Absolutely!
Very low ESR parts can make a voltage regulator unstable. They may also resonate with other capacitors depending on trace inductance between the two.
Very low ESR parts can make a voltage regulator unstable. They may also resonate with other capacitors depending on trace inductance between the two.
Actually have used solid polymer less, and less, the only ones I’ve gotten to work are lower values (slightly higher esr).
Clock power supplies seem to respond well with them, along with a tiny ceramic bypass that (hopefully) won’t resonate.
I have high hopes for the future of similar parts however.
Clock power supplies seem to respond well with them, along with a tiny ceramic bypass that (hopefully) won’t resonate.
I have high hopes for the future of similar parts however.
Hi Guys
DC servos seemed to become very popular with the "DC to green light" amps of the late '70s early '80s. I only use them if the amp does not have enough DC gain to make an offset <1mV. I prefer not to be able to measure the offset, so if my meter cannot read it then I'm happy.
The problem with simulations is that it is not easy to test capacitor dielectrics. maybe you can but I have really limited sim skills, really just what Bob Cordell explained how to do in his book. Thanks Bob.
I guess it will come down to trying different caps and measure if the DC offset changes - maybe make a boost amp in front of my meter akin to Cordell's "distortion magnifier".
So, here is the question of the day: What influences electrolytic leakage current?
Thanks
DC servos seemed to become very popular with the "DC to green light" amps of the late '70s early '80s. I only use them if the amp does not have enough DC gain to make an offset <1mV. I prefer not to be able to measure the offset, so if my meter cannot read it then I'm happy.
The problem with simulations is that it is not easy to test capacitor dielectrics. maybe you can but I have really limited sim skills, really just what Bob Cordell explained how to do in his book. Thanks Bob.
I guess it will come down to trying different caps and measure if the DC offset changes - maybe make a boost amp in front of my meter akin to Cordell's "distortion magnifier".
So, here is the question of the day: What influences electrolytic leakage current?
Thanks
Aluminum oxide layer, foil surface finish, and electrolyte composition.
The aluminum oxide layer is carefully done to support the intended purpose of the part. Try reversing the power to an uncharged electrolytic capacitor without making it explode, and then measure the leakage currents vs an undamaged part.
The aluminum oxide layer is carefully done to support the intended purpose of the part. Try reversing the power to an uncharged electrolytic capacitor without making it explode, and then measure the leakage currents vs an undamaged part.
For a Tant, reverse polarity is INSTANT DEATH. Aluminium Electrolytics are more robust and may survive .. maybe with a bit more leakage and/or noise. Modern cheapo Lo-Leakage Aluminium Electrolytics have leakage currents nearly as low as Tants IF OPERATED WITH A BIT OF THE CORRECT VOLTAGE ACROSS THEM.
One reason I like 5532/4 is that the high input bias currents mean you can usually arrange 100mV or so of voltage across the input & feedback electrolytics. This is a recipe for long life & low leakage.
Dunno about the 'polymer' electrolytics but for VLN applications, they are noisier than cheapo Aluminium Electrolytics.
Hi
I know for a fact that the voltage across an electrolytic can be reversed without damage provided the voltage is very low. This happens in DC blockers for mains where the cap is paralleled with anti-parallel diodes that clamp the voltage. And since the main topic we are talking about is using electros and polymer caps as signal couplers, the audio signal itself is reversing polarity and the caps survive for decades.
In the example of the gain circuit pot, the voltage across the cap was DC and tens of volts, with a signal voltage superimposed.
It seems then that polymer caps are not a good choice as audio coupling caps? which is just as well, since they are not available in as many values as standard electrolytics.
Tantalums should be kept many kilometers away from audio equipment 🙂
Marsh and Jung said so.
A guy I knew in an alarm system repair lab would sometimes put a tantalum backwards across a battery. The Top of the cap would turn bright orange and fire across the room. Fortunately they blew up in a very consistent manner.
I know for a fact that the voltage across an electrolytic can be reversed without damage provided the voltage is very low. This happens in DC blockers for mains where the cap is paralleled with anti-parallel diodes that clamp the voltage. And since the main topic we are talking about is using electros and polymer caps as signal couplers, the audio signal itself is reversing polarity and the caps survive for decades.
In the example of the gain circuit pot, the voltage across the cap was DC and tens of volts, with a signal voltage superimposed.
It seems then that polymer caps are not a good choice as audio coupling caps? which is just as well, since they are not available in as many values as standard electrolytics.
Tantalums should be kept many kilometers away from audio equipment 🙂
Marsh and Jung said so.
A guy I knew in an alarm system repair lab would sometimes put a tantalum backwards across a battery. The Top of the cap would turn bright orange and fire across the room. Fortunately they blew up in a very consistent manner.
You constantly ask if the Polymer is audio suitable;-) It helps just listening, hearing;-) Otherwise, the things have the typical bipolar capacitor properties that should be considered for their electrical connection;-)
How tantalums sound, I can’t say more, I think they had a clear harsh tone in the sound.
The one Polymer I mentioned above: he did not have the clarity, blackness, rhythm and contour as I wish, but he sounded very elaborate and colorful.
How tantalums sound, I can’t say more, I think they had a clear harsh tone in the sound.
The one Polymer I mentioned above: he did not have the clarity, blackness, rhythm and contour as I wish, but he sounded very elaborate and colorful.
Maybe that's a good thing? For all we know these could be simple linear effects you're describing, and you've previously tweaked your system to suit something that's not so good?
World = frequencies. Material vibrations remain as an explanation. And also diameters and shapes and physical layout. An attempt to explain them at the usual educational level of frequency response, or current as a minimal concept symbolically represented as lines and numbers, is not sufficient here. The ear does not perceive lines and numbers. Physics is required.
Hi
It became clear early in the thread that hybrid polymer caps are not suited to the signal path if you want lowest distortion and lowest noise. I was looking for quantitative and science-based information, as I still am regarding the leakage current of electrolytics when used as signal coupling caps.
I use polypropylene coupling caps wherever the circuit resistance values allow. These have measured performance that is well suited to audio circuits where low-THD is the goal. They have essentially zero leakage, too, and won't upset DC conditions. A milliFarad of PP is expensive and takes a lot of space. partly because you can only get very high values as high-voltage parts. In the "1k environment" I need 1mF for signal coupling as a minimum, but 3m3 or 4m7 work even better when going into parts-per-billion THD.
There are a lot of circuits and combinations of components that "sound good", and often it depends on the type of music played through it. My observation is that the proponents of simple high-THD circuits tend to listen to open airy music. Those circuits sound like mud when the midrange gets busy - lots of instrument overlap, no space between very fast notes, or distorted instrument tones. I am interested in many types of circuits using tubes, solid-state and hybrid, but I always find that the circuits with the lowest THD and flattest widest frequnecy response sound best. Anything that reduces THD alo reduces IMD, which is a much worse type of distortion as its artifacts are not musically related to the signal.
Subjective opinions have their place, but I like to begin with objective facts. Ultimately your perception of the sound excites you to keep listening or to keep searching for something better. When there are too many things in your system that "have colour" of their own, swapping pieces in and out is never ending. This goes down to the component level, as well. My stereo system has been stable for many years, but I would like to go back to having an active crossover and multiple power amps, as I had that previous to a move and always wanted to get back to it. In the crossovers and the PAs, I do not want to put in a coupling cap whose distortion will swamp the rest of the amp.
Speaker distortion is typically 1% or 5% - it can be higher with complex passive crossovers, or much lower for electrostatics like the Quad ESL at 0.1% - and indeed it is the highest THD of most system elements. However, if you believe that electronic distortions are unnatural, that we did not evolve to compensate for them, then you will want to get rid of as much of that distortion as possible. Yes, our senses can resolve ppb for things that can upset us (sounds, smells, tastes). Using high-THD circuits may mask or compliment speaker THD in your own world, but I am aiming for clarity.
Of course, our own hearing distorts what we are sensing, and our brain alters it further. Since aural compression kicks in in the presence of all sound, it may be impossible for us to hear anything accurately? But we try to control those distortions we have some sway over.
It became clear early in the thread that hybrid polymer caps are not suited to the signal path if you want lowest distortion and lowest noise. I was looking for quantitative and science-based information, as I still am regarding the leakage current of electrolytics when used as signal coupling caps.
I use polypropylene coupling caps wherever the circuit resistance values allow. These have measured performance that is well suited to audio circuits where low-THD is the goal. They have essentially zero leakage, too, and won't upset DC conditions. A milliFarad of PP is expensive and takes a lot of space. partly because you can only get very high values as high-voltage parts. In the "1k environment" I need 1mF for signal coupling as a minimum, but 3m3 or 4m7 work even better when going into parts-per-billion THD.
There are a lot of circuits and combinations of components that "sound good", and often it depends on the type of music played through it. My observation is that the proponents of simple high-THD circuits tend to listen to open airy music. Those circuits sound like mud when the midrange gets busy - lots of instrument overlap, no space between very fast notes, or distorted instrument tones. I am interested in many types of circuits using tubes, solid-state and hybrid, but I always find that the circuits with the lowest THD and flattest widest frequnecy response sound best. Anything that reduces THD alo reduces IMD, which is a much worse type of distortion as its artifacts are not musically related to the signal.
Subjective opinions have their place, but I like to begin with objective facts. Ultimately your perception of the sound excites you to keep listening or to keep searching for something better. When there are too many things in your system that "have colour" of their own, swapping pieces in and out is never ending. This goes down to the component level, as well. My stereo system has been stable for many years, but I would like to go back to having an active crossover and multiple power amps, as I had that previous to a move and always wanted to get back to it. In the crossovers and the PAs, I do not want to put in a coupling cap whose distortion will swamp the rest of the amp.
Speaker distortion is typically 1% or 5% - it can be higher with complex passive crossovers, or much lower for electrostatics like the Quad ESL at 0.1% - and indeed it is the highest THD of most system elements. However, if you believe that electronic distortions are unnatural, that we did not evolve to compensate for them, then you will want to get rid of as much of that distortion as possible. Yes, our senses can resolve ppb for things that can upset us (sounds, smells, tastes). Using high-THD circuits may mask or compliment speaker THD in your own world, but I am aiming for clarity.
Of course, our own hearing distorts what we are sensing, and our brain alters it further. Since aural compression kicks in in the presence of all sound, it may be impossible for us to hear anything accurately? But we try to control those distortions we have some sway over.
Hi nauta,
I haven't had the time to listen and measure poly vs electrolytic caps in working amplifiers yet. I have seen improvements in some power supply and decoupling circuits. Poly capacitors are still sort of new for audio work. Wurth has some test reports available. I wouldn't rule them out in audio signal paths yet or maybe in the future. I suspect that reports on the sound of these might depend more on the ideas in the listener's head than the reality. Informal measurements have been inconclusive so far, as have listening tests.
It comes down to putting the correct component type in the circuit depending on the characteristics. If you can hear it, you can darned well measure it using good enough equipment. These parts are mature in industry, but not in the audio industry.
If you measure DC offset that depends on the capacitor, the offset probably always goes in the same direction and it will be due to leakage currents in that case. If the part generates its own DC charge, interesting and probably best avoided for all applications.
I haven't had the time to listen and measure poly vs electrolytic caps in working amplifiers yet. I have seen improvements in some power supply and decoupling circuits. Poly capacitors are still sort of new for audio work. Wurth has some test reports available. I wouldn't rule them out in audio signal paths yet or maybe in the future. I suspect that reports on the sound of these might depend more on the ideas in the listener's head than the reality. Informal measurements have been inconclusive so far, as have listening tests.
It comes down to putting the correct component type in the circuit depending on the characteristics. If you can hear it, you can darned well measure it using good enough equipment. These parts are mature in industry, but not in the audio industry.
If you measure DC offset that depends on the capacitor, the offset probably always goes in the same direction and it will be due to leakage currents in that case. If the part generates its own DC charge, interesting and probably best avoided for all applications.
Hi
I believe most people use "poly" to refer to plastic dielectric caps, ie, polyester, polypropylene, polystyrene. "Polymer" would be a better contraction for aluminium hybrid polymer caps.
I believe most people use "poly" to refer to plastic dielectric caps, ie, polyester, polypropylene, polystyrene. "Polymer" would be a better contraction for aluminium hybrid polymer caps.
I note it on service reports as P-A. In the context of electrolytic capacitors, "poly" is pretty clear. For film caps I use the full name on reports.
Problem is the context is also capacitors, so 'poly' isn't great. Many people will assume poly and electrolytic are distinct kinds of capacitor... I don't think there is a good clear abbreviation...
Have a look at the link I posted in #4. Polymer electrolytics and Tants definitely perform worse than cheapo Aluminium electrolytics for VLN work. Noise is easily tested and/or listened to.
The link is about stuff at 0.28nV/rt(Hz) 😲 & below. But Tant noise can be heard on Electret microphone circuits too.
I often see them called "solid" capacitors, short for "solid electrolytic". It seems a reasonable name to me, as there aren't any other types of capacitor that are referred to as solid, even if they technically are.
Hi kgrlee,
That would be due to leakage current. Sitting there they don't have intrinsic noise. New ones seem to be getting better (no surprise). Poly-Aluminum are better than the solid Tantalum types were. I'm just saying that you never know how good they will turn out to be. I have tried a few Poly-Aluminum types compared in the same circuit to electrolytic types. No increased distortion or noise.
I haven't tested them for low leakage applications simply because the manufacturer's data suggests they are more leaky than electrolytic types. So far I haven't seen this. Units I have tested (many) have similar leakage currents to electrolytic. Who knows what happens over time or as temperature increases? Anyway, it isn't wise to use them in applications they are not intended for.
One thing I have seen, they don't tend to short. That's good because they are intended as power supply bypass capacitors.
That would be due to leakage current. Sitting there they don't have intrinsic noise. New ones seem to be getting better (no surprise). Poly-Aluminum are better than the solid Tantalum types were. I'm just saying that you never know how good they will turn out to be. I have tried a few Poly-Aluminum types compared in the same circuit to electrolytic types. No increased distortion or noise.
I haven't tested them for low leakage applications simply because the manufacturer's data suggests they are more leaky than electrolytic types. So far I haven't seen this. Units I have tested (many) have similar leakage currents to electrolytic. Who knows what happens over time or as temperature increases? Anyway, it isn't wise to use them in applications they are not intended for.
One thing I have seen, they don't tend to short. That's good because they are intended as power supply bypass capacitors.
At those noise levels - yes. Many wish their electronics were that quiet.
Hi
I basically put tantalums out of my brain decades ago when Marsh and Jung wrote their capacitor article. I cannot believe anyone would use them in an audio amp? I believe there is a crazy thread here about a 25k-British pound phono preamp that seems to be stuffed with tantalum caps.
I basically put tantalums out of my brain decades ago when Marsh and Jung wrote their capacitor article. I cannot believe anyone would use them in an audio amp? I believe there is a crazy thread here about a 25k-British pound phono preamp that seems to be stuffed with tantalum caps.
The first hybrid polymer aluminium caps that I heard of were Sanyo's OSCON series.
I think it was in Hi Fi World that they were extolling their virtues, mainly because they were more expensive, but for de-coupling rather than coupling.
It was a long time ago, though, so this debate has been going on for many years now.
It's "organic semiconductor", so maybe different to polymer hybrid, or maybe its just a fancy marketing term for the same thing.
I think it was in Hi Fi World that they were extolling their virtues, mainly because they were more expensive, but for de-coupling rather than coupling.
It was a long time ago, though, so this debate has been going on for many years now.
It's "organic semiconductor", so maybe different to polymer hybrid, or maybe its just a fancy marketing term for the same thing.
Last edited:
Hi nauta,
Back then, before the article came out, I was replacing Tantalum caps with normal electrolytics for coupling applications. They sounded awful and equipment back then didn't measure low enough to show the problem. At that time those parts were the darling of the high end industry.
Hi johnnyx,
Many publications pushed these capacitors. Oscon from Sanyo were supposed to be the next perfect thing. I still have some old samples.
Like anything, improvements are made over time. For now in coupling capacitor applications, I don't see why you would want to spend more to end up at the same place.
Back then, before the article came out, I was replacing Tantalum caps with normal electrolytics for coupling applications. They sounded awful and equipment back then didn't measure low enough to show the problem. At that time those parts were the darling of the high end industry.
Hi johnnyx,
Many publications pushed these capacitors. Oscon from Sanyo were supposed to be the next perfect thing. I still have some old samples.
Like anything, improvements are made over time. For now in coupling capacitor applications, I don't see why you would want to spend more to end up at the same place.