Which class of capacitor on this switch

I've had to take my Marantz 2226b front panel apart to clean some switches and as its difficult to get the board out, I'm refreshing what I can.

There's a paper in oil cap? 0.022uf accross the switch contacts going live to live and a 0.022uf poly caps going neutral to neutral.

Which class of capacitor would be the one to use.
 

Attachments

  • Screenshot_20250627_193206_Adobe Acrobat.jpg
    Screenshot_20250627_193206_Adobe Acrobat.jpg
    136 KB · Views: 79
These are not a "critical safety part". What they do is suppress the transient over the switch. If they fail (short) the switch is closed unintentionally, that is all. Sometimes there is too much recapping because why not, pavlov reactions and urban legends.
 
These are not a "critical safety part". What they do is suppress the transient over the switch. If they fail (short) the switch is closed unintentionally, that is all. Sometimes there is too much recapping because why not, pavlov reactions and urban legends.

They are a critical safety part. They are constantly subject to the mains voltage when the unit is off, but plugged in. They are renowned for randomly exploding, particularly PIO and Rifa style potted epoxy types. They crack, oxidize, become a highish resistance, heat up, catch fire or explode- all while the owner has no idea, unless they are lucky enough to hear the bang or see/smell the smoke.

Suppression cap failures are responsible in my opinion for a huge amount of perfectly otherwise fine gear going into landfill. Plenty of kitchen appliances, hifi, power tools, etc.

Any switch suppression caps on vintage HiFi gear should be replaced as a matter of course particularly ones like the 22xx, 50yo Marantz receivers..
 
I always use a safety cap with a voltage rating of at least twice your mains voltage
As tempting as that is the caps must be X1 or X2 rated. 1000vdc/630 vac rated caps might for example seem a safe bet but they absolutely are not.

An X1 or X2 cap will have an AC rating of typically 275vac which might seem to offer no safety margin but they do. They are designed to 'self heal' in the event of a breakdown in the dielectric due for example to a transient spike. Do not use anything else.
 
  • Like
Reactions: ejp
They are renowned for randomly exploding, particularly PIO and Rifa style potted epoxy types. They crack, oxidize, become a highish resistance, heat up, catch fire or explode- all while the owner has no idea, unless they are lucky enough to hear the bang or see/smell the smoke.
Well I assume you have references to these kinds of events. I am 55 years involved in electronics and electrical equipment. Both design and repair, both consumer and professional. But I never have witnessed anything else than a silent failure in which case the capacitor either opened or shorted. And that even a handful of times.

Now I happen to live in a climate which is extremely harsh on everything and especially on electronics. Due to high temperature, high humidity, dust, salt and UV. Everything oxidizes, breaks, deteriorates, crumbles, falls apart, transforms to chewing gum or evaporates. Lifetime of any electrical equipment used to be a third of that in moderate climates, with today's "quality" it is 4-5 times faster if it even survives initial switch-on. Still I have never seen such capacitors fail any more violently than die silently. Not even in the 1970-ies on equipment which was already ancient at that time. But I might be wrong.

Besides, in this particular example of Marantz the two capacitors are in series and each have half the mains voltage across.
 
A dielectric breakdown in a cap which is wired across the switch terminals cannot result in any more fault current than the unit normally draws when turned ON. The cap itself might be destroyed, but won’t result in uncontrolled current flow. Higher voltage rating is always better, and the value SHOULD be set such that the LCR circuit never results in crazy voltages. It will be less than what is developed across the switch when opened without the cap, for sure.
 
Well I assume you have references to these kinds of events.

https://www.transistorforum.nl/forum/index.php?mode=thread&id=20759

Old RIFA paper capacitors are notorious even though they are type-approved X capacitors. Apparently they met all the requirements when new, but after many years, they can develop cracks, moisture can ingress into the paper and then you are in trouble. They call them RIFA-rookbommen (RIFA smoke bombs) at the Dutch forum I just linked to.
 
Well I assume you have references to these kinds of events. I am 55 years involved in electronics and electrical equipment. Both design and repair, both consumer and professional. But I never have witnessed anything else than a silent failure in which case the capacitor either opened or shorted. And that even a handful of times.
I've seen it twice in IBM PC and AT power supplies, the capacitors are right across each power lead and ground at the power connector (not quite the same as being across the unit's power switch, but still). They emit lots of smoke and hiss, but don't pull the 15+ amperes required to throw the circuit breaker. Here's a quick Youtube search that brings up some surely exciting videos:
https://www.youtube.com/results?search_query=eevblog+rifa
 
Last edited:
Isn’t RIFA part of Philips? Nuff said.

As far as those non-X caps from line to ground throwing a fit while not drawing enough to throw a breaker - that’s one of the reasons behind AFCI’s. There may NOT be enough average/RMS current to throw a breaker, but the peaks in an arc fault can be pretty damn high.
 
https://www.transistorforum.nl/forum/index.php?mode=thread&id=20759

Old RIFA paper capacitors are notorious even though they are type-approved X capacitors. Apparently they met all the requirements when new, but after many years, they can develop cracks, moisture can ingress into the paper and then you are in trouble. They call them RIFA-rookbommen (RIFA smoke bombs) at the Dutch forum I just linked to.
I once replaced the capacitor in parallel with the motor in my Dual turntable. It had cracked and burned just like the one in the photo.
Ed
 
Type this into google and look at the images:

rifa capacitor fail

I have replaced many hundreds and cleaned up the mess from hundreds more. They fail due to the epoxy cracking (it shrinks) (whether they are used or NOS in a drawer), letting the air in and the foil oxidizes creating a leakage current which heats up the capacitor, shrinking the epoxy even more and the pattern continues until they fail. They were most often installed as arc-suppresion caps and subject to full mains voltage all the time the device was plugged in until it was turned on.

If you haven't seen a belt drive turntable from the 1970s/80s that doesn't turn off, or the motor pulley strangely kicks when the arm is returned, or a tuner where the power switch does nothing or a Bamix immersion blender with black smoke marks around the rotary speed switch, or a Kenwood chef stand mixer that went BANG just sitting on the kitchen bench, etc, then you haven't been in electronics or repair long enough. All epoxy Rifa style failures.

Often Rifas (or old PIOs) were also tied to chassis from neutral- we call them Death Caps, especialy for poor Americans with their rotatable two pin old-skool power plugs. Plenty of gear brought to Australia in the 70s and 80s was rewired here by the distributors and much of it two core unearthed. There's a 50-50 chance they ballsed it up and when they did, often Rifas to chassis (as well as the switch suppressors) go bad and make a real mess.

They should be replaced on sight. Most are 50+ years old and once you start taking the covers off your vintage HiFi and actually looking for the little bastards hanging off every power switch in appliances or even old power tools, you will find these time-bombs everywhere. Look at them- see the cracks and the oxidized aluminium inside. Cut them off and replace them now.
 
Last edited:
Isn’t RIFA part of Philips? Nuff said.

As far as those non-X caps from line to ground throwing a fit while not drawing enough to throw a breaker - that’s one of the reasons behind AFCI’s. There may NOT be enough average/RMS current to throw a breaker, but the peaks in an arc fault can be pretty damn high.
These ARE "X caps":
Andy Pugh says:
April 2, 2023 at 4:04 am
I think that there is a classification error here, in that Rifa make (or made: https://en.wikipedia.org/wiki/RIFA_AB ) many different styles of capacitors and translucent prismatic capacitors are made by many other manufacturers.
The problem, it seems, from literally seconds of Googling, is with the RIFA X2 type capacitors.
Report comment
Reply
  1. Sam says:
    April 2, 2023 at 1:23 pm
    In the retro computing world they’re just referred to as RIFA caps for ease, basically anyone will know what is meant by them because they’re absolutely notorious.
https://hackaday.com/2023/04/01/why-do-rifa-capacitors-fail/