ranging in price from $400 – $100,000+ for a single device. The life expectancies of even those devices are usually around 20-25 years. Compare these to your $50 dinky Best Buy routers, and the scale is probably accurate that if you get more than a year out of your home router, you’ve done well.
There are three differences:
1. Enterprise routers have built in fans to prevent overheating. Retail routers do not.
2. Enterprise routers are usually installed in a data room that is temperature controlled to be cooler than your standard 68-72 degrees.
3. The components in enterprise grade routers are of much better quality than the components within a home entertainment retail router.
4. In my experience, this has been the biggest home router degrader; every time you cut power and then restore power to the router, it has to go through reactivating all of it’s processes, and bringing itself back to full function. It’s a large burden on the components. So everytime the router loses power, you can expect it to be worse off when it comes back.
My advice would be to find a clean, open, cool space for your router, and perhaps buy a small clip on fan and let it blow on low towards the router while you can supervise. Most importantly though, when you turn it on, DO NOT kill the power to it unless it is completely unavoidable. You may even consider purchasing a UPS unit so that the router will not lose power in an outage.
I have had the same router for about 3 years. Recently my devices which would auto connect to it periodically won’t, even when I’m in the same room as the router. Its working as other devices are connected to it. But once I remove the power to it for about 20 seconds then repower my device will connect. This issue happens with ipad, 2 different android phones and 2 different PC’s. Router is a Linksys.
Lots of really good point made so far. I hadn’t considered that as more and more wireless devices get used they interfere with each other.
There is also the issue of inconsistent and ‘dirty’ power coming into the house via the power company (POCO). Down here in sunny, and stormy, Florida the power supply can have a lot of voltage spikes, sags, or drop-outs caused by lightning, power system switching, industrial power users, and, not uncommonly, heavy appliances within your own home.
I’ve seen an older AC compressor that threw a large voltage spike onto the line every time it turned off or on. The unit was short-cycling and always under heavy load. To get it to start reliably someone had stacked on capacitors. And, of course, when it shut down a defective relay meant the saturated motor coil discharged through the line. The homeowner had a long list of electronic that had gone bad in two years. TYs, stereos, microwaves and at least one computer.
One residence we worked on was directly down the line from an industrial welding shop. Another case where electronics didn’t last long.
Dirty power can cause a lot of wear. Voltage spikes cause arcing that poke holes in electrolytic capacitors, it degrades integrates circuits. Voltage sags cause motors, essentially constant power devices, to draw more current. This causes overheating and breaks down insulation. Momentary voltage interruptions cause electronics to have to reboot, forces ‘hard starts’ in appliances with compressors, and can cause overheating when fans stop before heat is dissipated.
My recommendation for Florida is that you install a large surge arrestor at or as near as possible to the power meter as possible. This takes the top off any large voltage spike. Follow this up by plugging your valuable electronics into a high quality surge arrestor strip. When you first install it note the month and year on the back in permanent marker. Change them out regularly. How often you change them out depends on how valuable the devices being protected are to you. Don’t throw them out prematurely. After a year a good quality unit might not be assurance enough for the computer working with important data. But that same arrestor likely has some life left in it and it will do for the computer your kids use to play Angry Birds on. Likewise the two year old unit on that computer can protect your microwave, or the TV in the guest bedroom. Once they have more than five years on them, or whatever you think is right, you can assume the surge arrestor function is shot and either toss them out or use them with power tools.
Voltage sags and interruptions are harder to deal with. The good news is that most computers store just enough power to protect the hard drives. The old ATs were notorious for bad things happening if you didn’t park the heads and shut down properly. That said, losing power doesn’t do your electronic devices any favors. Important electronic devices can be protected with an uninterruptible power supply.
If there is a pattern of electronics failing consider calling in an electrical contractor. Loose connections, more than one neutral-ground point, improperly sized conductors, overloaded circuits, sharing of circuits between electrically noisy appliances and delicate electronics can all cause problems. I got called in on an unrelated issue at an office and walked in in on a group of network techs having a fit because their network wouldn’t work. They had tried new network cards, different drivers, new cables, placing the wireless routers closer together. As a courtesy I checked the main panel, I had to go in there anyway. I found a couple lose neutrals and grounds. Once I snugged them up tight their network worked.
The power provide by most POCOs is remarkably smooth and reliable the vast majority of the time, and the wiring in most buildings is so consistent that it is easy to overlook problems coming in on the power lines or originating in the electrical system.
> consider calling in an electrical contractor.
Bingo. Make sure s/he will start by tightening the big setscrews behind the “electricians only beyond here” panel in the meter box, where the power line comes in and is connected. Those can slowly back out of their threads with repeated temperature cycling, reducing the contact area. That solved our problem. Only because I insisted it be checked after nothing else turned up.
Even electricians don’t like to touch those — they’re hot all the time unless the power’s been disconnected at the pole, they’re before the main circuit breaker/switch for the house.
Our electrician just used a big hex wrench with thick insulation on it, using her right hand only, left hand held behind her back — and each of the setscrewd creaked through maybe half a turn, tighter.
Problems gone.
Kids, adults, unlicensed people, do not mess with this.
Oh, and, a longtime expert explained to me long ago that semiconductors generally don’t fail flat out — they get a bit of a spike outside what they can handle, and the “semi” part — the level at which they change from insulator to conductor — drifts slightly when the spike hits, and doesn’t recover. Another spike, another drift — on that or some other component.
After a few years in a noisy environment, each of the affected semiconductors has drifted off its specification by some amount, so they interact — differently.
This was the explanation I got when I asked why so many “technicians” ignored the antistatic guidelines and said they’d been working on stuff for years and never broken anything by handling it. It’s because — it doesn’t happen that way, it’s slow and cumulative.
Same for noisy environment as for mishandling, it adds up slowly.
Could be. Or that could be some kind of religion. Hard to tell.
There is a variety of causes for the degradaion of electronics.
However the first consideration – is the problem environmental or the PC.
Are you being interfered with – I should be so lucky. No but have you recently had a neighbour go on line and using an interfering channel.
Have you recently had plasterboard fitted – possibly with aluminium foil to help with household heat retention.
One would suggest these causes could be quickly eliminated as they not cause a gradual degradation but could bee missed until there was a combination causing failure of the system.
Etcetera – Etcetera.
Anyway they are fairly easy to eliminate and now we come to the kit itself.
Three possible problems.
a) domestic contaminants
Cooking grease – nicotine, paint fumes ( depositing a possibly invisible layer of paint over the internals) excessive dust across terminals causing absorbtion of the signal.
b) Thermal effects on components.
Whilst it is unlikely that a domestic router would get hot enough to actually fail – any component would be likely to drift in its performance.
The networking bands are split up into numerous channels and due to the high frequencies involved and the narrow channel widths only a small shift in carrier frequency – perhaps 1/100 of a percent would push the frequency to between channels. _f the receiver does not drift to the same extent the system would suffer as the in channel signal would then be effectively be “overspill” of the ideal “clean signal.
The transmitter IC itself is unlikely to change its outpot. However this signal has to be fed to an aerial.
It has been proven that even small temperature rises cause the leaching of the plasticisers in the aerial cable. This could cause a change in characteristic impedance – hence a difference in conduction ability to the aerial itself.- a difference in the swr – standing wave ratio – and if this does not correctly match even if the signal is being generated in the transmitter it will never all get to be radiated.
c) Electrical effects
Whilst electrical noise would be more likely to destroy devices some degradaion would be caused and could cause out of band working and/or low output but I would not expect this to be gradual.as this type of noise damage would be expected to cause immediate damage to a device.
depending on how serious it may go unnoticed untill it had happened several times so could appear gradual but would in fact be a series of step changes.
Power levels. Whilst electrical these could be caused by heat effects on capacitors – gradually drying up and reducing their value thus reducing voltages and thus output levels. Low value capacitors could cause increased self interference – filtering capacitors failing and causing the various circuits to interact.
Equally resistors would tend to increase in value and the more than normal voltage drop across those in the power supply would reduce output. and any voltage regulators could change their output voltage.
Thus electronic device will degrade – either domestic or industrial.
The difference is the industrial kit is built with degradation in mind.
Temperature effects are “elliminated” by building in fans – contamination effects by installing in a “clean” room and usually the kit is regularly maintained and possibly even replaced without the client being aware of it as part of this maintenance cycle so even if it did degrade it would be be changed out before it became a problem. The client is interested in 100% performance- A good maintenance contractor fits 120% good kit and swaps it out before getting below the 100% performance rating – this kit is comparatively cheap – and would return to the manufacturer for overhaul.
Any or all of these effects could be present in a domestic location.
I work as a Geek Squad Agent part of my job is installing residential networks. Many of the routers I am replacing have started to degrade. One of the main causes I think is the heat. They are often times stacked on top of other devices and have little to no ventilation. At least once per month find a router that is so hot I can’t hold on to it. I also agree with the comment about the interference this is a common issues in apartment buildings and condos. The 2.4 GHZ band is such a crapshoot with all the other devices that use that same spectrum. Many of my clients with close neighbors I will recommend they go with the 5 GHZ band as their primary.
I find this interesting. I myself have been wondering the same this. As the above mentioned thread stated, it’s life cycle. I am a CCNA as well, worked on $10k+ enterprise routers and several WAPs (wifi for business). They just get old and degrade for some reason. My home router is about 5 years old. I get 12 mbps on average on wifi. My actual connection is 60 mbps which i hit easily on Cat5e. I Have done wi-fi interference tests and channel changes to no avail. Trust a tech when they say “that’s just how it is”. I just never bought a new one cause I don’t care is I get 12 mbps to my phone or laptop, that’s plenty for me.
The first post had it, I think… electrolytic capacitors. Of all of the components inside a router, they are the one most likely to fail, with the shortest specified service life, and the most sensitive to heat (and it gets hot inside the mostly unventilated consumer routers I’ve had). Such failures definitely can be responsible for gradual reduction of range (or speed) over time or other similar issues, as the noise floor gradually rises as the cap becomes increasingly unable to filter it out.
Caps can fail without visible signs, but chances are good that you will find them with bulging tops in any of these degraded-performance routers. If the device is in warranty, then RMA it, of course. If not, open it up and look for bulged caps– if you find them, they are definitely bad, no question about it.
If the caps inside the unit look good, perhaps the ones inside the AC adapter have failed, which would feed increasing amounts of ripple into the router. It can result in decreased stability and even shorten the life of components in the router that are substantially harder to replace than a simple capacitor.
If any electronic device fails or acts strangely, failed electrolytics is a pretty good guess. I’ve had a router with reduced speed, an LCD monitor that would not turn on at all. Both of them were fixed by replacing the bulged caps. I also have a PC PSU that exhibits behavior typical of capacitor issues, and it does have at least one bulged cap. I would bet replacing that cap would fix it too, and I may try it just to find out (I don’t actually need the PSU).
Replacing bulged caps (or unbulged ones that are many years old– you can test these with a capacitance/ESR meter) may or may not fix the device, but you won’t know until you try. Caps are cheap and relatively easy to replace, so why not, if the device is still worth using?
Yes this may be a case of electrolytics strike again. My most recent encounter was with a digital video recorder which runs pretty warm. In spite of this, the designers in their wisdom decided to sit some of the electros pressing against the heat-sinks of power diodes.
Rubbish design just never goes away.
i just trashed my linksys wifi router at work cause signal getting weaker,and just last month my netgear wifi router at home has the same problem..weaker signal overtime for a course of 3-5 years..yes routers do degrade overtime.they are kept clean,room temps 65-75f.both of routers have been bought 1 year apart..yes and both of them are the gigabit type wifi routers..
just to add,,at work i just replaced the linksys with the same model router but is kept in an old cabinet we found in one office..installed it and BAM!!! works like new again signal is strong again.so again routers fail !give it 3-5 years in my experience.
> that could be some kind of religion. Hard to tell.
Nope. I later discovered that the slow degradation I mentioned has a technical name in the semiconductor industry:
https://www.google.com/search?q=semiconductor+wounding
ElectroStatic Discharge : Practical Analog Semiconductor …
http://www.allaboutcircuits.com › … › Practical Analog Semiconductor Circuits
… an example of latent (“walking wounded”) ESD damage.
[PDF]Electrostatic Discharge (ESD) (Rev. A) – Texas Instruments
http://www.ti.com/cn/lit/pdf/ssya010
ESD can have serious detrimental effects on all semiconductor ICs and the system … There is no known practical way to screen for walking wounded devices.
In addition to ESD damage cited above, you can also experience breakage of the fine wires that join the semiconductor to its package pins, due to repeated thermal expansion/contraction. This can kill PCs that are turned on & off repeatedly — keep your PCs turned on.
You can also experience “dopant migration”, where the dopant elements added to the silicon to make them N-type or P-type migrate through & even leave the matrix, upsetting its semiconductor properties. This effect is increased with temperature, so it tends to kill things that are on all the time, but allowed to overheat.
Cold, steady temperatures are best.
Can this problem be addressed using a router booster?
Electronics can degrade, especially with temps and humidity affecting electronics. External influences of interference have to be considered as well. Adding smart devices, low energy bulbs that have bad circuits in them can cause severe interference. I bought several CFL bulbs when they came out, going to save the world you know. However the interference coming from them was atrocious and after discovering they were the cause they all had go and I replaced them with traditional incandescent bulbs. Personally, I never liked wireless anything, not wireless routers, Bluetooth, wireless mice or keyboards. They all have problems and nothing is as solid as a wired connection.