Smoke

[unitaudio]

I've got a Rotel power amp that one day recently gave a quiet pop and started smoking. It carried on working but obviously I switched off fast. On trying it again it worked perfectly but a cursory glance inside didn't even show any sign of where the smoke came from. It was, as I said, only a cursory glance but reading this thread convinces me this is the problem. I don't yet know what make of capacitors are used in the amp but I'll have a look (probably next week).

Regards,
Paul

[threeseven]

Being Japanese it won't be a RIFA cap. Although no cap in that application can be expected to last forever. Modern X & Y rated caps will fail safe though, instead of exploding!

[camtechman]

Rifa....Smoke....the mind boggles !

[threeseven]

Wouldn't know what you mean

[m0cemdave]

They were probably designed in the 1960's, when such things were intended to go up in smoke...

[kalee20]

Originally Posted by threeseven

Modern X & Y rated caps will fail safe though, instead of exploding!

Originally Posted by m0cemdave

They were probably designed in the 1960's, when such things were intended to go up in smoke...

I am fairly sure that Rifa still make these capacitors. They certainly did in the 1980's, with all the approvals on the side. And the 1990's. I just haven't looked as I have avoided them since a few failed on me early on in my career - on the bench luckily, not out in the field.

They are dreadful.

[Leon Crampin]

It's "Rifa" not Reefer.

The Japanese makers did use these as mains filters because of their approvals. I've found them in all sorts of unlikely places including a set of Swiss made horse clippers. Smoke here is a complication I would not care for...

Leon.

[Boater Sam]

Also used in Siemens electricity meters, made I think, by Landis & Gyr. They explode and take out the 100 Ohm resistor in series. Leave behind the usual filthy mess too.
Unfortunately this also alters the meter readings drastically and I have not worked out how to re-set them.
So much for approvals printed all over one side!

Boater Sam.

[Billy T]

Originally Posted by kalee20

I am fairly sure that Rifa still make these capacitors

They certainly do, as 'EVOX RIFA" they remain a catalogue item, along with a variety of blue and green encapsulations.

To be fair, they do not normally die prematurely, but they still don't enjoy a long and peaceful life. They are probably good for 15-20 years in a reasonably benign environment, but I come across them in 20-30 year old industrial gear.

Cheers

Billy

[Radio Wrangler]

The capacitors seem to be the limiting factor in quite a lot of equipment and they fail very dramatically. So they are becoming associated with equipment failures.

They might last a reasonable amount of time, but they cause collateral damage and frighten people.

The one in my sig gen burnt a hole through the metal can of a Schaffner mains inlet, so in Leon's case of horse clippers, I don't fancy dropping burning clippers into the straw in a stable.

One of the approvals hieroglyphs on the casing is UL, Underwriters Laboratories who grew out of the fire insurance associations in America. They are VERY interested in any failure mode which can initiate combustion, and at any age.

David

[AC/HL]

They are the first line of defence in a very dirty environment. For every one that fails there must be countless instances of much worse damage being prevented.
I somehow doubt that most of the Worlds standards agencies are completely wrong.

A new wonder plastic that completely eliminates the fire risk would be nice, but as other capacitor manufacturers have found, new plastics can have their own Achilles Heel.

[Radio Wrangler]

Unfortunately the most promising fireproofing agents for plastics got listed in ROHS...

The paper ones were designed to survive a flashover and heal shorts, which they do.

RIFA are certainly legit with these parts, but their name is getting a stinker of a reputation

David

[daviddeakin]

Exactly the same thing happened to me when I applied a reverse voltage to a small electrolytic capacitor a few years ago. Not knowing my mistake, I left it running for a few minutes. Suddenly there was a great gush of acrid blue-ish smoke spewing from this tiny capacitor like a stage effect in a heavy-metal concert. I thought it was never going to end! Stank something terrible for hours.

[julie_m]

Originally Posted by Radio Wrangler

One of the approvals hieroglyphs on the casing is UL, Underwriters Laboratories who grew out of the fire insurance associations in America. They are VERY interested in any failure mode which can initiate combustion, and at any age.

Hardly surprising, in a country with a lot of wooden buildings and 120 V (therefore, more amps per kW than the 230 V world) electrics .....

[Leon Crampin]

In terms of approvals, I can tell you that in my experience, Approvals Houses are clueless as to the real failure mechanisms of components. Working in the automotive sector in the '70s, it was obvious to us that they had no understanding whatsoever of the real issues.

Thermal fatigue, mismatched expansion rates and poor mechanical design were just not revealed by their procedures. We designed our own tests to sort them out.

One of my best tests was to wire 30 x 1N4002 is a series string, suspend it from the ceiling and hang a weight on the bottom - equivalent to the maximum allowable axial lead pull. The diodes were then subjected to maximum forward current as a string - 30 secs on, and 30 secs off. With a counter in the circuit, this continued to failure. The remaining diodes would then go into the dishwasher and then be measured for reverse leakage.

Many components with approvals failed miserably. The glass passivated under plastic components were better - the Japanese the best.

Failure of these diodes was a potentially serious business for us as so many were used as flywheel diodes across solenoids - eg. in 'bus gearbox controllers.

I guess with the "Rifa" test, I earned my month's salary...

Leon.

[Maarten]

You certainly did.

Originally Posted by Billy T

To be fair, they do not normally die prematurely, but they still don't enjoy a long and peaceful life. They are probably good for 15-20 years in a reasonably benign environment, but I come across them in 20-30 year old industrial gear.

While not prematurely, they do fail completely unnecessarily. Better dielectrics and encapsulations already existed in the 1980's, let alone now.

Originally Posted by AC/HL

They are the first line of defence in a very dirty environment. For every one that fails there must be countless instances of much worse damage being prevented.
I somehow doubt that most of the Worlds standards agencies are completely wrong.
A new wonder plastic that completely eliminates the fire risk would be nice, but as other capacitor manufacturers have found, new plastics can have their own Achilles Heel.

Such plastic is not needed in capacitors in which faillure modes are reasonable (according to the IEC and UL regulations, anything but opening up or shorting out is deemed unreasonable). The first line of defense means that a capacitor in an electrically dirty environment will open up slowly, or short in extreme cases but not that it will become a source of damage itself, damage that is larger then without the capacitor. No more, no less. I'm also at a loss as to why they were approved, but apparently the tests devised by boards that should know better don't cover all faillure modes (possibly including a mode which will have the capacitor survive the relatively short and extreme climate tests, but not real life less extreme long climate tests that include mechanical stress effects).

[McMurdo]

Certainly fitted to some jap equipment. I had one pop on a late 70's Yamaha organ a few months ago while I was fault-finding in another area...it had been in a shed for 5 years.

[Heatercathodeshort]

The real shame is that thousands of good quality electrical appliances are taken to the dump with little more than a faulty mains filter capacitor. The heavy smoke frightens the user [and who can blame them?] resulting in a new purchase and yet more wicked waste. John.

[ianhumph]

I bought a circa 1966 valved Ultra suitcase record player from a car boot sale (the model number eludes me...) around 10 years ago. This was before the days of me using the Internet and a good while before I joined here. However, logic told me that it wasn't the wisest thing to just plug in straight in. I paid £4 for it, so if the worst happened then I wasn't going to have lost much and already had a later Fidelity HF 45 which I still use often.

The deck spun after a tiny bit of manual encouragement from me and it got up to near enough its correct speed. So, I got an old mono single and popped it on - all was well for a couple of minutes. The sound was not great but listenable. When a fair old amount of smoke started to come from it, I decided that the only thing to do was run over to the socket and switch it off. A friend who was present said it was probably 'dust on the valves' and I said that it thought it could be a capacitor. I wasn't aware of them exploding back then! I opened a window, let it 'cool down' for a while and switched it back on. The smoke lessened and stopped.

Days later, I tried the machine again and it had a little smoke once more - but when I used the player over the coming weeks it didn't do it again. I put it in storage and it got pinched, unfortunately!

[WaveyDipole]

They don't use jumbo versions of them in electricity sub-stations by any co-incidence do they?

http://www.bbc.co.uk/news/uk-england...shire-28084409

I did hear and see and electrolytic let rip inside a computer monitor once with a loud bang, a bright spark and a cloud of smoke shortly thereafter. Curiously the monitor carried on working although, of course, we did promptly turn it off. I don't think I have ever seen one of these RIFA types go though.