Dielectric Constants of film & its impact on HiFi

[onanysunday]

As with other materials in this hobby, such as various metals used for plugs and conductors and their (IACS) conductivity rating from least conductive to most conductive and whether they are alloys or pure metals, there's also a hierarchy of film materials used as insulators or dielectrics to help implement the function of a cable or capacitor. When an insulator makes direct contact with a conductor, there are some negative effects and interactions that occur- namely dielectric loss and added capacitance to the signal. Because of this, what goes in is not equal to what comes out. However, it is very close. For instance, unless you had a bare speaker wire from I/O with no insulator surrounding it, the sound is being somewhat colored by the insulator in contact with it when the speaker receives the signal. This is because an insulator interacts with and leeches the signal of the conductor to a small degree. It is a necessary evil we have to put up with. Unless you are "listening" for this, this phenomenon does not really matter outside the audio realm as far as I'm aware. While there is science to back up this notion, it mostly only matters for audio reproduction, specifically regarding "sound".

However, some films perform better than others. Not necessarily in the sense they're adding something "magical", but that they aren't taking as much away -or coloring the sound as much, think "EQ". Many films add a slight overlay or veil to the sound. Teflon/ptfe is a good example of a superior film that is better than most other alternatives -just slightly above polypropylene. Because of this, teflon capacitors are generally regarded as best-in-class. This could matter in a speaker crossover or coupling cap (for instance) among other applications such as choosing wiring/ cabling insulated with these better film materials for less interaction/ degradation of the sound. From my own trials and empirical experiments, a film with a dielectric constant (K) of 2.XX or less sounds better than a film with a higher-K than this. Some examples of films with a relatively higher-K compared to the best alternatives would be acrylic, vinyl, polyvinyl chloride (PVC), polyester and all its derivatives. I've always liked the sound of mylar the most of all polyesters which makes sense, considering it has a lower K (3.1) than base polyester (3.6). Somewhat related, and take it fwiw, regarding a "cable break-in period", insulator molecules may draw energy from the conductor, re-arranging themselves (and the conductor itself) into a more uniform structure over time.

Some examples of films with lower-K include teflon/ptfe, polystyrene, polypropylene, PE (among others). There are even mixtures, such as PE Foam, or other films made into "foam insulation" cable material. The advantage of this is that the tiny air bubbles injected into the film have a K of only 1, so when combined with a low-K film, the resulting K is even lower than its base film material (but still not quite 1 which is the reference point for air/ vacuum). Other materials besides film that make good insulators for cables/ wires include cotton, air (as mentioned) and perhaps others.

Long story short, whenever you are able to choose film materials with a lower dielectric constant of around K=2.XX (or less) the sound will be truer to its source. Plastics make it possible.

Dielectric constants (averaged) of commonly used films/ materials in audio:

Air = 1.006
Acrylic = 3.1
Butyl Rubber = 2.3
Cotton = 1.3
EPR = 2.2
FEP = 2.1
Foam PE = 1.6
Fluoropolymers = 2.0
MPPE = 2.2
Mylar = 3.1
Oil = 2.4
Paper = 3
PE = 2.3
PFA = 2.1
Polyester = 3.6
Polycarbonate = 3.1
Polypropylene = 2.2
Polystyrene = 2.6
Polyurethane (PU) = 1.1
Polyvinyl Chloride (PVC) = 3.6
PPS = 3
Silicone = 2.8
Teflon/ptfe = 2.1
TPE = 2.2
Vacuum = 1
Vinyl = 4
XLPE = 2.3

 

[Thorsten R.]

And you have verified this via blind A/B tests?

 

[mcgjohn]

on caps only looking at the dielectric is only half the problem. Also need to look at the electrode material. Also interesting to note many folks who have updated their gear with Teflon caps either sell the gear a short time afterwards or yank the caps out.

a person may want to return and read the Marsh article on caps written back in the 1980s.

 

[rcs16]

I've always liked the sound of mylar the most of all polyesters

Mylar is a trade mark for polyester?
I usually get banned responding to these threads posted by hobbyists because they are opinions without any scientific evidence. Pretty much all electronics uses PVC coated wire because it’s cheaper. Last report has Richard March giving up on these audio forms and living in Thailand.
Name one consumer grade audio component manufacturer using Teflon caps? And who even makes or sells them, good luck with that.

 

[onanysunday]

Mylar is a trade mark for polyester?
I usually get banned responding to these threads posted by hobbyists because they are opinions without any scientific evidence. Pretty much all electronics uses PVC coated wire because it’s cheaper. Last report has Richard March giving up on these audio forms and living in Thailand.
Name one consumer grade audio component manufacturer using Teflon caps? And who even makes or sells them, good luck with that.

Not all polyester is mylar, but mylar (PET) is a registered trasmark of DuPont and is a type of polyester. Other types of polyester that are not mylar include PEN, PBT, PTT. The dielectric constant/ loss is the science behind why some insulators behave differently than others. They do not all behave equally to one another (electrically speaking). Audio Research uses Teflon caps they've branded as "ARC". I don't know if they made them or simply re-badged them. I have a bag of them I've been using for speaker crossovers and as coupling caps. I also have some Russian military surplus "K72" teflon caps as well that I use.

 

[onanysunday]

Indeed, the leads and electrode material of capacitors will have an effect on its sound as well, but all else being equal, the the lower-K insulators provide better sound when all other materials remain constant. The difference I've heard and noticed is that lower-K films sound clearer and have better musical separation than higher-K films. These differences are only noticeable when you compare them directly. For instance, OFC speaker wire sheathed in PVC may sound just fine by itself. It's not necessarily the case you're listening to that wire and thinking to yourself it sounds "bad", not at all. You get used to it and it sounds just fine to your ears. It's only through direct comparison to other materials that differences in sound emerge. I've never gone back in and yanked teflons out. They consistently sound better than other alternatives I've tried.

 

[onanysunday]

The point of this (I am hoping) is that this just becomes another data point to consider with your material selection. I am not here to argue or prove anything to anyone. Take it for what it's worth to you, which may be nothing at all. It should not just be ignored outright or discredited either as it impacts sound for those that are interested in sound as it effects everything, everyone, and all equipment. I've been listening to and comparing different materials in audio for many years. Most internal wiring in equipment is regular non-ofc stranded copper insulated in PVC (or similar). It's also the case that the wiring going into our homes could be hundreds of miles worth of non-ofc copper insulated in lossy insulators.

The thing I've found though, is that with the introduction of different materials into your system at different entry points near the end of the chain, it's never an "all or nothing" scenario; meaning that making small changes to some materials used at the endpoint (i.e. your equipment) greatly effects the sound that is heard, even when everything leading up to it may just be standard or nothing different or special. I think of all the times I've added silver at some point in the signal chain and the drastic change in sound it immediately provides -regardless of all the other materials leading up to that point in the signal chain that have remained constant. Films similarly impact sound on a scale from clarity at one end to muffled on the other. If we were using cables with insulators that were higher-K than types commonly available to us, it would become immediately apparent to anyone listening their potentially detrimental effect to sound. But the ones that are available to us all sound pretty decent which is why they are commonly used and are our only reference point. While the differences may not be as drastic as they could otherwise be, they are still impacting sound -whether we choose to believe it (or not).

 

[onanysunday]

One example of this would be think of all the times an mkt polyester cap presented more detail or musical separation than an mkp polypropylene cap. I can't think of a single time. This is why. We may prefer the polyester cap, saying it's a bit more natural, rounded, smooth, averaged, in the middle, easier to listen to, what we're used to hearing and is therefore our preference, and that is perfectly fine as it's all just a matter of choice or personal preference anyway. But the maximum fidelity or performance ceiling heard from using polyester or pvc (in an absolute sense) is going to be less than teflon or polypropylene when other contributing materials to the component or equipment remains constant.

 

[gadget73]

Teflon caps? And who even makes or sells them,

The only semi-common source I see of these is old Soviet surplus. I have some that I'd like to try in something, unfortunately they aren't of a value that anything I own uses so they're just sitting in their box. I'm not much of a believer in the sound of caps, but I like to keep an open mind.

 

[arts]

Name one consumer grade audio component manufacturer using Teflon caps? And who even makes or sells them, good luck with that.

Reliable Capacitors (USA); Rel Cap RT
Solen (SCR, France); FEP Metallized Film, FEP/SN Film-Tin Foil

 

[Capitol C]

Capacitive effects in speaker wires are negligible unless the wire's designer made a special effort to screw things up. All that matters for speaker wires is that the dielectric has low losses (the imaginary part of the dielectric constant is small) and that the dielectric constant is independent of frequency over the audio frequency range. If these criteria are met, the speaker wire will not affect the signal.

 

[John3655]

Hmmm don't see wood listed, wooden capacitors.

 

[Aerobat]

A coupling cap's sole purpose is to filter out DC. There's a formula, I forget which, that determines the appropriate value in microfarads and it should be as low as possible. Or, you could go with a Nelson Pass design with zero DC offset and no coupling caps.

Since the AC signal is going through the coupling cap it has to have an impact on the sound IMO.

 

[gslikker]

Hmmm don't see wood listed, wooden capacitors.

You could try in combination with the sound interconnects materials test in another forum, just google "copper banana mud interconnect blind test"
I could imagine, the wood could be a bit leaky if humidity is high.

 

[onanysunday]

Here's another correlation for you that's somewhat similar in that it has the same "general" type of effect or outcome. As touched on earlier, there's the connection between conductivity rating and the level of detail or musical separation heard in the audio signal. For instance, pure silver IACS 106% (not an alloy) consistently provides more clarity and detail compared to pure red copper IACS 100%.

Now, keeping that in mind, imagine what it's like to recap an old receiver from the seventies. Really listen to it beforehand. Focus on the low-end, midrange, and upper frequencies as separate parts of sound taken and heard in isolation from one another. How detailed is the sound? Take notes if you need to so you'll remember. Does the bass sound boomy, soft around the edges, thick, syrupy, one-note, or lack definition? What are the highs and midrange like? Now, recap it and run the same listening test. What you'll notice if your old caps were out-of-spec is that suddenly musical separation, imaging and clarity have returned to the music. I notice it most especially with imaging clarity, and in the bass where things really seem to get cleaned up and beats have a more succinct attack and decay to them afterwards. This is the same general change observed using different films as capacitors or cable insulators.

The lower-K films allow better musical separation or clarity of individual instruments to emerge from the music in isolation from the surrounding musical information, higher-K tends to blend all that same music info together more. It's this "blending" or smearing of musical information that's really the crux of it (as far as I can tell) and what the difference is that you can expect to hear, but only when two different (K) materials are compared against one another, directly, and in real-time. Otherwise, everything is (and will always sound) just fine and remain the same. Nothing has to change. It's just "fine" the way it is, but it's a fun eye-opening experiment to run nevertheless. Also, the nice thing about it is that a person does not have to swap out all materials to notice a change or improvement in sound. It's shocking sometimes just how little you can change to hear an audible difference in sound (for the better) when everything else remains constant. Just think of all the times you noticed a change in sound when only swapping out a pair of cables.

 

[audiotemp]

Here's another correlation for you that's somewhat similar in that it has the same "general" type of effect or outcome. As touched on earlier, there's the connection between conductivity rating and the level of detail or musical separation heard in the audio signal. For instance, pure silver IACS 106% (not an alloy) consistently provides more clarity and detail compared to pure red copper IACS 100%.

Now, keeping that in mind, imagine what it's like to recap an old receiver from the seventies. Really listen to it beforehand. Focus on the low-end, midrange, and upper frequencies as separate parts of sound taken and heard in isolation from one another. How detailed is the sound? Take notes if you need to so you'll remember. Does the bass sound boomy, soft around the edges, thick, syrupy, one-note, or lack definition? What are the highs and midrange like? Now, recap it and run the same listening test. What you'll notice if your old caps were out-of-spec is that suddenly musical separation, imaging and clarity have returned to the music. I notice it most especially with imaging clarity, and in the bass where things really seem to get cleaned up and beats have a more succinct attack and decay to them afterwards. This is the same general change observed using different films as capacitors or cable insulators.

The lower-K films allow better musical separation or clarity of individual instruments to emerge from the music in isolation from the surrounding musical information, higher-K tends to blend all that same music info together more. It's this "blending" or smearing of musical information that's really the crux of it (as far as I can tell) and what the difference is that you can expect to hear, but only when two different (K) materials are compared against one another, directly, and in real-time. Otherwise, everything is (and will always sound) just fine and remain the same. Nothing has to change. It's just "fine" the way it is, but it's a fun eye-opening experiment to run nevertheless. Also, the nice thing about it is that a person does not have to swap out all materials to notice a change or improvement in sound. It's shocking sometimes just how little you can change to hear an audible difference in sound (for the better) when everything else remains constant. Just think of all the times you noticed a change in sound when only swapping out a pair of cables.

This stuff you're posting in this thread is not widely accepted as factual. Are these just your opinions?

 

[John3655]

Wood actually isn't totally daft, paper in oil, effectively wood pulp, also Pioneer produced bamboo capacitors. Burnt paper/carbon was used in Black Gate capacitors.
Google tells me the DK varies between 1.5 and 4.5.

 

[onanysunday]

This information is what I've found for myself after doing my own trials and experiments in materials and sound over the years. I couldn't find an explanation for it (at first) why some films consistently sounded not just different, but better than others. Not just in the sense of my particular taste in music, but in an absolute sense of overall resolution, clarity and a higher performance ceiling. Finally, after researching conductivity in-depth and how that impacts sound with regards to conductors, I stumbled upon the science of insulators as well, i.e. dielectric constant (K) / dielectric loss / added capacitance, etc. As it turned out, it all fits together. I found that my findings and listening observations coincided almost directly with the (K) of these films, whereas a lower (K) consistently provides a clearer sound. This is/was observed when other parameters and materials remained constant during testing. Since then, I've used information provided here to make different component choices in my projects and system equipment (both internally i.e. component or wiring choices, and externally -cables, connectors) that have consistently helped me achieve a higher level of performance in sound quality than was possible before.

 

[pete_mac]

And you have verified this via blind A/B tests?

You may wish to read the rules of this section of the forum...

 

[Capitol C]

This stuff you're posting in this thread is not widely accepted as factual. Are these just your opinions?

Silver has a conductivity that is 6 percent higher than copper. This means that if you have a silver wire and a copper wire of the same diameter and length the with the same diameter and length, the silver wire will have a 6% lower resistance (actually 5.66% if you insist on using a calculator). You can get the same effect using copper by increasing the diameter by 3% (again, 2.96% if you use a calculator). Whether you use silver or slightly thicker copper, the effect is negligible.