If they have a tighter tolerance I think its not snake oil. I just replaced some regular electrolytic 100uf caps in a crossover with a 5% tolerance metalized polypropylene caps. All the imbalance I heard between the speakers disappeared. I always measured no imbalance with a dB meter but it always sounded imbalanced to me. I suspect it was the tolerance and the electrolytic caps were opposite of each other on the tolerance range. Anyways the metalized poly caps are the size of my fist, and I immediately regretted buying them when I got the box from russia, but I installed them anyway. Some day I will try some audiophile electrolytic caps because of space constraints. I will look for the smallest tolerance and I'm guessing it will be similar performance.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Are "audio grade electrolytic capacitors" snake oil?
- Thread starter thyristor
- Start date
solderdude
Grand Contributor
seems to work though. 
Was that about the failing caps? What exactly is that supposed to protect against?seems to work though.
solderdude
Grand Contributor
Yes, the small caps fail before the caps on the secondary side do. You know what happens when caps on the secondary side do when they dry out.
The increased ESR on caps (and voltages) that are not used for regulation can have substantial spikes on those voltage lines. These can cause active components to fail even with flames as a result.
That's why I was told that it is fine to replace the faulty cap but they mentioned to also replace the still good caps on the secondary side and why they still recommend not to use long lasting caps near heatsinks as that part of the 'safety measure' would not work any more.
It made sense to me when they told me. Since then I replaced all secondary caps also when I have to replace a start-up cap.
The increased ESR on caps (and voltages) that are not used for regulation can have substantial spikes on those voltage lines. These can cause active components to fail even with flames as a result.
That's why I was told that it is fine to replace the faulty cap but they mentioned to also replace the still good caps on the secondary side and why they still recommend not to use long lasting caps near heatsinks as that part of the 'safety measure' would not work any more.
It made sense to me when they told me. Since then I replaced all secondary caps also when I have to replace a start-up cap.
It makes sense insofar it is likely what actually happens, not in the sense of being good idea. In my opinion, safety devices should never be reliant on accidental behaviour that is normally unwanted.
solderdude
Grand Contributor
I agree but understand the desire to produce something as cheap as possible with 'simple' safety measures however crude they may be.
mhardy6647
Grand Contributor
- Joined
- Dec 12, 2019
- Messages
- 11,393
- Likes
- 24,704
Electrolytics generally don't fare well (although they can sometimes be reformed and work satisfactorily). Other types can fare much, much better over time.
Some of the old "oilers" (oil-filled capacitors) are pretty darned superb in crossover use.
Here's a canonical example. I quite literally cannot afford a pair of these, but, boy howdy, if I could, I would.
(Marky now quickly dons a Nomex suit and heads for the hills!
)
DSC_7330 (2) by Mark Hardy, on Flickr
Some of those older caps have PCBs in them.Electrolytics generally don't fare well (although they can sometimes be reformed and work satisfactorily). Other types can fare much, much better over time.
Some of the old "oilers" (oil-filled capacitors) are pretty darned superb in crossover use.
Here's a canonical example. I quite literally cannot afford a pair of these, but, boy howdy, if I could, I would.
(Marky now quickly dons a Nomex suit and heads for the hills!
mhardy6647
Grand Contributor
- Joined
- Dec 12, 2019
- Messages
- 11,393
- Likes
- 24,704
Yes indeedily-do.Some of those older caps have PCBs in them.
I think that is why they sound so good.
As long as those chlorinated biphenyls stay inside 'em -- it's all good.
Way more dangerous things in most folks' homes -- e.g., the automobile they drive in (well, you know, when not locked down).
Life is all about risk management -- and relative risks. It's the relative part that many folks don't seem to 'get' very well.
Dangerousness relative risks of activities by Mark Hardy, on Flickr(from an article in the Boston Globe)
EDIT:
I used to have a better more comprehensive morph of this from a lecture by Bruce Ames - of "Ames' Test" fame, but it's lost to the ages now. Ames was chagrined by the misuse and abuse of "Ames Test" data (which essentially uses mutagenicity as a surrogate marker for carcinogenicity and/or teratogenicity) to eliminate the use of many compounds that otherwise might be very valuable, and at least in the 1980s used to conclude his lectures with a discussion of making rational choices when faced with a pallet of risky alternatives. I share his frustration.
I should also add that, to the best of my understanding, most of the challenge with PCBs was (is) environmental, and probably has more to do with the profligate, careless, and short-sighted dumping of excess/waste materials (often by manufacturers or large-volume endusers) than the intrinsic hazards of the material (when properly used and stored).
Last edited:
- Joined
- Jul 21, 2019
- Messages
- 2,365
- Likes
- 3,552
I believe you are drifting off into pure subjectivity based on no controlled testing?
Where passive crossover networks are needed, large value film capacitors can be had from Solen at prices which won't break the bank too badly. Polypropylene is an excellent capacitor dielectric for many audio applications.
Maybe not exactly relevant, but Bruce Hofer (of Audio Precision) here explains some of the causes of distortion caused by different types of resistors and capacitors. It shows that there is definitely a difference in the TYPE (electrolytic, film, ceramic, tantalum you name it). Since the causes of the non-linearity are the same but the specification isn't the same, I can imagine that the performance is also different.
However I must also add that most manufacturers don't specify the mentioned parameters. To objectively tell if one capacitor is better than the other you'd need to measure it.
I would love to see a shoot out of various kind of types of capacitors, differences between industrial manufacturers and audiophool branded ones and the effect in powersupply decoupling and signal path. Amir already did it for low inductive resistors and actually DID measure a difference (albeit at very high frequencies). Whether it's audible is left for the reader to be judged.
Simple schematic could be made for all of these using two unity gain buffers and the DUT in between them.
solderdude
Grand Contributor
It is already known that capacitors measure different and resistors also in some respects. This has been shown in measurements through decades already.
The question is whether -90dB and -100dB is actually audible (of sonic consequence), and even more so with actual music.
So while it is easy to show there are measurable differences it is less obvious how these differences work out with real music.
IME you need quite a lot of distortion for this to be audible.
And a second thing is one should use components that are suited for the purpose.
I mean it would be very silly to use a Z5U ceramic if you want to make a stable filter or oscillator when one could have gotten the same value in C0G/NP0 ceramic.
When I was young I built 2 exact same pre-amps (using OP27 opamps in those days) where one was DC coupled and the other one was AC coupled with a bunch of electrolytics and film caps in series (high enough value not to cause roll-off and when AB ing I could not tell any difference.
I am certain that when I could measure that deep down in those days I would likely have measured difference performance.
The whole 'timbre changing' nonsense I read here and there when used in pre-amps, DACs, or as coupling/feedback caps in amplifiers is highly unlikely.
The question is whether -90dB and -100dB is actually audible (of sonic consequence), and even more so with actual music.
So while it is easy to show there are measurable differences it is less obvious how these differences work out with real music.
IME you need quite a lot of distortion for this to be audible.
And a second thing is one should use components that are suited for the purpose.
I mean it would be very silly to use a Z5U ceramic if you want to make a stable filter or oscillator when one could have gotten the same value in C0G/NP0 ceramic.
When I was young I built 2 exact same pre-amps (using OP27 opamps in those days) where one was DC coupled and the other one was AC coupled with a bunch of electrolytics and film caps in series (high enough value not to cause roll-off and when AB ing I could not tell any difference.
I am certain that when I could measure that deep down in those days I would likely have measured difference performance.
The whole 'timbre changing' nonsense I read here and there when used in pre-amps, DACs, or as coupling/feedback caps in amplifiers is highly unlikely.
eliash
Senior Member
Just did a recapping of my fathers Phillips CD100 CD Player from 1983, still made in the Netherlands.
The power supply consists of a bunch of linear regulators with lots of thermal dissipation (especially since running with 230V, instead of 220V as being designed for - changed it to 240V now, but still getting quite hot on the rear heatsink).
Being interested in this issue of cap aging as well, I compared the capacitances of the "bigger" power supply related caps and some others, not loaded with ripple current and not surrounded by high temp sources with new ones.
All the power supply caps had less than 50% of their rated capacity, some loaded with ripple current had less than 30% capacitance, none was completely dead. Other blocking and coupling caps were still fine from the perspective of capacitance, no idea if the ESR had raised, since measurements were taken on an even older ballistic meter...Also replaced the power supply related ones with similar mech. size ones with 105°C and higher voltage rating, as offered today (by Panasonic - 40pcs in total...)
Photo taken before complete teardown...still playing pretty well after extensive alignment and some version upgrade, even though it has some CD-format problems with some newer released CDs, not obeying the Red Book standard completely, as I suspect, or certain ones with refraction anomalies...anyway, now permanently connected to my stereo as the only CD-player...
Capxcon capacitors by the way
ZeDestructor
Active Member
- Joined
- Jul 28, 2019
- Messages
- 119
- Likes
- 68
CapXon have been perfectly fine for 10 years or so, now that the capacitor plague is mostly done. This has been evidenced from various server SMPS with CapXon caps lasting 5-10 years in that time.
You have to go quite far down capacitor tiers to reach the bad capacitor region.
ZeDestructor
Active Member
- Joined
- Jul 28, 2019
- Messages
- 119
- Likes
- 68
Very likely some old stock from the later 2000s with the incomplete stolen, unstable electrolyte formula (read more here). And even then, good designers like Delta Electronics make a shit Capxon cap last years, while a less good designer like Fortron/FSP will squander high-quality Hitachi or Rubycons. It's just like DAC chip implementations in that respect - you can have all the top components, and still make an unimpressive product.
As it stands right now, in 2020, the capacitor plague is truly over, and tier 2 caps like CapXon, Teapo are fine, and even tier 3 caps are fine in less space constrained or less performance-conscious applications. It's when you hit the tier 4 and lower companies, like Fuhjjyuu that you should start getting leery.
Sorry but I disagree. Capxon capacitors, besides the capacitors plague, are constantly associated with failing devices. They were exploding two years ago, they were in the Samsung tvc, they are in Behringer products with high fail rate, they are in failing SMPS. They are all over the internet and from a customer standpoint I would not buy a device with CapXon capacitors inside, ever.