IEC Mains filter failures

[cprobertson1]

Good morning folks!

I've had an awesome weekend repairing lots of old equipment - and what struck me as odd was that several of them had blown IEC mains filters...

-Fluke PM6669B - an ebay special - IEC filter blew up on first use
-Racal-Dana 5002 - IEC filter already self destructed (and molten!)
-Hameg 5010 - also an ebay special - IEC filter gone to meet its maker (and spewed it's guts out over the side of the case for good measure)

My Fluke PM6669A also had a blown IEC filter when I first got it (what a bang THAT made!)

Now have I just had bad luck - or is this a common failure mode I was unaware of? Incidentally, every one of these is a schaffer brand filter; and every one of them is 25+ years old (which I suspect may have more to do with it than anything else)


Has anybody else found these little filter blocks to be a problem - or, again, was it just bad luck on my part that I happened to amass a collection of equipment with the same failure!

[VK5RC]

It is a very common failure, I think the real problem was the capacitors they used. If I recall correctly it was 1st described by an antique radio group. Schaffner does have a big proportion of the IEC filter market - I think the recent ones ie 10 yrs or younger are generally ok. A lot of guys change them along with the electrolytics in older good gear. Below is a photo of an HP 3245 I bought recently and its iec filter! Rather bizarrely still working with all normal values!

[dmills]

Totally well known, and the brand is no surprise... Replace on sight in old gear.

See also those class Y caps in the yellow semi transparent resin (Which I suspect are the root cause of the mains filter issue).

Regards, Dan.

[madires]

It's a known problem and it's caused by bad Rifa caps which disintegrate over time. When you replace a mains filter don't buy NOS which might have the same bad Rifa cap.

[IanMacdonald]

Even in the days of CRT TVs mains filter failure was a common problem. One of the more common fire hazards too.

The modern ones seem more reliable so if replacing make sure it is a modern one.

If in doubt, whip it out. Better a little RFI than joining the homeless squad down under the viaduct. 

[cprobertson1]

Even in the days of CRT TVs mains filter failure was a common problem. One of the more common fire hazards too.

The modern ones seem more reliable so if replacing make sure it is a modern one.

If in doubt, whip it out. Better a little RFI than joining the homeless squad down under the viaduct. 

Modern replacement - noted! Also, I love that saying xD

The fluke frequency counters both use the somewhat standard filter modules - but the Racal-Dana has a somewhat fancier assembly. It's actually quite well-made -  it has a voltage-selector drum with a number of cams on it, which when rotated makes and breaks contacts to match the voltage selection - quite a neat little system... molten plastic notwithstanding

Anyway - it's quite an interesting failure mode some of these seem to have - in two of them they've managed to get hot enough to melt the resin inside them - which has then been blown out (probably just pressure from boiling resin escaping) to judge by the spatters on the side of the case, and the molten resin that's flowed down the back panel.

Either way, the smell is exquisite! I'll cut one of them in half and let you folks have a look at some point! Nothing beats a good dissection

[FrankE]

32 quid! Ouch!

[dexters_lab]

Has anybody else found these little filter blocks to be a problem - or, again, was it just bad luck on my part that I happened to amass a collection of equipment with the same failure!

yes, they are notorious, especially the Shaffner ones!

stinkin' horrible things when they expire, had several go fizpop on me

[KJDS]

They are most likely to go when the get some moisture into them. If powering up something recently bought, especially if it's been sat in a damp auction house for weeks, it's worthwhile having it in a warm dry house for a few days first.

[babysitter]

They are most likely to go when the get some moisture into them. If powering up something recently bought, especially if it's been sat in a damp auction house for weeks, it's worthwhile having it in a warm dry house for a few days first.

My yet 2 devices that popped had been sitting heated and well used before the MSRE^*. Got me to start exchanging all of them.

BR
Hendrik

^* Magic smoke release event

[cprobertson1]

They are most likely to go when the get some moisture into them. If powering up something recently bought, especially if it's been sat in a damp auction house for weeks, it's worthwhile having it in a warm dry house for a few days first.

I thought they were all epoxied/potted up to prevent moisture ingress? Or does it start to become permeable over time?

This particular example was dead on arrival at any rate;) I still can't get the smell out my lab! I wonder if magic smoke inhalation is bad for your health...

[WA1ICI]

I've worked on lots of test equipment over the years, many with IEC filters, and have never come across blown capacitors in them, nor is this a "thing" among my technical associates.  Could this be because I live in the land of 120V power?  Less stress on the capacitors?

- John Atwood

[babysitter]

1. Almost every potting compound slows influx of humidity but securely keeps it inside
2. Yeah, 120V might be friendlier to the caps I suppose. I did a bit of googling when mine exploded, the problem gained traction around 1995. Time-fused smoke bomb.

[stj]

does anybody know if these schaffner filters are used in HP5 laser printers?
i have one i suspect could be randomly tripping an RCD.

[FrankE]

I can't recall seeing one in my HPLJ5.
Have a look at the youtube channel 'Brokenprinter'

[mmagin]

What I find especially fascinating is that the capacitors in these line (mains) filters are paper capacitors, decades after paper capacitors were known to develop leakage in the signal path of old radios and audio amplifiers (detrimentally affecting the DC biasing).

[CNe7532294]

They are most likely to go when the get some moisture into them. If powering up something recently bought, especially if it's been sat in a damp auction house for weeks, it's worthwhile having it in a warm dry house for a few days first.

I had a debate about the exact mechanism of failure in my own threads. I can attest that moisture definitely had something in someway to do with filter cap failure.

Long story short, we enlisted every available demodulating receiver (ie. spectrum analyzers and radios) and modulating transmitter (ie. signal and function gens) for use during Hurricane Harvey. We left the more handy radios and transceiver at another location. Now the garage wasn't flooded thank goodness but there was definitely high humidity during and after the storm. My place of stay literally became an island (basically Houston turned into Venice) and I had no boat/watertaxi. Turned on to test a spectrum analyzer we thought had been inspected. Date coded around '88-89. It worked and it had worked several times before during drier days. Left it plugged in and zzzzzt about 20mins later. I "instinctively gunned for" pulling the plug off the wall socket.

Moral of the story, replace any old filter caps you find at all times. Rifa X and Y caps in my case.

https://www.eevblog.com/forum/chat/old-rifa-capacitors-and-a-disaster-story/msg1302802/#msg1302802