What's new

Do Routers Get Slower As They Age?

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

Bulldog

Regular Contributor
My Asus wireless router [RT-N56U] is about two years old. Throughout this time it has been paired with an Asus wireless client [USB-N53.] Both router and client are 802.11n and both operate on the 5GHz network.

Until recently my downstream internet speed with wireless (from Ookla) was about 115 Mbps. However, recently it's dropped to about 85 Mbps and I've switched to Ethernet because it's faster.

Yes, I've re-booted the router. Firmware and drivers are up-to-date.

I know there are all kinds of ways to measure internet downloads and you can only put so much faith in speed measurement websites. It's the relative decrease in Mbps that concerns me. Doesn't make sense. So here's my question and pardon me if it's dumb:

Do routers and/or clients get slower as they age?
 
Slower with age? Not due to electronics, with a few very rare exceptions such as a semi-fault CPU clock oscillator.

Some reasons you might see slowing:
  • You are using your ISP's DNS server (the default) and it is slowing. Cure: change router settings to use a DNS like 8.8.8.8
  • ISP service declining. Use SpeedTest.net 6 times, different days, take an average. Prove you are getting what you pay for.
  • WiFi slows due to new neighborhood congestion and competition for air time on the channel you are using, or within 2 channels. Cure: Try changing (on 2.4GHz) to 1, 6 or 11.
  • WiFi client has weak signal
  • WiFi slows due to non-WiFi signal interference. Cure: change, as above.
  • Malware on computers that burden the CPU. In windows, look at CPU usage with performance monitor within windows.
  • Ethernet switch semi-failed - excessive blinking (traffic indicator) on one or more ports. Cure: replace switch
  • Ethernet cable flaky. Replace

Other possibilities a plenty.
 
Thanks very much for your help. I can't imagine why a router would get slower with age. I only thought of that because my previous router (DLink) got very slow before it finally died.

After checking I discovered that other wireless devices on my home LAN are still downloading very quickly. This suggests to me that the Asus wireless client [USB-N53] is behind the slower wireless speed I am observing. This is a refurbished device which Asus sent me to replace a same-model client that stopped working altogether.

Not blaming Asus necessarily, but you can only expect so much from consumer electronics.
 
Those USB WiFi adapters are dirt cheap now.
I recommend using a USB extension cable so the adapter can be elevated for a more favorable RF location.
 
Aging electronics can cause the router radio signals to lose in quality, which could translate in an increase in packet retransmit. An aging capacitor no longer properly filtering electrical noise in a filter stage, for example.

We're talking about devices that are getting quite old there however, not something that's just 2 or 3 years old.
 
Thanks to both of you for your help.

There's little doubt in my mind that my wireless client [the Asus USB-N53] is the culprit here. Switching between my wireless client and Ethernet produces very clear differences.

EDIT:

And I am wrong!

Here's what was causing my wireless speed to drop: I was using an adapter [from Belkin] between my wireless client and its USB port; this adapter bends so that my wireless client sticks out from my computer at a 90-degree angle. I thought it was a smart idea to do that because it meant that my wireless client 'faces' my router instead of a wall. I guess I'm not so smart because once I removed the adapter and connected my wireless client directly to a USB port, my internet speed came roaring back.
 
Last edited:
That Belkin USB swivel adapter.. faulty or badly designed, e.g., rather high current consumed by WiFi client may not be accommodated in the swivel adapter's wiring.

If your signal strength is not excellent, I do again suggest getting/using a good quality USB extension cable (some have wire gauge too small) - to get client device/antenna better positioned.
 
SteveCH and rMerlin have brought up good points and I agree...

Outside of the SNB/WiFi/Router/AP stuff here - I am a home audio hobbyist - and I've had to re-CAP several receivers/amplifiers just because capacitors have a life cycle - they're made/they work well/they die...

We've seen the first hand in our little group - one of the big factors is cheap wall AC Adapters - you'd be surprised when they start to fail before going down hard - rather than relatively clean DC, they start putting spurs into the DC rail, introducing noise into the Power Supply Rails - this impacts things in both the digital, as well as the analog domains...

Because this is a sub-one dollar part, and one of critical importance, most folks buy a new router at 40 to 250 dollars, rather than replace the AC adapter at a few dollars...

And perhaps you get better performance, as Moore's Law says it will be as such...

just saying - just know that for the most part, the SoC/WiFi MAC-Baseband/PA's along with the Ethernet switches - this is totally silicon - it's mostly power supplies that cause stuff to die... and the analog components to some extent - and there it's most electrolytic capacitors...

sfx
 
Any electrical device that generates heat has the capacity to wear out. I generally stay away from devices that reviewers remark are 'running hotter than expected'. While the engineers who made the device probably considered heat in the design, all it takes is one slacker to mess up everyone else's work.

The remarks about capacitors as a failure point are spot on. Since all devices use them, it's logically a non-issue to worry about when picking out a new device. Unless they run hot.

Also, some devices are just bad out of the box, only you don't know about it for several years. I have an Onkyo AV receiver that may have a factory defect that takes years to appear. Onkyo just 'allowed' a recall of 'some' devices made from 2009 through 2012 because of audio problems. Still looking to see if mine (with problems) is on the recall list.

Electrical failure's don't all look alike. An intermittent failure may look like a slowdown if it's short in duration.

ME, I always buy the cheapest acceptable electrical item and treat it as disposable. I will cry less about a $70 refurb AC router than I will a $200+ state of the art powerhouse if it breaks.
 
Last edited:
aging could be the wire or wifi chips from heat. Motherboard links use copper and the circuits themselves use silicon. silicon only wears out to really extreme stress and the last reported one was from overvolted pentium 4s. Electronics will last for many many years, more likely the cable between you and your ISP could've degraded or your wifi chip could've degraded as RMerlin said.
 
Depends on the hardware used inside this devices. If made well the device can last a very long time. Once you started to stress it out and push the firmware beyond the OEM limits then your shorting the lifespan of the hardware.
 
I disagree. Choice of firmware doesn't affect life.

The wall transformer is often the first to soft- or hard-fail.
 
I disagree. Choice of firmware doesn't affect life.

The wall transformer is often the first to soft- or hard-fail.


It does if the firmware has glitches in the code. That could effect the performance of the router.

DD-WRT had a feature that would overclock WiFi amp like in Buffalo G54 HP. I had told the author of DD-WRT that pushing the WiFi amp over 70 would hard the router. He did agree and removed the feature.
 
Aging electronics can cause the router radio signals to lose in quality, which could translate in an increase in packet retransmit. An aging capacitor no longer properly filtering electrical noise in a filter stage, for example.

We're talking about devices that are getting quite old there however, not something that's just 2 or 3 years old.

This is true of electrolytics and especially true considering many electrolytic caps have very poor dielectric fluids. For the last decade motherboards have been plagued with crummy caps. I had to replace the caps on my Linksys RV-042 router due to the caps being of very poor quality. They were used on a DC-DC voltage converter and their instability caused the 5V rail to be noisy.

I suspect this router uses a switch-mode power supply as a wall wart. If there is noise, it will be there. Easy test, try another compatible supply. I have had this issue with many other devices using similar construction, including a Silicon Dust HDHomerun tuner and a cheap Android tablet power supply.
 
one of the big factors is cheap wall AC Adapters - you'd be surprised when they start to fail before going down hard - rather than relatively clean DC, they start putting spurs into the DC rail, introducing noise into the Power Supply Rails - this impacts things in both the digital, as well as the analog domains...

Because this is a sub-one dollar part, and one of critical importance, most folks buy a new router at 40 to 250 dollars, rather than replace the AC adapter at a few dollars...

- and there it's most electrolytic capacitors...
sfx

Electrolytic caps in switching power supplies get the cwrap beat out of them and unless they have very low ESR (equivalent series resistance) which makes them expensive, they will fail eventually. Cheap wall-warts are generally sealed so you can't see the crappy quality caps inside. First rule of troubleshooting: Check the AC connection. It isn't always easy to do with non-standard DC plugs and voltages.
 
Electrolytic caps in switching power supplies get the cwrap beat out of them and unless they have very low ESR (equivalent series resistance) which makes them expensive, they will fail eventually. Cheap wall-warts are generally sealed so you can't see the crappy quality caps inside. First rule of troubleshooting: Check the AC connection. It isn't always easy to do with non-standard DC plugs and voltages.

This is actually a small part of the reason why I swap the power supply on my router (I'll grant, I am living with the wall wart on my AP). I use a Belkin AT&T Uverse UPS that I got for $25 shipped. First off, at the modest power draw levels my Archer C8 has, it has about 12hrs of run time on the battery backup...but it also outputs pretty smooth DC power (I checked it on an oscilliscope). At some point I'd love to get a second one to plug my AP in to, though the downside is that is rather large and bulky (same foot print as the AP, but roughly 4x the thickness). The battery is also pretty easy to replace in the UPS and is only a ~$30 battery. Also again, because of the low power draw, unless the thing up and fails entirely, even with old age it should be able provide at least an hour or two of battery backup for years and years before the battery in it needs replacing.
 
overclock WiFi amp like in Buffalo G54 HP.
The amplifier is an analog device. One cannot "overclock" as there is no clock to the amplifier.

The low level manufacturer firmware will ignore requests from higher/user interface code for more power than can be supported for the necessary distortion for any one of the modulation rates. No hardware harm, but if this were not done, the transmitted signal would cause much higher error rates at the receiver, for some modulation modes.

But this is a different topic than the thread title.
 
The amplifier is an analog device. One cannot "overclock" as there is no clock to the amplifier.

The low level manufacturer firmware will ignore requests from higher/user interface code for more power than can be supported for the necessary distortion for any one of the modulation rates. No hardware harm, but if this were not done, the transmitted signal would cause much higher error rates at the receiver, for some modulation modes.

But this is a different topic than the thread title.

Well, to be technical there are analog devices that have different clock rates, but an amplifier is not one of them. It just amplifies the incoming analog signal by a certain amount. Heck, a clock source is an analog source, and that can certainly be changed to a higher frequency, which would be "overclocking" it. Technically speaking.

One would say increase the amplifier power though, not overclock.
 

Similar threads

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top