Faulty Electrics

Aug 26, 2004 39 Replies

Umm, this sounds like spectacular component failure of something storing rather more energy than yer average (or even two-standard-deviations-above average!) mains spike carries - maybe your PC/server supplier had a dodgy batch of hefty electrolytics, *and* had them mounted - unusually - direct on the board rather than enclosed in a bought-in power supply? Or maybe you had a lot of lightning strikes in your part of the world? Not to belittle the value of UPSen - my home machines and their less power-hungry peripherals typically sit behind one, as much for the convenience of a single power button to turn off all the little bits of kit as for the power conditioning and keep-goingness of the UPS - but the type of failure you mention is atypical, and unlikely to have been prevented by a (consumer-grade) UPS...

Stefek

In article , Handy Andy writes

It works like this. The chip amp that is in the PC speakers more often than not responds to frequencies way beyond the audio spectrum and usually well into the radio spectrum. The speaker leads act as aerials and bring all manner of RF "noise" back into the amp unit in the PC speakers where it gets demodulated and then presents as audio, hence the noise.

You may well find that same thing happens if you use your mobile phone close to them as well. You may well have noticed this effect if your mobile rings when using your land based phone which is about as bad for picking up RF "rubbish".

The Yamaha ones on my PC right in front of me now are quite good for this sort of thing......

Is it all lights or just one and if its one is it a flourescent light. These can often cause arcing at the switch contact, ruin the switch and then generate lots of RF energy that will probably be picked up by the amplifier/speakers

Well in 30 years or so of working with computers I've never ever seen that. I've seen the odd, single, fried component but that's all.

Hmm, they'd have to be ***** big spikes to have the effect of "literally half of the components had blown off the motherboard and other cards". A direct lightning strike could have that sort of effect but they're pretty rare.

We had a very severe over voltage at our previous house, it was due to a failed neutral at the sub-station. It blew a few lamp bulbs and the microwave. The only effect on the (several) PCs in our house was to make two of the switched mode power supplies fail, neither failed spectacularly, the just died quietly and silently.

You have to do something really serious to get any sort of damaging overvoltage through a PC power supply, it will die itself before anything of great consequence gets beyond it.

Well, I guess you are right because the ones I saw were all at the same site of a very large factory with some pretty heavy machinary and furnaces working 24/7. I have to say I was amazed to see it, but the cases were like rattles with all the bits at the bottom of them.

Replace 'spam' with 'org' to reply

I would also think it likely that they had some heavy duty arc welding equipment.

Colin Bignell

There has been an issue with certain cheap low esr electrolitics, that wern't and boiling, this could explain your server melt down

Never saw that myself when I was doing PC support. Not sure what value a UPS alone is if you're backing up. Cost of UPS vs. cost of components broken by spike vs. likelyhood of said spike occuring over the lifetime of components.

As I remember UPS's never tended to be too reliable - lot's of expensive battery problems.

top-of-the-range

Even when off?

I wouldn't.

Decemt PC speakers?

Is it more likely to be the

Try a new light switch, a mains filter on the speakers or new speakers.

capacitors conduct high frequencies, not block them....

top-of-the-range

Repeated power failures can cause extensive damage- no spikes. I had a 10mA trip in a new office that had not had computers in it before. The PC kept tripping it and eventually the board, processor, FD and DVD drives failed. I was blamed for using a kettle! New PC and new trip (30mA) and it's been fine.

Peter Scott

You can get small ferrite rings to reduce spikes and HF interference. You thread the signal cable through the ring a few times. As someone said earlier capacitors pass HF, so inductors block them. By winding the witres round you make a low value inductor.

Peter Scott

Certainly you can, but that isn't an appropriate response to the OP's problem.

Normal mains wiring and normal PC sound systems don't have that problem, so what the OP has is a definite *fault*. What you do with faults is find them, and then fix them.

There is a place for added ferrite rings and such (and I know it well) - but that place is not here.

Most external PC sound systems use an external amp too - and this is far more likely to be the source of the problem through an inadequately filtered power supply. PCs themselves generally have a well filtered PS for obvious reasons.

Out of 3 systems I know of, 2 do.

The problem is a design one. Almost all suggestions have been right off track. Theres nothing you can do about it unless you mod the design on the audio side, or buy some decent kit.

This is uk.d-i-y not sci.electronics.

Regards, NT

You can buy plug in mains filters.

It would be sad to exclude simple electronics.

However, it would help the the OP actually read the replies and answered the points arising.

If the same thing happens when using headphones direct from the computer, then I doubt external filtering on its supply will help. But if it stops, then the fault lies with the external sound system.

Can you say that in English please!

I think he meant low ESR electrolytics which didn't have a low ESR, as for the boiling, he's probably talking about the cup of tea required to enhance the effect of the sharp intake of breath

I get exactly the same on my computer speakers, it's especially evident when the humidistat fan in the bathroom spends 20 seconds or so deciding whether or not to switch off.

If I put my hand near, or touch, the very thin (2mm maybe) signal cable which takes the audio from PC to speakers while this is happening, then the effect stops. If I remove the signal cable from the speakers, the effect stops.

The cable is basically acting as an aerial!

I keep meaning to either replace the cable with one with better quality screening, or maybe put a ferrite core on it. But it's not a major problem.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required