tp-link router died

Thanks, but I don't think a sample size of 1 as going to influence me one way or the other.

Apart from that, you havn't given us the model number.

In message , Graham. writes

Indeed, my TP link wireless AP has been running ok for a few years now

year. It was working but I could not access its http console (visitor want ed the pass phrase), so I restarted it (power cycled). After that it acted as an ethernet switch but nothing else. It was as if the control part of th e device had failed.

resets possible.

Oh I didn't think I'd bother. I didn't think a sample size of 1 was going t o influence you one way or the other. Don't you recognize and excuse to ven t ? (model TD-W8961ND Ver 3.0 S/N 12986703642). A sample size of 1 may be statistically insignificant, but there is a signi ficant chance that if something happens, it happens more than once (because its due to an underlying design factor). Or is it because I split my coffe e in it ? Simon.

a year. It was working but I could not access its http console (visitor wa nted the pass phrase), so I restarted it (power cycled). After that it acte d as an ethernet switch but nothing else. It was as if the control part of the device had failed.

presets possible.

to influence you one way or the other. Don't you recognize and excuse to v ent ? (model TD-W8961ND Ver 3.0 S/N 12986703642).

nificant chance that if something happens, it happens more than once (becau se its due to an underlying design factor). Or is it because I split my cof fee in it ?

Those few typos make me sound more angry than I am. My contact lenses are g etting a bit blurry. Simon.

And mine 18 months + ...

I'm sure mine failed earlier than most, but I was still not inclined to go out and buy the same model again ! Simon.

My TP Link router replaced my Belkin router which died earlier this year but as each of them only cost about a tenner I tend to think of them as consumables like paper and ink. I could buy a really nice router for Belkin/TP price x n where n is a big number but I'm not convinced a nice router would stay alive n-times longer than my cheapies or even that computers will still use the same standards in n-times years time.

Nick

I used the same logic when buying wrist watches, preferring supermarket cheapies over the likes of Omega etc. If it breaks down, replace it. Now I don't even use one.

Hmm, most odd. I thought the innards of these things tended to be the same just badge engineered tese days. Brian

year. It was working but I could not access its http console (visitor wante d the pass phrase), so I restarted it (power cycled). After that it acted a s an ethernet switch but nothing else. It was as if the control part of the device had failed.

esets possible.

I've given up with my TP-Link WR1043D after about 18 months - it falls over if it gets slightly too warm and has a number of irritating minor bugs, fo r example some, but not all, internet radio station start buffering after h alf an hour or so. I assumed it was issues with the streams, but having swi tched back to my bombproof old Netgear DG834G, that streams all stations f ine and doesn't fall over if the ambient temperature reaches 25C or whateve r. The TP-Link was very cheap for the functionality but looks like for once I got what I paid for.

I suppose as this is Uk d-I-y I ought to flash the TP-link with DD-WRT or s omesuch third party firmware but frankly I don't have that much time to was te.

In article , snipped-for-privacy@gmail.com writes

Check out the voltage of the plug top power supply under load, I've had a couple fail early. They show fine off load, it was the old bad caps story

Thanks for the tip, might try running it off a proper stabilised PSU for a week then and see what happens.

Mine did have a bug where it would drop the wireless connection. However, I saw a beta firmware release mentioning it had fixed this (before I'd fully realised - I thought it was my laptop being silly) and it's been fine ever since.

Personally I'm quite impressed with the feature set on mine.

Yup had a bunch of D-Link switches fail for just that reason (internal PSU rather than soap on a rope, but same principle and fix).

Probably had half a dozen TP-Link switches in the field now for a few years - not had any failures.

That's a bit of a sweeping statement. It's surprising just how different they can be. However, you do see badge engineered versions of Zyxel(?) and Netgear kit re-marketed by the likes of Peak and (spit!) Belkin.

If I had to make do with a badge engineered version, I'd choose Peak over Belkin every time. The hardware might be the same but the OEM customer (Belkin) will usually employ someone to write, in Belkin's case, a s**te version of the firmware so full of security lapses as to turn a moderately bad piece of kit into a festering PoS so bad that anyone with a moderate level of knowledge wouldn't want to touch it with a bargepole.

====snip====

If there's a DD-WRT firmware available for that model, I'd be taking it apart to replace any suspect caps and enlarge any existing vent slots and adding a couple of square inches worth of ventillation holes in the base of the case (7 to 10 mm drill bit sized) and, if necessary, drill a few in the top of the case to supplement the existing vent slots if the effective CSA can't be increased to match the holes in the base[1]. _THEN_ and only then, would I deem it worthy of such a firmware upgrade.

In case its any use, some info on DIY cap replacement etc:

That's actually an excellent observation. That could well be the root of Simon's problems with the TP-Link.

However, since the use of switching regulators in these cheap commodity routers and switches has displaced those ancient analogue voltage regulators for well over a decade now (purely on cost cutting grounds), it's a simple enough problem to fix since the only matching requirement of a substitute wallwart is plug size and VA rating over a voltage ranging from a low of 6 to a max of 15 volts DC.

The switching regulator will (unlike the specials used in laptop power management switching regulators) cheerfully accept a wide range of input voltages equal to or greater than the output voltage threshold (typically 5 volt) that's within the upper input voltage limit, usually defined by the input capacitor voltage rating (conservatively, a maximum of 15 volts).

You can safely power 99% of all such kit from any 10VA rated nominally 12 volt wallwart that happens to have a suitably matched plug on the end of its DC output lead (if necessary, you could always use the plug off the old wallwart to fix this issue if needs must).

====snip====

If you've got a variable voltage bench psu to hand, you can start off at the 'rated' input voltage and slowly wind it down, observing the current draw. It's extremely unlikely that you won't observe the negative slope in the volt/current plot that's characteristic of a switching regulator.

If you observe this, you can then increase the voltage above the 'rated' voltage, perhaps limiting it to a max of 15 volts just to remain on the safe side of the input cap's voltage rating.

Once you've ascertained the use of a switching regulator by the wallwart powered router or switch, you'll know for sure that it's safe to make use of almost any old wallwart with a sufficient VA rating as a replacement for the originaly supplied wallwart.

The only remaining issue being the question of the DC lead's jack plug size. IME, such jack plug issues don't occur all that often since there seems to be a singular size favoured by the makers of such kit (unlike the laptop makers who deem out of warrantyreplacement mains charging bricks to be their version of inkjet refills).

HTH