|
Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets. |
|
Thread Tools |
8th Dec 2018, 2:30 pm | #21 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Would you trust these caps?
This is exactly what they’re designed to do. They are sacrificial. They protect the device and eat up some of the noise going into it.
Healing here means it’s still a capacitor and not a short or open after a transient. This it’s still able to perform its duty. Capacitance of course will decline because every flashover means the plate size is reduced permanently. So when you design with them you usually over specify the capacitance by 3-5x. Ceramics go short then explode. Much worse outcome. Stuff with these things tends to have a service life of around ten years and that's plenty to eat up all the transients over time. We only hate the things usually because we own the stuff after a decade. I’ve learned to respect the sacrifice and just replace them. |
8th Dec 2018, 3:07 pm | #22 |
Dekatron
Join Date: Aug 2007
Location: W.Butterwick, near Doncaster UK.
Posts: 8,935
|
Re: Would you trust these caps?
Yes I remember those blue ceramics in a certain Phillips chassis tv that used to burn up very well!
__________________
G8JET BVWS Archivist and Member V.M.A.R.S |
8th Dec 2018, 3:28 pm | #23 | ||
Dekatron
Join Date: Dec 2007
Location: Haarlem, Netherlands
Posts: 4,203
|
Re: Would you trust these caps?
Quote:
|
||
8th Dec 2018, 3:32 pm | #24 | |
Dekatron
Join Date: Dec 2007
Location: Haarlem, Netherlands
Posts: 4,203
|
Re: Would you trust these caps?
Quote:
Also, there's lots of equipment we fully expect to be around after 10 years, especially if it sees little usage. Measuring equipment and sewing machines are two categories where conservative or unknowing designers often specified Rifa that are especially prone to see premature paper capacitor failure. Last edited by Maarten; 8th Dec 2018 at 3:41 pm. |
|
8th Dec 2018, 4:38 pm | #25 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Would you trust these caps?
The issue with the RIFAs failing is the encapsulation material degrades and lets moisture in. The capacitor has a paper dielectric and this gets damp, turns into a resistor, dissipates power and goes kawoosh. The new Kemet ones have a different encapsulation epoxy. The bit inside the capacitor is top notch. The bit that kept the inside In wasn’t
Measuring equipment gets inspected regularly unless you’re one of those horrible companies that can’t be bothered and sticks “for indication only” on everything and let’s it rot for two decades. Sewing machines for sure. My mother’s food mixer blew up a few months back with a RIFA going off. Took 30 years. That’s not bad for anything going across the line. |
8th Dec 2018, 6:48 pm | #26 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Would you trust these caps?
That’s interesting research, Hugo. Useful to know what I vaguely suspected is actually true.
And yes, I did wonder by what mysterious mechanism caps could be ‘self-healing’. It sounds like a marketing ploy.
__________________
Al |
8th Dec 2018, 7:03 pm | #27 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Would you trust these caps?
As I was just literally replacing one this afternoon here’s an old vs new RIFA comparison for reference:
|
8th Dec 2018, 8:37 pm | #28 |
Dekatron
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,195
|
Re: Would you trust these caps?
Hi, from what is written above it looks like RIFA's are considered OK now.
Has anyone any idea when they were improved? Ed |
8th Dec 2018, 8:39 pm | #29 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Would you trust these caps?
They redesigned them when everything went lead free. Not sure when that was precisely.
|
8th Dec 2018, 8:57 pm | #30 | |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 14,007
|
Re: Would you trust these caps?
Quote:
A bad reputation is easy to earn but takes a long time to live down. |
|
8th Dec 2018, 9:48 pm | #31 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Would you trust these caps?
Quote:
So if there is a component added to a pcb with some life-cycle maybe 10 years, it is merely another example of built in obsolescence. I regard these capacitors in a similar light to say a soldered in pcb battery, or perhaps an internal battery backed up non volatile ram, or perhaps an electrolytic capacitor too. They are used by manufacturers knowing full well they will set a limit to the life of the device or appliance and that device will definitely be in land fill before the decade is out, albeit some of it recycled. For those wanting to save our planet from pollution and E-waste, it is much better if our appliances last longer, as they once did when there was less life-cycling and superior quality components and built with the service person in mind and with support documentation. Everyone seems to forget about the entropy increases of manufacturing new goods from scratch and the recycling of old ones. We are encouraged to throw away the old and buy the new. One reason the X2's are popular though is their small size for the capacity and voltage ratings. As with any electronic part, it will last longer if it is physically bigger. Imagine a transformer the size of a house only being asked to deliver 50W. Probably (if protected from weather) it would still be working 1000 years later. The same for massive resistors, capacitors or practically any part you can think of. Clearly a 2kV rated cap will be over double the physical size of a standard X2. So maybe we just need a simple equation for the capacitor: Size/Lifetime = constant. This also applies to electrolytic caps where it has been observed by many that the physically larger ones have a much longer life. I have particularly noticed this in vintage computers with analog power supplies where the very physically large and large uF value (>10,000 uF) caps are as good as new over 40 years later. And of course the worst electrolytics with the shorter life are tiny surface mount types. The only way to improve X2's is to use thicker insulation, in other words increase their voltage ratings and they would last longer and be physically bigger. Last edited by Argus25; 8th Dec 2018 at 9:56 pm. |
|
9th Dec 2018, 12:23 am | #32 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Would you trust these caps?
Quote:
In the middle of the night a Rifa capacitor cross the foot pedal controller system (inside the foot pedal) spontaneously shorted out, probably after a voltage spike. The machine started on full speed. I woke up wondering what the heck was going on and it really freaked me out. After I investigated it, of course it all made sense in the light of day, but half asleep in the middle of the night, when a machine does that, it was quite disconcerting. |
|
9th Dec 2018, 1:32 am | #33 |
Moderator
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,967
|
Re: Would you trust these caps?
This isn't at all unusual. Singer rheostat pedals from the 30s were fitted with two TCC interference suppression caps. My 1934 Singer 201K has this arrangement, and one day when I forgot to switch it off at the mains I found it running at full speed. I changed the original caps for X2s and all was well.
|
9th Dec 2018, 1:14 pm | #34 | |
Dekatron
Join Date: Dec 2007
Location: Haarlem, Netherlands
Posts: 4,203
|
Re: Would you trust these caps?
Quote:
P.S. 30 years is indeed possible for such a capacitor, and if you draw up a graph of the average lifespan it may well be that most will live those 10 years you mentioned as a recommendation. That doesn't really address my problems with using them, but I can see why it would be tempting to do so if there would be any other small advantage or percieved advantage, even if only because they look so pretty (my first impression when I saw a few of them in a Swedish Philips set). Last edited by Maarten; 9th Dec 2018 at 1:25 pm. |
|
9th Dec 2018, 2:00 pm | #35 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Would you trust these caps?
This is all good and that but I've seen the Epcos MKT X2 catch fire in an HP server power supply as well. And that was under three years old with huge upstream transient filters.
This is high energy stuff. It's going to have a working life. Stay in the curve and you're fine. |
9th Dec 2018, 3:58 pm | #36 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Would you trust these caps?
I had a similar experience with a RIFA in the pedal of a sewing machine which went with a loud bang followed by a horrible smell... The machine didn't start to run at full speed though as some have described. It had been purchased in the 1970's and I had 'rescued' it when I cleared out my dad's house. Although the problem could easily had been resolved with a replacement cap, the experience left such an impression on the other half that she insisted on not having the thing in the house again. It wasn't that reliable when new, so I guess it was just as well to let bygones be bygones.
It seems that I am not the only one to be unconvinced about the "self-healing" terminology. As per the preceding comments, I had always assumed that a transient or spike that exceeds the design tolerances is likely to damage the foil surfaces/dielectric to some extent and I also could not imagine any mechanism by which this damage might self repair, so I always took the expression to be marketing hype. I hadn't quite thought of it as "progressive death" either but I suppose that progressive deterioration is by nature a feature of any capacitor, just that some deteriorate quicker than others depending on materials used and the application in which they serve. Last edited by WaveyDipole; 9th Dec 2018 at 4:16 pm. |
9th Dec 2018, 4:00 pm | #37 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Would you trust these caps?
An analogy. Your body is self healing. To a point.
|
9th Dec 2018, 4:21 pm | #38 |
Dekatron
Join Date: Nov 2003
Location: Heckmondwike, West Yorkshire, UK.
Posts: 9,642
|
Re: Would you trust these caps?
Capacitors were mentioned, and as night follows day the topic has reverted to RIFAs!
These capacitors have been fitted by a wide range of manufacturers, both domestic and professional, for many years and seemingly still are. Their designers must have reasons for using them and the manufacturers must be aware of relevant statistics. They can't all be wrong surely? |
10th Dec 2018, 12:29 pm | #39 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Would you trust these caps?
Sorry, my fault! I started the original discussion about unknown HV ceramics purchased from eBay that were the subject of my concern, but then, as an aside, asked whether such HV caps would be suitable to replace RIFA Y-class caps. I received clear advice on that (they are not suitable) for which I am grateful, but it seems that my question triggered a diversion to the general subject of RIFA's.
Regarding the original unknown brand ceramics, a member has offered to soak test them at up to 2kV and I have sent him a couple of samples of each of the 471s and 472s. I will report back on the result. |
12th Dec 2018, 2:13 am | #40 | |
Dekatron
Join Date: Dec 2007
Location: Haarlem, Netherlands
Posts: 4,203
|
Re: Would you trust these caps?
Quote:
Ceramic capacitors weren't really suited as a full range alternative to paper capacitors back then either. I think ceramic Y-capacitors were only introduced in the early 1970's and pulse rated ceramic capacitors such as about which this topic was started probably even in the 1980's. @AC/HL: After roughly 1980, they all were wrong, but didn't necesarilly know or care. After all, faillure within warrantee would have been reasonably low. Last edited by Maarten; 12th Dec 2018 at 2:20 am. |
|