Dave, doesn't that first switch just change the rise time between 5ns and 10ns? You seemed to be expecting it to change the pulses between 100ns and 10ns, but I'm sure you mentioned rise time in the previous video and the "Tr" notation seems to allude to that. Good score by the way. I've been looking to get one of these for a while, but very few show up in Australia, and shipping from overseas is often far too expensive.
Ah, good catch, yes, Tr would equate to rise time. Pays to engage brain before talking. Yes, would cost a fortune to ship from overseas, it's seriously heavy, and any suitably padded box would be huge. Cubic weight calculation would be huge.
@@EEVblog That's normal, your mind becomes less sharp when you smoke rifa's a lot. Also you might become paranoid if you smoke too much of these, so keep an eye out for that symptom.
Some background: RIFA is Swedish and stands for RadioIndustriernas FabriksAktiebolag, which translates into The Radio Industry's Manufacturing Incorporated. This was a company launched by the Swedish government during WWII, because it feared there might be a blockade and a shortage on electrical components. The government wanted everyone to have a working radio so they could receive messages in case of an invasion and beyond. RIFA's initial charter was essentially to manufacture as many components (mainly caps) as quickly as possible. Apparently, the initial runs had some serious quality issues, and a saying in Sweden at that time was "If your radio stops working, just replace the RIFA caps". Anyway, RIFA became pretty successful over the years and did pretty well. After a while, they started producing ICs and other components as well and eventually got bought by Ericsson to produce mainly ICs for their AXE telephony switches. I worked there in the 80s, right in the transition between RIFA and Ericsson. It was my first job. A summer job that kind of transitioned into a "real" job as a programmer for their test engineering department. It was a great place to work. I don't think there's much left of it anymore. The caps production is definitely long gone.
Evox bought the capacitor business and later Kemet bought Evox-Rifa. They still make these crappypotted paper capacitors, and I think they kept the production faciity in Kalmar for quite some time. I didn't even know for sure they closed it down. Unfortunately they apparently did't think that was a good moment to scrap the old production lines...
@@mjouwbuis Seems like this type of caps are produced in Finland nowadays (Evox). There's one factory in Gränna, Sweden producing electrolytic caps though (Rifa).
I don't know about Kalmar, but Kemet still produces automotive capacitors in Gränna, in one of RIFA's old factories! As far as I know, it's the last passive component factory left in Sweden, so I'm super happy that they're keeping that alive.
This is a common problem with vintage computers. Especially in the power supplies. Just get an old machine, turn it on, and boom! Magic smoke! First thing you gotta do with vintage equipment before you turn it on the first time, is check the caps. You see these paper-wound X/Y caps, you don't question it, just replace them right away.
Dave, replace the mains input IEC connector as well! It also contains the same type Rifa paper capacitors, according to the comment overlaid on this video. Have a look at that comment now, I asked him, and he confirmed. He tossed out a whole NOS box of these IEC connectors, because they fail the same way!
I agree. The Schaffner IEC connector is almost certain to fail in the same spectacular fashion. I haven't opened one up but I suspect they have same RIFA caps inside and exhibit the same failure mode. In my experience the RIFA caps drop to quite a low insulation resistance before failure so an insulation resistance test from line to earth can be useful for finding dodgy Y caps in an IEC module. I treat anything with an insulation resistance of less than 10 megohms as an imminent failure and replace the filter module. However, finding dodgy X caps with an insulation resistance test from line to neutral is less successful since the filter modules commonly have a in-built discharge of a few megohms which masks the leakage of a failing X cap. It is safest to just immediately replace any Schaffner IEC connectors of that vintage.
In the past couple of years Yamaha have done a product recall on 30 year old amplifiers which were fitted with RIFA smoke bombs. On the same audio forum someone had posted pictures of a RIFA capacitor that had exploded despite having never been fitted into a board. No smoke but the thing had completely split apart. Stories from me about RIFA capacitors. 1). Hope your vacuum cleaner doesn't have one fitted. With all the air going through you won't see smoke but you will smell it everywhere. 2). If equipment fails check for these. My Mother threw away all the accessories for her expensive sewing machine and asked me to take the machine to the tip thinking the motor had burnt out. Out of curiosity I looked inside and it was a RIFA capacitor next to the motor had released all its smoke. 3). I'm watching this after having a second Philips Oscilloscope release all the magic smoke from one of these.
Hahaha these are such a classic thing to have go bad!I happen to live in the city where they were manufactured and know one of the guys who developed those caps. A fun little side note about them is if you leave them powered on they will basically never go bad, because they were designed with some amount of inefficiency, causing them to run ever so slightly warm and dry themselves out.
Nice to hear from someone who actually knows an insider. It's indeed known that under light working conditions, keeping paper capacitors powered up is good for their survival. Did you ever ask this guy why they kept on making those, seemingly to the exact same design, for decades on end even long after the design was outdated and probably known by the quality control department as needing improvement or discontinuation? Is it just because customers kept buying them?
You DO need all the chokes and caps to make sure that the pulse you generate on the output doesn't contaminate the rest of your instrumentation. All of those devices ARE really necessary. You really don't want to test ALL your lab equipment, only the one device on the DUT side of things. Fascinating about the caps though.
@Pamperchu They are very often failing to short. Just a momentary short, because when they blow up, they eventually go open (usually, but that's like Russian roulette). I've seen dozens on mains fuses blown up by them!
@@poptartmcjelly7054 They are "flame proof". Of course that doesn't mean they can't emit copious amounts of smoke while depositing nicotine like gunge all over adjacent components. I saw it for a fourth time this week while testing an old Philips Oscilloscope.
Yeah, but these aren't "safety capacitors" as they aren't Y class, they're only X thus are designed to fail short - the circuit designer needs to take this into account... most didn't. I've never had a RIFA Y fail short.
I should have disputed Dave's assumption it was the transformer / choke. I used to rewind motors & transformers... and lacquering and baking them. There's no way the lacquer could re-liquify and drip upon being reheated. Yes, in stripping motors we used to heat up the stator by powering it without the rotor... and that would make the windings easier to pry out after air-chiselling the tops off. The lacquer would become sticky... but never liquidy. Maybe after 6 hours at 400 degrees... But my electronics knowledge is very limited. However in this very narrow example my motor/transformer winding knowledge is more than Dave's!!! Yay!!
Bill R the copper was still bright around the turns, I observed the same thing but trusted this Aussie ‘s big, experienced-nose to sort out the culprit by smell on site, tho evidently mistaken were we both, or he planted a trap for us new players at home for the sequel.
Same story here. A RIFA cap released its magic smoke in my 30-year old Studer A810 tape recorder. A big puff of white smoke and a strong odor of burned newspaper 😯
I had 2 of these RIFA caps fail in an old sewing machine. (The one in the machine went first, after 15 min of use, and after another half hour, the one in the pedal had enough also) Not a pleasant smell at all!
And with some of these old x and y caps if the fuse doesn't cut quickly enough, they release this sort of weird tar like vapor........which forms a sticky coating on adjacent parts and smells like burning phenolic.....vigorously....oh boy......ethanol/spirits remove it just fine, though. And i think it wasnt just RIFA.....WIMA made...and STILL makes caps that look virtually identical. Me thinks those also blow. Let's hope the new ones don't cause the same shit in 20-30 years -.-
I serviced some Luxman stereo equipment from the late 70s that had similar Rifa caps. None had failed, but it was clear the cases were deteriorating like you demonstrated. Needless to say, I replaced them with some new poly x-types along with all of the original electrolytics, and I'm glad I did!
In europe they might be underrated by now also since we have raised the grid voltage. In my sewing machine there were 220v caps and our grid is now 250. I dont know how much that would matter but the new ones are properly rated anyway.
@@Eddies_Bra-att-ha-grejer Det är nominellt. kan vara så låg som 210 vid hög belastning och 260 270 andra gånger. Sen kan det ju spika när man statrar såna grejer som just det kondingarna sitter i tex symaskiner då kan det lokalt vara väldigt mycket högre än den nominella nätspänningen.
I was going through some old (new) parts I have and I found a largish Rifa cap, never used, but cracked just like the rest. I always thought they swelled up and cracked after many years of service - Not so, actual usage does not seem to matter, just the time from manufacture to now. Years ago, I worked in a Telstra repair shop (TSG Collingwood) and we had trouble with GEC Terminet printers catching fire, it was eventually tracked down to these Rifa caps in the power supply. These printers could easily be left to run for days monitoring things, unattended, so if one of these caps smoked up there would be no one there to switch it off before the flames started after all the initial smoke had been let out.
Yeah that PSU should have had a fusible resistor in series, a common penny pinching design omission I've found in several PSUs over the years. Modern filter capacitors just lose their capacitance over time and may as well be open circuit after a few years. The RIFA design was supposed to be "self healing" however, the epoxy doesn't have flexibility nor the same thermal expansion characteristics of the metallized paper roll of the capacitor and cracks form, these allow water ingress and then it's game over, it's akin to wrapping aluminium foil over toilet paper, once it gets wet, it turns into a good conductor and boom... 😱
@@davidhunt240 What is the brown nicotine like gunge that comes out of them ? The ones in the video looked quite clean compared to one I have just had fail.
Those hadn't failed under load. Typically the X capacitor fails and goes short circuit blowing the main fuse. This prevents the other capacitors from failing in quick succession. The gunk that comes from the capacitor is the glue, paper and moisture (often melted aluminium blobs too) and those things also get hot enough to burn and create the characteristic smoke. In larger PSUs > 1kW there's often a fusible resistor in series with banks of these damn things and they are like Chinese firecracker ropes, if there was a ghost in the machine, it's gone and done a runner now 😝
Yep, always blame the Rifa, I had one go bang inside a Hoover Junior vacuum, the stink was immense and the stink stuck on the vac for months even after cleaning the mess out. Where there's a Rifa, there'll be trouble... Biggest problem with them is they don't bother encasing the cap in a plastic box, like everyone else does, so the resin is exposed and breaks down faster, and a contemporary cap would still work fine when a Rifa goes bang, because Rifa just got it completely wrong when cheaping out on materials, not that they care, they got their money, and these days someone else is getting money too for the replacements...
twocvbloke moisture in compounds remains a modern problem, even with hermetic and conformal coating. Shell up what you may, when things short and go bang the case becomes a potential projectile no matter the construction. Place them in locations where they won’t blind a peeping tech like so and that’s all the safety there is.
The lack of enclosure is only half of the problem. The other half is that they insisted on using a paper dielectric instead of polypropylene like most other series and brands. It's VERY hard to make a reliable paper cap, especially if you use anything less than a glass or a soldered metal can with oil filling. Unfortunately RIFA is still making money from this crap, since some people think that using original parts is the best way to repair equipment.
and Philips scopes, BBC Micro's, Husqvarna Sewing Machines, Electrolux vacuum cleaners (that one was smelly 😞). I've had failures in all of those ! I don't own a Yamaha amp but they have done a product recall an 30 year old equipment because of these.
Few months ago I put a comment because I repaired a ye olde Swiss made Meteor winder from mid 70's. Electromechanical motor controller was virtually blown out by those Rifa caps. Now I had another popcornish Rifa caps madness. The victim was an auxiliary SMPS assembled in Mexico inside a mid 90's QUAD IVc MK2 pick and place machine "proudly made in the USA" (cuac!). In this ocassion those common mode filter components don't caused massive destruction, but in our cold humid Buenos Aires winter conditions, became leaky and the nearby Laser aligment controller gone crazy. So, the moral is: If you see an old Rifa cap, just change it! By the way. Dave, I liked so much your differntial probe! If my coutry's economy goes better (and my own), I may consider to get a pair of those. My lab colleagues have a devotion for frying oscilloscope probes as if they were gifted them!! :-(((
Just yesterday I powered up a Grundig V2000 format video recorder for the first time in about 10 years. Oh yes, lots of magic smoke. The smell is similar to that from the old phosphorus matches. Replaced capacitors with modern high quality X type ones of the correct value, even though, as you say, the unit will actually function without them. They are branded WIMA.
Haha I have an old HP LCR meter that I’ve used for years. One day I heard a sizzle and a sudden BANG followed by smoke. But the unit kept working. Yep - good old RIFA cap on the X cap.
Used as filters in singer sewing machine pedals made in clydebank Scotland. Got my mums down from loft after 20 years and erupted in smoke. Replaced and its fine now.
Might be a good hint to remind people to measure those old safety caps before using them in a new circuit, quite often they slowly fail open and the capacitance rating can be nearly zero after a "one product lifetime" of use.
I've had this happen with an old amber monitor. Back then I didn't know what it was and threw it out. I kind of regret that because it still worked, even after that smoke escaped. So I think it also was a filter capacitor.
Thanks for your sharing. I just found out an old Haag Streit slit lamp power supply also using RIFA filter caps (both X2 and Y2). The X2 220n and Y2 1n0 RIFA filters caps were blown out. Its time to change them all.
I had some RIFA PME271M caps fail in the power brick for a 1988 Compaq SLT/286 "lunchbox" laptop. Along with all the silver coloured Sprague caps. Went up with a sizzling "puff" and tons of smoke. Smelled like burning wood. All replaced and running great now :D
Is it legal to smoke them?! They made us watch that movie to discourage pot smoking among us high schoolers. As if that would work in Northern California! Haha!
Had 2 long unused Apple II power suppliess fail because of RIFAs, One, my spare, right after the first. Both a few minutes after power up. Fortunately I was around to hear the zap-crack before too much smoke escaped. Easily replaced.
Bloody RIFAs. I've witnessed several of these go off, but none as dramatic as inside a Grundig VCR from the 90s. It dented the can containing the PSU and even lifted and catapulted the fuse right out of its socket! The fuse was surprisingly still ok, but the whole PSU compartment was littered with shrapnel. By now I've learned to preemptively replace them; it takes minutes compared to the hours you'll spend cleaning up the mess, particularly the sticky flameproof resin!
I find only the transparent case versions will go bang, the ones potted in the thicker coloured housings are a lot more reliable, they only fail as they age from self healing down to the picofarad range. The transparent ones do not have an outer water resistant housing, and the bare epoxy degrades as it is exposed to air, the cased ones the fill at the bottom is generally thick enough that they just slowly degrade. IIRC X class are allowed to pass current as they self heal, but Y class are required to be double layer construction, so that there are 3 foils inside, one floating and acting as an intermediate, so that even if the one self heals there is no current flow through the terminals, just the single part blows out and is isolated, but no DC current flow.
RIFA caps are the most common part to pop in the ACORN BBC MICRO and MASTER computers and probably every other computer from the 80s that has a built in power supply, they look exactly the same as those and same result if not replaced after a circa 20 year dormant storage
My first ever magic smoke incident as a young'un was with an old Mac power supply board. I saved it as a memento. Looking at it right now... yep, the cap that burst looks just like the ones Dave took out!
Yep, I can verify that. I have 2 BBC Micros and a Master 128, in each of which I have replaced the RIFA caps. Only 1 of them had actually failed, with the others showing signs of imminent failure. The amount of smoke and stench these caps release is almost off the chart!
Last year, while my friend was guiding a tour in computer museum, one of Master Compact psu went kaboom. I though that I've replaced these caps in all units. Oops :-)
@13:25 Maybe someone half pulled the IC to make it appear broken in order to get a new unit, I have come across that kind of thing before on broken gear I have purchased, where it was obviously made to appear broken, ( which made it easy to fix). I quite like it when people do that to get new lab gear, as it seems I can then be lucky enough to buy the "broken" one on eBay.
The Defpom's Repair Channel More likely that the IC was pulled during diagnosis or by a smart tech to prevent theft or usage before a repair was effected, in wait for a purpose which never came. U are abominable.
Equipping my second workbench at home, I had three Tektronix Oscilloscopes that I had previously on store spreading fumes in a row just in the span of one or two days!
Good information on the Rifa capacitors. They were commonly used in older high-end equipment, but I'm not sure if they or their successors are used anymore. Anyway - I'd say that you now have a decent tool to test any serious build you make.
Neat video, I have removed all kinds of the newer plastic caps from stuff, threw in a bin, never knew there was a difference, to me they all seemed the same.
Third video in the series: Dave hooks up RIFA caps to the mains outside and entertains the crowd with clouds of smoke. Afterwards on EEVBlog2, Dave and Mrs EEVBlog show how to get burn marks off the patio pavers.
BTW, to remove glares from non-metallic surfaces on video get a Circular Polarizer filter in front of the lens, the only problem is that you might also loose the picture of LCD displays, but it depends on the angle.
Same thing happened to me with an old Apple ][... Powered it up after being off for 20+yrs, heard a huge BANG! Thought the thing was completely blown, turns out it was those RIFA caps...
It’s a shame you didn’t end the video by demonstrating the effectiveness of a healthy X class cap across the output of the device under test. That would have been cool!
Dave, you really need to look at Mr.Carlson's Lab. His basic approach would have flagged the problem very quickly. I suggest you become a Patreon and build some of his test equipment so that you can better understand and explain to your audience the common modes of failure and how to identify them.
Hi Dave, would you please make a video demonstrating and explaining what these main caps and chokes really do? Many of us specially beginners can't simulate the situation where they are needed. Most of the time we neglect the fact that there might be glitches in the mains and that our devices affect it.
You looked at the pulse fall time and there was no difference. You needed to zoom in more to see the difference is rise time as you where on 500ns per division.
I suspect that the scope pulses are a result of the capacitance and inductance etc of your 'load'. The unit outputs an specific width energy pulse. You would need a resistive load to see the actual pulse profile.
I make my own cap filter with x2 cap type and smd resistor on pcb board😊 I must replace a lot of this caps on WEKO drivers in offset printing machines. They always blows up or crack and some parts don’t works
Damn RIFAs again :D i see alot of these and yeah they dont always fail spectactularly but about _EVERY_ one i have seen so far has at least started cracks
Yes Dave they are supposed to crack and vent. The idea is to take out surges and power spikes, they generally can only ever handle a few surges before they start to vent. The plastic blue potted ones tend to go with a enormous bang as they are sealed however the Rifa type crack and vent without exploding violently. They are a sacrificial capacitor, they take the hit and protect the rest of the circuit.
Could you explain the importance of X capacitors? I have plug in surge protectors that self-consume 29-43 mA, and I suspect it is the X capacitors leaking current! Was wondering what effect if I remove them?
@@mrtechie6810 The keep high frequency/ sharp transient interference in and out of the device. Which might cause malfunction in logic circuits, noise in radio devices etc.
A couple of observations and queries: 1. The new cap you show is potted inside a solid plastic shell, and it looks more robust than the old ones. Do modern film caps actually last longer? 2.) I noticed that the 'scope traces of short and longer pulses show a secondary 'peak' followed by a slight bulge on the pulse's trailing edge. I have a hypothesis that explains that, maybe the apparent insensitivity in the pulsewidth settings. I wonder if in reality a train of 'ringing' oscillations followed the main pulse. Some kind of filtering in the measurement instrumentation would partially remove the oscillations, yet leave an 'envelope' signal that you observed. So, while the true pulse width varies with the set time, the ringing may not, giving you the apparently constant pulse width on the scope. Also, are you certain that your 'scope has enough bandwidth to track oscillations as fast as the ringing would be?
You should have called out the people in your company who thought it was not only a good idea to keep making paper capacitors while the competition had already switched to polyester and polypropylene, but to actually make them worse than they needed to be even by 1960's standards. By the 1980's, polypropylene capacitors had gotten reliable enough to replace the last paper capacitors in safety and pulse applications, so the very least that RIFA engineers could have done, was make a _superior_ paper capacitor if they wanted to stick with paper for sentimental or marketing reasons. Instead they kept on producing this old, proven unreliable design. Bringing that up, unfortunately, isn't hate but being flabbergasted. I can't think of a single reason why a company would continue doing this for decades (Evox continued, and now Kemet still sells them). To be fair, Wima did the exact same thing, but they were slightly (a few percent) more durable and they did make and enthousiastically market some of the best polypropylene capacitors.
Would your high voltage probe be good for hooking a tube amp (400v) up to my scope? I don't trust my little probes to not let some nasty voltage through to my front ends.
I have Phillips PM2811. One day, EMI filter Shurter exploded. Problem is i can't find any adequate replacement. There were a RIFA caps, just like on video.
Did they never test those caps in some sort of environment/ageing chamber? Or did they, and just went "Pah, 15years lifetime? We will have flying fusion powered cars until then" ? I can't image those caps didn't show some of the issues we see with ALL of them today...
I've read that they indeed didn't pass mechanical tests (not even aging tests) at some equipment manufacturers, but obviously not every design department had such strict quality control.
Could you explain the importance of X capacitors? I have plug in surge protectors that self-consume 29-43 mA, and I suspect it is the X capacitors leaking current! Was wondering what effect if I remove them?
Because the X capacitor is wired across the mains, it’s expected to draw reactive power of incoming mains voltage divided by its reactive impedance. I’ll guess that it’s about 1uF based upon those currents.
The PME labeled caps were used extensively in older Tektronix switching power supplies. You must shot gun replace these because they are a ticking time disaster for old gear.
I'm guessing that this is what happened to my old Hameg oscilloscope. I powered it on and it was fine for about 5 minutes, and then a loud BANG and a ton of smoke, but the scope still stayed on. Gave me a heart attack lol.
@@MarkTillotson except that Rifa-Evox-Kemet still make exactly the same capacitors, including the crappy housing that causes them to fail even faster than any paper capacitor.
I had a beautiful, new old stock HP 5328B that went pop and a small amount of magic smoke came out. Looked inside for an exploded electrolytic, found none, and set it aside after some head-scratching. I just wonder if it's suffering from RIFA madness....?
You can know the 'health' of the capacitors (discharged) measuring the C-Rp or C-Rs parameters with an RLC meter (at 50 Hz or 60 Hz). Rp modelates leaks (between layers). When the voltage 110 V or 230 V (worse case because double current pass) is applied, there is a power dissipation in form of heat. As the time pass, the temperature goes up, and if the heat is high, the magic smoke scapes from the capacitor. You can see the delta tangent from the datasheet, and from this value and the capacitor value, isolate the Rp. "X" capacitors could fail in short circuit mode, and if the current flowing the capacitor is high enough, the well dimmensioned fuse of the equipement will breaks. At lower current, magic smoke goes out. "Y" capacitors could fail but in open circuit mode, otherwhise they could short the L(ine) to the earth of the equipment. In this case, is the differential breaker who breaks the line. A bad earth of the instalation could be dangerous the the user. If there is no a linear effect with voltage (measuring C-Rp at , you can 1 V p-p) you can use your differential probes to measure the current and the phase using the capacitor in series with a known resistor. It is a three element network: C-Rp + Rs_test. You can measure it at different voltages to learn about the linear effect with voltage.
Could you explain the importance of X capacitors? I have plug in surge protectors that self-consume 29-43 mA, and I suspect it is the X capacitors leaking current! Was wondering what effect if I remove them?
@@mrtechie6810 X capacitors are for regulatory: EMC, mainly to reduce the conducted noise from the equipment to the mains. If your equipment consumes e.g. 30 mA and is a reactive current only (is 90° out of phase of the voltage), it does not consume active power and then the series resistance is small and the capacitor does not heat up. If the voltage is 230 V, the frequency 50 Hz; I = 2 * pi * f * C * Vac; C = 30 mA / (2 * pi * 50 Hz * 230 V) = 415 nF, that is usual in mains filters. If you remove them, the EMC compliance could be compromised. If the capacitors heat up, change them for a new ones.
@@user-xlario the SPDs I opened have 0.68uF, 0.47uF, 0.33uF X2 caps. I measured 29-43mA using the clamp meter UNI-T UT210E. The higher capacitance had higher leakage current. I did not check with an oscilliscope. My concern is not heating but phantom power waste, with multiple of these SPDs around the house, always plugged in, for many years!
@@mrtechie6810 An ideal rective component: capacitor or inductor, does not consum active power (W), but reactive power (VA). This has no effect on the power consumption (does not cost money). You only pay for the active power (W).
@@user-xlario so what is the correct way to measure this? How would a wattmeter read? I had a wattmeter ProGrade 512 (now broken) and as I recall it measured around 5-10 watts per SPD. So I think you are saying only commercial or smart metered customers are billed for reactive power. And to understand this, requires an oscilloscope or a fancier power meter. Is that correct? On the other hand, if running off a UPS/solar storage battery, an additional 70 watts of reactive power would use/drain the battery!
My tek2445 smoked some time ago leaving a nasty smell in the air. One of those RIFA capacitors had blow up. Upon replacing, spotted a few more that I had to replace also, before it all happen again. A photo: www.flickr.com/photos/jcbarros71/40894499174/in/dateposted-public/
My Zanussi washer dryer had one of these but instead of cracking right away the plastic expanded with the capacitor inside it until it finally cracked and started turning off my rcd all of the time
Dave, doesn't that first switch just change the rise time between 5ns and 10ns?
You seemed to be expecting it to change the pulses between 100ns and 10ns, but I'm sure you mentioned rise time in the previous video and the "Tr" notation seems to allude to that.
Good score by the way. I've been looking to get one of these for a while, but very few show up in Australia, and shipping from overseas is often far too expensive.
Ah, good catch, yes, Tr would equate to rise time. Pays to engage brain before talking.
Yes, would cost a fortune to ship from overseas, it's seriously heavy, and any suitably padded box would be huge. Cubic weight calculation would be huge.
@@EEVblog That's normal, your mind becomes less sharp when you smoke rifa's a lot. Also you might become paranoid if you smoke too much of these, so keep an eye out for that symptom.
@@Tommyinoz1971 ha! Pissa.
@@6c45pi Yeah, I'm Kean... I mean keen. If you google me you should find my email or other contact details pretty easily
@@Tommyinoz1971 - In Canada it's now legal to smoke Rifas.
They smell horrible though...
- Eddy
Some background: RIFA is Swedish and stands for RadioIndustriernas FabriksAktiebolag, which translates into The Radio Industry's Manufacturing Incorporated. This was a company launched by the Swedish government during WWII, because it feared there might be a blockade and a shortage on electrical components. The government wanted everyone to have a working radio so they could receive messages in case of an invasion and beyond. RIFA's initial charter was essentially to manufacture as many components (mainly caps) as quickly as possible. Apparently, the initial runs had some serious quality issues, and a saying in Sweden at that time was "If your radio stops working, just replace the RIFA caps". Anyway, RIFA became pretty successful over the years and did pretty well. After a while, they started producing ICs and other components as well and eventually got bought by Ericsson to produce mainly ICs for their AXE telephony switches. I worked there in the 80s, right in the transition between RIFA and Ericsson. It was my first job. A summer job that kind of transitioned into a "real" job as a programmer for their test engineering department. It was a great place to work. I don't think there's much left of it anymore. The caps production is definitely long gone.
Evox bought the capacitor business and later Kemet bought Evox-Rifa. They still make these crappypotted paper capacitors, and I think they kept the production faciity in Kalmar for quite some time. I didn't even know for sure they closed it down. Unfortunately they apparently did't think that was a good moment to scrap the old production lines...
@@mjouwbuis Seems like this type of caps are produced in Finland nowadays (Evox). There's one factory in Gränna, Sweden producing electrolytic caps though (Rifa).
RIFA caps are notorious for blowing out in certain vintage keyboards.
I don't know about Kalmar, but Kemet still produces automotive capacitors in Gränna, in one of RIFA's old factories! As far as I know, it's the last passive component factory left in Sweden, so I'm super happy that they're keeping that alive.
I just bought a RIFA paper filter cap last week, so they are still in market.
In German vintage audio forums, these RIFA caps are known as "Knallfrosch", which translates to firecracker.
Lol das is a geile Bezeichnung.
In Dutch: Rifarotjes (also firecrackers with some alliteration thrown in) or rifarookbommen (smoke bombs, also with the alliteration)
Not surprising to me at all when I go bang sure sounds like a firecracker.
RIFA PME capacitors is pretty much the electronic version of Surströmming.
Both come from Sweden, and both explode and stink.
😂
I've smelled surströmming. I don't want that to ever happen again!
Shame you can't open these in a bucket
This is a common problem with vintage computers. Especially in the power supplies. Just get an old machine, turn it on, and boom! Magic smoke! First thing you gotta do with vintage equipment before you turn it on the first time, is check the caps. You see these paper-wound X/Y caps, you don't question it, just replace them right away.
Dave, replace the mains input IEC connector as well! It also contains the same type Rifa paper capacitors, according to the comment overlaid on this video. Have a look at that comment now, I asked him, and he confirmed. He tossed out a whole NOS box of these IEC connectors, because they fail the same way!
I agree. The Schaffner IEC connector is almost certain to fail in the same spectacular fashion. I haven't opened one up but I suspect they have same RIFA caps inside and exhibit the same failure mode.
In my experience the RIFA caps drop to quite a low insulation resistance before failure so an insulation resistance test from line to earth can be useful for finding dodgy Y caps in an IEC module. I treat anything with an insulation resistance of less than 10 megohms as an imminent failure and replace the filter module.
However, finding dodgy X caps with an insulation resistance test from line to neutral is less successful since the filter modules commonly have a in-built discharge of a few megohms which masks the leakage of a failing X cap. It is safest to just immediately replace any Schaffner IEC connectors of that vintage.
This has happened to me twice in a year, just now on the IEC im choking on the sickening smoke
In the past couple of years Yamaha have done a product recall on 30 year old amplifiers which were fitted with RIFA smoke bombs. On the same audio forum someone had posted pictures of a RIFA capacitor that had exploded despite having never been fitted into a board. No smoke but the thing had completely split apart.
Stories from me about RIFA capacitors.
1). Hope your vacuum cleaner doesn't have one fitted. With all the air going through you won't see smoke but you will smell it everywhere.
2). If equipment fails check for these. My Mother threw away all the accessories for her expensive sewing machine and asked me to take the machine to the tip thinking the motor had burnt out. Out of curiosity I looked inside and it was a RIFA capacitor next to the motor had released all its smoke.
3). I'm watching this after having a second Philips Oscilloscope release all the magic smoke from one of these.
Hahaha these are such a classic thing to have go bad!I happen to live in the city where they were manufactured and know one of the guys who developed those caps. A fun little side note about them is if you leave them powered on they will basically never go bad, because they were designed with some amount of inefficiency, causing them to run ever so slightly warm and dry themselves out.
Nice to hear from someone who actually knows an insider. It's indeed known that under light working conditions, keeping paper capacitors powered up is good for their survival. Did you ever ask this guy why they kept on making those, seemingly to the exact same design, for decades on end even long after the design was outdated and probably known by the quality control department as needing improvement or discontinuation? Is it just because customers kept buying them?
You DO need all the chokes and caps to make sure that the pulse you generate on the output doesn't contaminate the rest of your instrumentation. All of those devices ARE really necessary. You really don't want to test ALL your lab equipment, only the one device on the DUT side of things. Fascinating about the caps though.
Yes, I think the 1kV pulse was only about 750V because some of it was pushing back into your lab mains!
"safety capacitor" and "notorious for failing" don't go together too well.
@Pamperchu by bursting into flames?
@Pamperchu They are very often failing to short. Just a momentary short, because when they blow up, they eventually go open (usually, but that's like Russian roulette). I've seen dozens on mains fuses blown up by them!
Safety co-pastor failure first safety second it will work fine as a safety capacitor in till it's failed unsafely
@@poptartmcjelly7054 They are "flame proof". Of course that doesn't mean they can't emit copious amounts of smoke while depositing nicotine like gunge all over adjacent components. I saw it for a fourth time this week while testing an old Philips Oscilloscope.
Yeah, but these aren't "safety capacitors" as they aren't Y class, they're only X thus are designed to fail short - the circuit designer needs to take this into account... most didn't. I've never had a RIFA Y fail short.
I should have disputed Dave's assumption it was the transformer / choke. I used to rewind motors & transformers... and lacquering and baking them. There's no way the lacquer could re-liquify and drip upon being reheated. Yes, in stripping motors we used to heat up the stator by powering it without the rotor... and that would make the windings easier to pry out after air-chiselling the tops off. The lacquer would become sticky... but never liquidy. Maybe after 6 hours at 400 degrees... But my electronics knowledge is very limited. However in this very narrow example my motor/transformer winding knowledge is more than Dave's!!! Yay!!
Bill R the copper was still bright around the turns, I observed the same thing but trusted this Aussie ‘s big, experienced-nose to sort out the culprit by smell on site, tho evidently mistaken were we both, or he planted a trap for us new players at home for the sequel.
Same story here. A RIFA cap released its magic smoke in my 30-year old Studer A810 tape recorder. A big puff of white smoke and a strong odor of burned newspaper 😯
I had 2 of these RIFA caps fail in an old sewing machine.
(The one in the machine went first, after 15 min of use, and after another half hour, the one in the pedal had enough also)
Not a pleasant smell at all!
And with some of these old x and y caps if the fuse doesn't cut quickly enough, they release this sort of weird tar like vapor........which forms a sticky coating on adjacent parts and smells like burning phenolic.....vigorously....oh boy......ethanol/spirits remove it just fine, though. And i think it wasnt just RIFA.....WIMA made...and STILL makes caps that look virtually identical. Me thinks those also blow. Let's hope the new ones don't cause the same shit in 20-30 years -.-
I serviced some Luxman stereo equipment from the late 70s that had similar Rifa caps. None had failed, but it was clear the cases were deteriorating like you demonstrated. Needless to say, I replaced them with some new poly x-types along with all of the original electrolytics, and I'm glad I did!
mishmash, Smooth move... charcoal brickets don't fetch much on eBay!
@@BruceNitroxpro, no Ebay here! Those amps are making sweet music in my living room.
In europe they might be underrated by now also since we have raised the grid voltage. In my sewing machine there were 220v caps and our grid is now 250. I dont know how much that would matter but the new ones are properly rated anyway.
Det är väl 230v i Sverige, inte 250? Högsta som finns är väl 240v?
@@Eddies_Bra-att-ha-grejer Det är nominellt. kan vara så låg som 210 vid hög belastning och 260 270 andra gånger. Sen kan det ju spika när man statrar såna grejer som just det kondingarna sitter i tex symaskiner då kan det lokalt vara väldigt mycket högre än den nominella nätspänningen.
Merci beaucoup pour cette vidéo. Je vais changer tous les Rifa de mon TEKTRONIX des années 80. Merci encore et Bonjour de France.
I was going through some old (new) parts I have and I found a largish Rifa cap, never used, but cracked just like the rest. I always thought they swelled up and cracked after many years of service - Not so, actual usage does not seem to matter, just the time from manufacture to now.
Years ago, I worked in a Telstra repair shop (TSG Collingwood) and we had trouble with GEC Terminet printers catching fire, it was eventually tracked down to these Rifa caps in the power supply.
These printers could easily be left to run for days monitoring things, unattended, so if one of these caps smoked up there would be no one there to switch it off before the flames started after all the initial smoke had been let out.
Yeah that PSU should have had a fusible resistor in series, a common penny pinching design omission I've found in several PSUs over the years. Modern filter capacitors just lose their capacitance over time and may as well be open circuit after a few years. The RIFA design was supposed to be "self healing" however, the epoxy doesn't have flexibility nor the same thermal expansion characteristics of the metallized paper roll of the capacitor and cracks form, these allow water ingress and then it's game over, it's akin to wrapping aluminium foil over toilet paper, once it gets wet, it turns into a good conductor and boom... 😱
@@davidhunt240 What is the brown nicotine like gunge that comes out of them ? The ones in the video looked quite clean compared to one I have just had fail.
Those hadn't failed under load. Typically the X capacitor fails and goes short circuit blowing the main fuse. This prevents the other capacitors from failing in quick succession. The gunk that comes from the capacitor is the glue, paper and moisture (often melted aluminium blobs too) and those things also get hot enough to burn and create the characteristic smoke. In larger PSUs > 1kW there's often a fusible resistor in series with banks of these damn things and they are like Chinese firecracker ropes, if there was a ghost in the machine, it's gone and done a runner now 😝
Yep, always blame the Rifa, I had one go bang inside a Hoover Junior vacuum, the stink was immense and the stink stuck on the vac for months even after cleaning the mess out. Where there's a Rifa, there'll be trouble...
Biggest problem with them is they don't bother encasing the cap in a plastic box, like everyone else does, so the resin is exposed and breaks down faster, and a contemporary cap would still work fine when a Rifa goes bang, because Rifa just got it completely wrong when cheaping out on materials, not that they care, they got their money, and these days someone else is getting money too for the replacements...
twocvbloke moisture in compounds remains a modern problem, even with hermetic and conformal coating. Shell up what you may, when things short and go bang the case becomes a potential projectile no matter the construction. Place them in locations where they won’t blind a peeping tech like so and that’s all the safety there is.
@@HighestRank, LOVE the "peeping tech" visual... BANG! OMG, did they say "Wear safety goggles?"
The lack of enclosure is only half of the problem. The other half is that they insisted on using a paper dielectric instead of polypropylene like most other series and brands. It's VERY hard to make a reliable paper cap, especially if you use anything less than a glass or a soldered metal can with oil filling. Unfortunately RIFA is still making money from this crap, since some people think that using original parts is the best way to repair equipment.
Speaking from experience you wouldn't have been any better off with an Electrolux. The fastest way of making a whole room smell of RIFA!
yep i found those caps in mine the hard way to.. BANG !! was the first thing they said
I got bit in the butt with those caps. WARNING/NOTE... The older Tektronix scopes have these caps in there. One blew up on me in my 2213.
and Philips scopes, BBC Micro's, Husqvarna Sewing Machines, Electrolux vacuum cleaners (that one was smelly 😞). I've had failures in all of those ! I don't own a Yamaha amp but they have done a product recall an 30 year old equipment because of these.
Yeah, I've had a Husqvana sewing machine and an older Kenwood food mixer release the magic smoke.
I see these, I immediately take them out and replace them without even checking if they are good or not. I had so many of them explode in the past.
Apple II owners are VERY familiar with these. Love the title!
thanks for the info about x and y class caps, my mechanical understanding of the world was advanced a tad
Roger Beck *has advanced a bit.*
Roger that, but be sure to subscribe to Mr Carlson's lab to level up even more ;)
@@benbaselet2026, I SWEAR Paul is in his 90's! How ELSE could one person amass such knowledge! LOL de KQ2E
@@HighestRank It's A "tad "
When I solder in hotel rooms i find the shower cap provides a nice cover for the smoke detector 😎
Few months ago I put a comment because I repaired a ye olde Swiss made Meteor winder from mid 70's. Electromechanical motor controller was virtually blown out by those Rifa caps. Now I had another popcornish Rifa caps madness. The victim was an auxiliary SMPS assembled in Mexico inside a mid 90's QUAD IVc MK2 pick and place machine "proudly made in the USA" (cuac!). In this ocassion those common mode filter components don't caused massive destruction, but in our cold humid Buenos Aires winter conditions, became leaky and the nearby Laser aligment controller gone crazy.
So, the moral is: If you see an old Rifa cap, just change it!
By the way. Dave, I liked so much your differntial probe! If my coutry's economy goes better (and my own), I may consider to get a pair of those. My lab colleagues have a devotion for frying oscilloscope probes as if they were gifted them!! :-(((
I just smoked a RIFA. The whole circuit went up in magic smoke.
Just yesterday I powered up a Grundig V2000 format video recorder for the first time in about 10 years. Oh yes, lots of magic smoke. The smell is similar to that from the old phosphorus matches. Replaced capacitors with modern high quality X type ones of the correct value, even though, as you say, the unit will actually function without them. They are branded WIMA.
2:59 That broken cap at the lower left looks like a painting. Pretty colours.
Cheers!
Haha I have an old HP LCR meter that I’ve used for years. One day I heard a sizzle and a sudden BANG followed by smoke. But the unit kept working. Yep - good old RIFA cap on the X cap.
Wow! I have a few of these in my store of parts. I will take them out and mark as do not use, for museum use only. Thanks Dave!
Used as filters in singer sewing machine pedals made in clydebank Scotland. Got my mums down from loft after 20 years and erupted in smoke. Replaced and its fine now.
RIFA caps and VARTA memory backup cells.... Match made in hell.
@Dick Fageroni sometimes the same thing.
@Dick Fageroni RIFA isn't German, it's Swedish.
Nice one. A very sailient point on capacitors and closer inspection. Cheers Dave.
Might be a good hint to remind people to measure those old safety caps before using them in a new circuit, quite often they slowly fail open and the capacitance rating can be nearly zero after a "one product lifetime" of use.
and look closely for cracks. That happens even on New Old Stock.
Good to see that you did a follow up on your train coductor device :)
(that's what Schaffner roughly translates to)...
I've had this happen with an old amber monitor. Back then I didn't know what it was and threw it out. I kind of regret that because it still worked, even after that smoke escaped. So I think it also was a filter capacitor.
Thanks for your sharing. I just found out an old Haag Streit slit lamp power supply also using RIFA filter caps (both X2 and Y2). The X2 220n and Y2 1n0 RIFA filters caps were blown out. Its time to change them all.
Nice cinematographic effect at 4:16 :)
Good on you for looking into this further. All very interesting. Many thanks for the great videos.
I had some RIFA PME271M caps fail in the power brick for a 1988 Compaq SLT/286 "lunchbox" laptop. Along with all the silver coloured Sprague caps. Went up with a sizzling "puff" and tons of smoke. Smelled like burning wood.
All replaced and running great now :D
@10:00 that 820n 500V MKP is certainly not a Y class cap but being across the bridge rectifier it's probably the pre-PFC smoothing cap.
RIFA MADNESS!!
100% folio rate!
Is it legal to smoke them?! They made us watch that movie to discourage pot smoking among us high schoolers. As if that would work in Northern California! Haha!
i had 2 in my Tek 2710 SA. but to reach them I had to remove even the crt and back pannel. That was a lot of work.
Had 2 long unused Apple II power suppliess fail because of RIFAs,
One, my spare, right after the first. Both a few minutes after power up.
Fortunately I was around to hear the zap-crack before too much smoke escaped.
Easily replaced.
We were already at the point where we called it "common mode choke", now we are back to "transformer" again. :D
Future destructive testing of the cheap china gear? You'll probably need a fume hood for the magic smoke though :D
A classic reefer capacitor, I see what you did there with the madness :)
Hi, I repair domestic appliances in the UK and I replace these on a regular basis.
Even ones that aren't cracked will let the smoke out eventually.
Electrolux vacuum cleaners had them. you won't see the smoke but you will certainly smell it as the capacitor sits in the airflow.
Bloody RIFAs. I've witnessed several of these go off, but none as dramatic as inside a Grundig VCR from the 90s. It dented the can containing the PSU and even lifted and catapulted the fuse right out of its socket! The fuse was surprisingly still ok, but the whole PSU compartment was littered with shrapnel.
By now I've learned to preemptively replace them; it takes minutes compared to the hours you'll spend cleaning up the mess, particularly the sticky flameproof resin!
Ooh caps... looks like I was right after all. :-)
So was I, but as Dave I was thrown by the red herring.
I find only the transparent case versions will go bang, the ones potted in the thicker coloured housings are a lot more reliable, they only fail as they age from self healing down to the picofarad range. The transparent ones do not have an outer water resistant housing, and the bare epoxy degrades as it is exposed to air, the cased ones the fill at the bottom is generally thick enough that they just slowly degrade.
IIRC X class are allowed to pass current as they self heal, but Y class are required to be double layer construction, so that there are 3 foils inside, one floating and acting as an intermediate, so that even if the one self heals there is no current flow through the terminals, just the single part blows out and is isolated, but no DC current flow.
@@SeanBZA, I wondered what the difference was... thanks for that lucid update!
RIFA caps are the most common part to pop in the ACORN BBC MICRO and MASTER computers and probably every other computer from the 80s that has a built in power supply, they look exactly the same as those and same result if not replaced after a circa 20 year dormant storage
Luckily I shake up me old-eBones every dozen years or so.
My first ever magic smoke incident as a young'un was with an old Mac power supply board. I saved it as a memento. Looking at it right now... yep, the cap that burst looks just like the ones Dave took out!
Yep, I can verify that. I have 2 BBC Micros and a Master 128, in each of which I have replaced the RIFA caps.
Only 1 of them had actually failed, with the others showing signs of imminent failure.
The amount of smoke and stench these caps release is almost off the chart!
Last year, while my friend was guiding a tour in computer museum, one of Master Compact psu went kaboom.
I though that I've replaced these caps in all units. Oops :-)
@@0x9066 The "auto-finding" function was enabled... it gives both an auditory and visual signal. LOL
@13:25 Maybe someone half pulled the IC to make it appear broken in order to get a new unit, I have come across that kind of thing before on broken gear I have purchased, where it was obviously made to appear broken, ( which made it easy to fix). I quite like it when people do that to get new lab gear, as it seems I can then be lucky enough to buy the "broken" one on eBay.
The Defpom's Repair Channel
More likely that the IC was pulled during diagnosis or by a smart tech to prevent theft or usage before a repair was effected, in wait for a purpose which never came. U are abominable.
Great idea Defpom. I need a new scope at work. Does that make me abominable?
It isn't my idea, it is what I have found a few times when I have purchased broken gear off eBay, so it is obviously happening somewhere.
You crack me up sometimes Dave, keep up the good work
Equipping my second workbench at home, I had three Tektronix Oscilloscopes that I had previously on store spreading fumes in a row just in the span of one or two days!
Airgead Holy pow, Badman!
Good information on the Rifa capacitors. They were commonly used in older high-end equipment, but I'm not sure if they or their successors are used anymore.
Anyway - I'd say that you now have a decent tool to test any serious build you make.
Neat video, I have removed all kinds of the newer plastic caps from stuff, threw in a bin, never knew there was a difference, to me they all seemed the same.
Third video in the series: Dave hooks up RIFA caps to the mains outside and entertains the crowd with clouds of smoke. Afterwards on EEVBlog2, Dave and Mrs EEVBlog show how to get burn marks off the patio pavers.
BTW, to remove glares from non-metallic surfaces on video get a Circular Polarizer filter in front of the lens, the only problem is that you might also loose the picture of LCD displays, but it depends on the angle.
Same thing happened to me with an old Apple ][... Powered it up after being off for 20+yrs, heard a huge BANG! Thought the thing was completely blown, turns out it was those RIFA caps...
It’s a shame you didn’t end the video by demonstrating the effectiveness of a healthy X class cap across the output of the device under test. That would have been cool!
Dave, you really need to look at Mr.Carlson's Lab. His basic approach would have flagged the problem very quickly.
I suggest you become a Patreon and build some of his test equipment so that you can better understand and explain to your audience the common modes of failure and how to identify them.
Cheek...
Hi Dave, would you please make a video demonstrating and explaining what these main caps and chokes really do? Many of us specially beginners can't simulate the situation where they are needed. Most of the time we neglect the fact that there might be glitches in the mains and that our devices affect it.
I once replaced a bunch of those RIFA capacitors in analog dimmer packs of an old theatre, those things really stink when blown!
i can remember seeing lots of those caps and ones with resistors in parralel too the telephone exchange had lots of them back in the day
I'm replacing a few of these in a TRS-80 Model III Power supply. The magic smoke hasn't come out yet, but they look just like Dave's.
Audiophiles would buy that RIFA "vintage" for a hundred dollars, just named it, vintage RIFA, RARE ITEM!
Thanks for heads up. I have a welding machine that doesn't work full of those caps. I didn't think much about them. I need to check them. Thanks.
Apparently machine failed a while after it was switched on. First it had intermittent operation and then completelly stopped maintaining arc.
I always choose wima caps, they're HiQ and i like the red color.
I've encountered Soooo many exploded / crusty caps like this on late 60's to 80's electronics
Thankfully the repair is usually quite straight forward
You looked at the pulse fall time and there was no difference. You needed to zoom in more to see the difference is rise time as you where on 500ns per division.
Yeah, but I won't see a difference with that probe bandwidth anyway.
@@EEVblog, Do you have any "exotic" probes which would?
That reminds me, 70s Grundig TVs used them. Fixed one up from the dump ( dry joint I think) in the 80s went kaboom few months later this explains it.
I suspect that the scope pulses are a result of the capacitance and inductance etc of your 'load'. The unit outputs an specific width energy pulse. You would need a resistive load to see the actual pulse profile.
I make my own cap filter with x2 cap type and smd resistor on pcb board😊 I must replace a lot of this caps on WEKO drivers in offset printing machines. They always blows up or crack and some parts don’t works
Took a while to get to 500K subs, but it appears your momentum is picking up, as well as views. It didn't take long to go from 500K to 627K. :-)
Yeah I had to replace one of those in my Yamaha DX7. The synth started smoking ....very pungent smell indeed.
I find these all the time in old Numatic vacuum cleaners.
Damn RIFAs again :D i see alot of these and yeah they dont always fail spectactularly but about _EVERY_ one i have seen so far has at least started cracks
That probe looks so nice!
Great play on words.. keep up the good work and cheers
Yes Dave they are supposed to crack and vent. The idea is to take out surges and power spikes, they generally can only ever handle a few surges before they start to vent. The plastic blue potted ones tend to go with a enormous bang as they are sealed however the Rifa type crack and vent without exploding violently. They are a sacrificial capacitor, they take the hit and protect the rest of the circuit.
Could you explain the importance of X capacitors? I have plug in surge protectors that self-consume 29-43 mA, and I suspect it is the X capacitors leaking current! Was wondering what effect if I remove them?
@@mrtechie6810
The keep high frequency/ sharp transient interference in and out of the device. Which might cause malfunction in logic circuits, noise in radio devices etc.
Hello Dave, you are doing a good job, thanks
A couple of observations and queries:
1. The new cap you show is potted inside a solid plastic shell, and it looks more robust than the old ones. Do modern film caps actually last longer?
2.) I noticed that the 'scope traces of short and longer pulses show a secondary 'peak' followed by a slight bulge on the pulse's trailing edge. I have a hypothesis that explains that, maybe the apparent insensitivity in the pulsewidth settings. I wonder if in reality a train of 'ringing' oscillations followed the main pulse. Some kind of filtering in the measurement instrumentation would partially remove the oscillations, yet leave an 'envelope' signal that you observed. So, while the true pulse width varies with the set time, the ringing may not, giving you the apparently constant pulse width on the scope. Also, are you certain that your 'scope has enough bandwidth to track oscillations as fast as the ringing would be?
Don't hate on RIFA! I used to work there in the 80s when they were still making capacitors!
You should have called out the people in your company who thought it was not only a good idea to keep making paper capacitors while the competition had already switched to polyester and polypropylene, but to actually make them worse than they needed to be even by 1960's standards. By the 1980's, polypropylene capacitors had gotten reliable enough to replace the last paper capacitors in safety and pulse applications, so the very least that RIFA engineers could have done, was make a _superior_ paper capacitor if they wanted to stick with paper for sentimental or marketing reasons. Instead they kept on producing this old, proven unreliable design. Bringing that up, unfortunately, isn't hate but being flabbergasted. I can't think of a single reason why a company would continue doing this for decades (Evox continued, and now Kemet still sells them). To be fair, Wima did the exact same thing, but they were slightly (a few percent) more durable and they did make and enthousiastically market some of the best polypropylene capacitors.
Would your high voltage probe be good for hooking a tube amp (400v) up to my scope? I don't trust my little probes to not let some nasty voltage through to my front ends.
I have Phillips PM2811. One day, EMI filter Shurter exploded. Problem is i can't find any adequate replacement. There were a RIFA caps, just like on video.
Did they never test those caps in some sort of environment/ageing chamber? Or did they, and just went "Pah, 15years lifetime? We will have flying fusion powered cars until then" ? I can't image those caps didn't show some of the issues we see with ALL of them today...
I've read that they indeed didn't pass mechanical tests (not even aging tests) at some equipment manufacturers, but obviously not every design department had such strict quality control.
Where do you find rare schematics like that? or even schematics in general?
Back in the '60s and ' 70s we had the same with WIMA brand capacitors.
I'm glad you took my comment into consideration. Now if i didn't make that stupid typo :D
Yep, my vintage Quad 33 pre-amp has a RIFA across the mains switch - cracked and smells awful on power up. Noisy at switch on/off so must be replaced.
Could you explain the importance of X capacitors? I have plug in surge protectors that self-consume 29-43 mA, and I suspect it is the X capacitors leaking current! Was wondering what effect if I remove them?
Because the X capacitor is wired across the mains, it’s expected to draw reactive power of incoming mains voltage divided by its reactive impedance. I’ll guess that it’s about 1uF based upon those currents.
Good video, the old RIFA caps of that style are known to fail often now.
funkyironman69, butter put her in dah ovens at 400F° for several hours before trying it 0n.
Not only now, they have been known to fail since a few years after market introduction.
The PME labeled caps were used extensively in older Tektronix switching power supplies. You must shot gun replace these because they are a ticking time disaster for old gear.
I'm guessing that this is what happened to my old Hameg oscilloscope. I powered it on and it was fine for about 5 minutes, and then a loud BANG and a ton of smoke, but the scope still stayed on. Gave me a heart attack lol.
Believe it or not, these RIFA caps are still available!
I'd argue it's not "because" those are Rifa, but those metallized paper mains-caps are indeed notorious for failing.
Khron's Cave •Rifa infused with 1/3% fail, a full ° more of fail than mere 33.3% Rifas!
All paper caps are timebombs. Paper absorbs moisture and becomes brittle and acidic over time. Noone makes them any more for good reason.
@@MarkTillotson except that Rifa-Evox-Kemet still make exactly the same capacitors, including the crappy housing that causes them to fail even faster than any paper capacitor.
I had a beautiful, new old stock HP 5328B that went pop and a small amount of magic smoke came out. Looked inside for an exploded electrolytic, found none, and set it aside after some head-scratching. I just wonder if it's suffering from RIFA madness....?
If it stinks yes.
You can know the 'health' of the capacitors (discharged) measuring the C-Rp or C-Rs parameters with an RLC meter (at 50 Hz or 60 Hz). Rp modelates leaks (between layers). When the voltage 110 V or 230 V (worse case because double current pass) is applied, there is a power dissipation in form of heat. As the time pass, the temperature goes up, and if the heat is high, the magic smoke scapes from the capacitor. You can see the delta tangent from the datasheet, and from this value and the capacitor value, isolate the Rp.
"X" capacitors could fail in short circuit mode, and if the current flowing the capacitor is high enough, the well dimmensioned fuse of the equipement will breaks. At lower current, magic smoke goes out.
"Y" capacitors could fail but in open circuit mode, otherwhise they could short the L(ine) to the earth of the equipment. In this case, is the differential breaker who breaks the line. A bad earth of the instalation could be dangerous the the user.
If there is no a linear effect with voltage (measuring C-Rp at , you can 1 V p-p) you can use your differential probes to measure the current and the phase using the capacitor in series with a known resistor. It is a three element network: C-Rp + Rs_test. You can measure it at different voltages to learn about the linear effect with voltage.
Could you explain the importance of X capacitors? I have plug in surge protectors that self-consume 29-43 mA, and I suspect it is the X capacitors leaking current! Was wondering what effect if I remove them?
@@mrtechie6810 X capacitors are for regulatory: EMC, mainly to reduce the conducted noise from the equipment to the mains. If your equipment consumes e.g. 30 mA and is a reactive current only (is 90° out of phase of the voltage), it does not consume active power and then the series resistance is small and the capacitor does not heat up.
If the voltage is 230 V, the frequency 50 Hz; I = 2 * pi * f * C * Vac; C = 30 mA / (2 * pi * 50 Hz * 230 V) = 415 nF, that is usual in mains filters.
If you remove them, the EMC compliance could be compromised.
If the capacitors heat up, change them for a new ones.
@@user-xlario the SPDs I opened have 0.68uF, 0.47uF, 0.33uF X2 caps. I measured 29-43mA using the clamp meter UNI-T UT210E. The higher capacitance had higher leakage current. I did not check with an oscilliscope.
My concern is not heating but phantom power waste, with multiple of these SPDs around the house, always plugged in, for many years!
@@mrtechie6810 An ideal rective component: capacitor or inductor, does not consum active power (W), but reactive power (VA). This has no effect on the power consumption (does not cost money). You only pay for the active power (W).
@@user-xlario so what is the correct way to measure this? How would a wattmeter read? I had a wattmeter ProGrade 512 (now broken) and as I recall it measured around 5-10 watts per SPD. So I think you are saying only commercial or smart metered customers are billed for reactive power.
And to understand this, requires an oscilloscope or a fancier power meter.
Is that correct?
On the other hand, if running off a UPS/solar storage battery, an additional 70 watts of reactive power would use/drain the battery!
My tek2445 smoked some time ago leaving a nasty smell in the air. One of those RIFA capacitors had blow up. Upon replacing, spotted a few more that I had to replace also, before it all happen again. A photo: www.flickr.com/photos/jcbarros71/40894499174/in/dateposted-public/
My Zanussi washer dryer had one of these but instead of cracking right away the plastic expanded with the capacitor inside it until it finally cracked and started turning off my rcd all of the time