amp new PSU capacitors

[sm_jamieson]

My approx 9 year old AV amp started buzzing / humming sounds like

100Hz, slowly got louder until drowning out the music ! PSU caps I should think. They are labeled "kenwood for audio". I've got the bigs caps out, they are 10000uF 80V. Lots around for about 20quid each, or some ebay ones which I try to avoid if possible. Does this look OK ? the life seem a bit short (I know this is at the max temp) ?

Also saw this but the lead spacing and layout is wrong:

did not realise the life was so limited though. I've left the amp on

24 hours a day on standby for years !

Also a couple of smaller 3300uF 50V jobs, I guess for a 5V or similar supply for the ICs, DSP chips etc. Probably worth changing those as well.

OR .. Is this really worth my time and up to 50 quid fixing ??? A new replacement would cost 350 quid but would support blue ray audio, HDMI etc.

Simon.

[Dave Plowman (News)]

At that price I'd want to be certain it was those caps. Can you isolate the main amps or perhaps short the inputs to be certain it isn't coming from elsewhere?

I've got several amps here a lot older than that on the original smoothing caps.

[Tabby]

Electronic equipment is complex in terms of the number of possible faults. Guessing is a poor repair strategy, especially when it costs =A350 a guess. Lytics normally last many decades, though there have been some duds on the market causing problems.

The simplest way to test the caps is to measure the ripple on them with the power on. If you dont have the proper equipment, you can use a mmultimeter on ac volts setting, connecting it to the big cap via a small 1uF non-polarised cap (ie polyester, not electrolytic). Once the meter reading settles down from 80v or so, you should get a steady reading of the ripple.

Its more likely to be something else than the caps.

PS what sort of amp would/could fail to support blueray?

NT

[The Natural Philosopher]

Seems overly expensive for reservoir caps.

Definitely worth fixing tho.

But you shouldn't be paying much more than a fiver..try fleabay.

[The Natural Philosopher]

Just solder a new 80v cap of almost any flavour over 1000uf across whats there.

Identical issue in 50 year old valve radio. Fixed it.

Bad caps abounded 7-10 years ago.

They all dried out..

[The Natural Philosopher]

?

> Does the life seem a bit short (I know this is at the max temp) ?

wrong:

>>> I did not realise the life was so limited though. I've left the amp on

It isn't.

Either a bad solder joint to the cap or the cap itself has gone high resistance..but the classic 'it slowly gets worse' is almost 100% sure to be the cap.

Easily tested by temporarily connecting something of adequate volatge rating across it.

Caps are just about the only components that fail gradually and non catastrophically.

Those and valves/CRT tubes.

Everything else tends to be either so low stress it lasts almost indefinitely, or so high stress that it goes bang and the set goes dead.

[sm_jamieson]

There are some caps (more expensive) marked "for audio" etc. Is this just marketing, or is there anything different ? Surely in the power supply they cannot make much difference ? I am sceptical myself. Simon.

[sm_jamieson]

The way the circuit boards are layed out with various daughter boards, ribbon cables etc, its very hard to have the circuit working and out of the case. I would really have to change the caps and put the whole thing back together to test it. Simon.

[Dave Liquorice]

One that doesn't have HDMI... A nine year old AV amp just predates HDMI 1.0 which came out in Dec 2002.

A graceful failure into hum is almost certainly down to dead or dying reservior capacitors.

80v is a bit of a odd rating, 63v or 100v are the normal ones.

CPC have: CA05298 10kuF 100v =A317.03 CA05334 10kuF 100v =A315.01 "computer grade" higher ripple current capability than the above.

[Archivist]

Quote: Surely in the power supply they cannot make much difference ? I am sceptical myself.

You're right to be sceptical of all 'golden ear" claims. However if you dra= w an "equivalent" circuit of any amplifier, you will see that the power su= pply shows as being in series with the load. In other words, the signal ret= urn from the speaker goes through the power supply. This means that you sho= uld choose ps components as carefully as you would for example speaker cros= sover components or any other component in the signal path. Two more points:

1. You don't need expensive caps, nor the exact capacitive value to test yo= ur theory. Hang the cheapest on temporarily.
2. I personally wouldn't spend anything on equipment of that era...

J

[John Williamson]

Because of the way they're made, electrolytics may have a significant self inductance. I remember in the past buying low self inductance versions for critical applications, which cost more.

[tony sayer]

Extend the leads from the replacement cap to the one inside the case for test?..

[tony sayer]

In article , Archivist scribeth thus

Well you might want to re cap something like a QUAD 303 which is a very worthwhile project around 30 40 odd quid for a replacement set:)..

We've got one here for a study setup, mainly plays audio off the PC soundcard a much better external one that that usually fitted. Still sounds fine :)..

[Dave Plowman (News)]

I'm using one to drive a pair of subs in the living room. It's never been touched. Over 40 years old. ;-) No hum from it - although it's quite possible it is out of spec in one way or another.

[GB]

Why would these need to be 80v out of interest? What sort of DC voltages do these amps typically operate on? I know there's a margin for peaks, but 80v across a nominal 8 ohm resistance would provide 800w. This is an amp, not a room heater, surely?

[sm_jamieson]

Ouch. Others posts were suggesting I should be paying about a fiver ! Anway, I like the units ! "kuF" ! You should have quoted the price in deci-mega-pennies ;-) Simon.

[The Natural Philosopher]

generally able to handle high peak currents and maintain low resistance.

[The Natural Philosopher]

an "equivalent" circuit of any amplifier, you will see that the power supply shows as being in series with the load.

No, you wont.

supply. This means that you should choose ps components as carefully as you would for example speaker crossover components or any other component in the signal path. No, it doesn't.

theory. Hang the cheapest on temporarily.

At least you got that one right.

20 quid for a pair of new caps is not a lot.

[The Natural Philosopher]

No, low series resistance. If inductance is an issue, you put a smaller cap across them.

That takes care of HF impedances.

>

[sm_jamieson]

I thought about something like that. I'll probably solder in some type of connector so I can replace the capacitors easily from the front of the board. Then I can stick the existing one back in temporarily and see if a small cheapy cap in parallel reduces the hum Simon.