Main Area and Open Discussion > Living Room

A Problem for A Real Expert - the Haunted Router(s)

<< < (4/5) > >>

40hz:
I'm going to have to cast my vote for gradual heat damage being the primary culprit.

Power surges to the actual router shouldn't be too much the issue since your devices all use "wall warts" for their power supply. Although these devices aren't surge suppressors or line voltage conditioners, they do provide a fair amount of protection to the devices they power. So unless they've been damaged, I'd look elsewhere.

If you check the output coming off your router's power supply with a multimeter you should be able to see if the correct voltage is being supplied. I'd be especially on the lookout for specs that read lower rather than higher. Low voltage can often cause a device to run hotter than it should.

The situation could also be brought on purely by environmental heat factors. Either the router is getting too warm by being where it is; or the power adapter is providing the wrong (likely low) voltage since many power devices will produce lower supply voltages when they overheat.

Suggestion: if you do test your adapter, try running it in its normal environment for an hour or two and then try to test it immediately after you disconnect it from the router. If there's a bad voltage condition, brought on by environmental or runtime factors, this will help you catch it.

So either way, I guess I'm going with heat damage as my answer. ;D

Luck! :Thmbsup:

4wd:
If you check the output coming off your router's power supply with a multimeter you should be able to see if the correct voltage is being supplied. I'd be especially on the lookout for specs that read lower rather than higher. Low voltage can often cause a device to run hotter than it should.

...

Suggestion: if you do test your adapter, try running it in its normal environment for an hour or two and then try to test it immediately after you disconnect it from the router. If there's a bad voltage condition, brought on by environmental or runtime factors, this will help you catch it.-40hz (July 15, 2009, 07:03 AM)
--- End quote ---

According to Linksys' site, the PSU for that router is switchmode so you won't be able to get an accurate reading unless it's under load - and it still won't tell you how much ripple is on the output which is a problem when the capacitors in a switchmode start going high-ESR due to, for example, heat strain and voltage fluctuations.

It would be easier to replace the PSU with a known working one.

Shades:
Now those old but very handy oscilloscopes could come in handy (to find out if and how much dirt the input/output takes and puts out)...too bad they always were so damn expensive, which is also true for the recent digital ones.

40hz:
It would be easier to replace the PSU with a known working one.
-4wd (July 15, 2009, 08:09 PM)
--- End quote ---

True. You could add a dummy load, but that really wouldn't be all that accurate either. Still, a 20Ω @5W or 10W resistor in parallel with the output should provide enough of a load that you could do a quick & dirty multimeter check to see if you're at least in the ballpark.

Then again, it would be easiest to just try a different PSU.

-------------

Sean (one of my hardware tech buddies) pointed me towards this article, which discusses wall warts in detail, along with a project to safely add voltage regulation and filtering to an inexpensive plug-in power module.

So, since we usually want to power an auxiliary device at a steady specified voltage and since we usually need a very clean power supply, what I wanted Bill’s help on was in creating a filtered, regulated power DC supply. We accomplished this by adding a module between the wall wart and its DC output plug that contained a fixed or adjustable voltage regulator and a network of filter capacitors.
--- End quote ---


Link to article: http://www.dxing.info/equipment/wall_warts_bryant.dx

The circuit and construction are well within a hobbyist's abilities. Sean swears by the fixed voltage version of this project. He's used it to get solid performance out of otherwise flaky consumer-grade router/switch products.

Might be worth a read. :Thmbsup:

(I'm going wire one up and give it a try on an RVS4000 that's been giving me some intermittent 'attitude.'  I figure it's got to be cheaper than taking a hammer to the little bugger. Which is something I've been tempted to do lately.  ;D)

4wd:
Sean (one of my hardware tech buddies) pointed me towards this article, which discusses wall warts in detail, along with a project to safely add voltage regulation and filtering to an inexpensive plug-in power module.-40hz (July 15, 2009, 09:32 PM)
--- End quote ---

That article is for your basic transformer style plugpack.  The circuit that is detailed is normally, (just to clarify - if the manufacturer wants a cleaner supply he may incorporate a simple switchmode design in the device itself but generally a lot of equipment uses basic linear regulators, eg. 7805, 7812, etc), what you find connected to the power input socket of the device - so in effect you are connecting another circuit that's the same in series.

The only problem with that is you've just disabled the internal voltage regulating circuit of the device.  The regulators mentioned in that article require at least an input voltage 1.2V above the required output voltage in order to regulate correctly otherwise they just drop out.

eg.  If your device requires 12V DC there will typically be a 7812 regulator connected to it's input power socket - it's relying on the normally unregulated voltage provided by the transformer plugpack of around 16V DC allowing it regulate correctly.

If you build the circuit as described and set its output to 12V DC and then plug that into your device, where does the internal regulator get its required voltage drop in order to regulate?

I would say he has success probably because of the extra filtering he's added by way of the capacitors.

Switchmode PSUs are self-regulating and don't require it, they only require the requisite load for regulation - and the latest versions don't even require that.

EDIT: Also note that article is from 2005, switchmode plugpacks have become more prevalent in the last 2-3 years.

Addendum: @raybeere: If the PSU is indeed a switchmode type, (they're generally lighter than transformer types is the easiest way to tell), then you should change it for the same type with an equivalent rating if you intend to.  The unregulated output from a generic transformer based PSU might damage the router, (unless the replacement specifically states it is regulated).

Navigation

[0] Message Index

[#] Next page

[*] Previous page