Coupling Cap Types and Voodoo Parts?

[makinrose]

I was wondering what people's opinions were on coupling cap brands and materials.  There lots of expensive parts out there that in the fine print seem to be  polyester or polypropylene caps in fancy large packages.  Is that any difference in sound? Can there be? or is it just a money grab? 


My experience has been that I can hear difference between different material types.  To me polyester/mylar caps sound a little warmer and less detailed than polypropylene.  Ceramic tend to more gritty and harsh. New paper type caps sound a lot like polyester but maybe a little less bright.


The other thing I've found I really cannot tell much if any difference between brands and construction types of the same material.  For example the little Panasonic radial polyester caps sound much the same as the Mallory 150 film and foil polyester. For that reason I'm really skeptical that spending $5 to $15 on coupling caps is worth while. Does the size or construction have noticeable tonal effect? 

What's the lowdown on it?   Thanks! I'm really interest to hear everyone's thoughts.

[K O'Connor]

Hi Guys

Polester/mylar adds much more distortion than polypropylene, which is essentially distortion-free. Polyester adds odd-order harmonics which will be heard in different ways insofar as these harmonics become part of the input signal to the following gain stage, which distorts the fundamental and all these extra tones, so "fatter" is a possibility. The extent to which the cap distorts depends on the signal voltage across the cap.

If you want minimal ceramic cap distortion, use NP0 or C0G; for maximum grit use X7R.

Within a given dielectric range, the physical size of the cap will effect parasitic elements contributing to tone, as TUT3 states.

[makinrose]

Hi Guys

Polester/mylar adds much more distortion than polypropylene, which is essentially distortion-free. Polyester adds odd-order harmonics which will be heard in different ways insofar as these harmonics become part of the input signal to the following gain stage, which distorts the fundamental and all these extra tones, so "fatter" is a possibility. The extent to which the cap distorts depends on the signal voltage across the cap.

If you want minimal ceramic cap distortion, use NP0 or C0G; for maximum grit use X7R.

Within a given dielectric range, the physical size of the cap will effect parasitic elements contributing to tone, as TUT3 states.

That  explains my affinity for polyester caps!  
 
The  line  that stuck out to me in TUT 3  was in 4-12 and reads " All that is required to make small parts sound like large ones, is an airy layout".  So what I'm inferring is that that state in conjunction with what you wrote on the post is that the parasitic elements are a function of layout (which is influenced by the size of the parts)  rather than the  size directly?

Which lead to one more question:  when you layout your amps are there rules of thumb that you use to ensure the spacing used minimizes parasitic element?

[K O'Connor]

Hi Greg

Do you mean "you" as me or "you" as anyone?

I don't use a rule of thumb for that but you can look at industry standards that follow a rule of life inasmuch as "empty space does not remain empty, which also applies to time. Take any hand-wired Fender amp. The circuitry is very simple yet it manages to occupy the entire internal space of the chassis, from Champ to Twin Reverb. There is lots of space around the components even though they seem cheek-by-jowl and the main filter caps on nominally "outside" the chassis. If you look more closely there is more space than occupation, so the parasitics in this assembly are low.

In more modern amps, costs are considered differently and most components are physically smaller than vintage types, combining to make tight assemblies with true empty space around the boards. In some cases, the boards are laid out airily to keep wiring short, as in re-issue Fender amps. The tight board has small parts and proportionately the parasitics might not have increased very much, while the airy modern board with small parts likely has similar parasitics to the eyelet board. The board materials will be different but the spacings are also different.

TUT3's guideline is an easy one for any hobbyist to follow, and choosing physically large components can force an airy layout, but could also lead to tight spacing if everything is large. There is a balance when you consider how the component carrier layout influences how the wired connections to the carrier must be laid out, sometimes beneficially and sometimes not.

In a test case, you might have a reference amp who's tone you wish to capture in a new build, but using entirely different component types and assembly methods. Intuitively you might begin with the same circuit, but you might find that you have to add some new circuitry or make modifications to be true to the tone. That is fine. That's what I do as I see each new format as a different amp even if it is the same circuit.

[makinrose]

Kevin,

Thanks for explanation. I had meant you/London Power when I had asked since I'm curious abut your design methodology.
One of the reasons I ask was because I'm in the situation I think lots of your readers are in: as they learn more about proper design they find the common pre-made boards and some chassis for popular amp models won't accommodate the extra de-coupling and galactic grounding. It's just better to build a board and sometimes use different chassis too. The last paragraph is particularly helpful to me. Thank you!

[jmcd]

If you want minimal ceramic cap distortion, use NP0 or C0G; for maximum grit use X7R.

Where do Y5P, SL, and Z5U fall on the ceramic cap grit continuum?

[I suggest this thread be moved to the preamp mods section, or perhaps a new section for components.]

[K O'Connor]

Hi Guys

Ceramic "tone" as it were is not necessarily linear with type. In a broad stroke, you could put them from cleanest to dirtiest corresponding to their temperature stability, where the most stable types are the cleanest.

Also, the higher the capacitance the more ceramic distortion there will be. This is why for the most high fidelity applications, ceramics over 100pF are rarely used.

[Champ81]

Hi,

I found this post interesting. Concerning ceramic disc caps i have been trying to obtain some from digikey and mouser but it seems the lead lengths are much shorter to fit them into the turret spacings I got.   unless I piggyback them onto another component. I settled for some mica caps but I heard also people don't like them for some reason and they have a particular "tone". Any thoughts on the sound of mica vs ceramic disc's? And why some people prefer the ceramic disc's? Is it for the reasons above? Like the ceramic "grit".

[K O'Connor]

Hi Guys

Champ81: What values are you looking for? In cases where you want long leads, as we always do for hand wiring, you end up going to high-voltage types (1kV+) and maybe to safety caps. Myrata has many long-lead ceramics that are typically X or Y tolerances.

When you do the searches at Digikey or Mouser, select only one parameter at a time and then 'apply'. Selecting too many restrictions often brings back zero results.

With the short-lead caps, use bus wire to extend them to reach the terminals, or solder it across a small resistor then cut out the resistor.

Mica is more neutral compared to ceramic, so you see hobbyists swapping a mica for the 270oF treble cap in Fender EQs. In a Marshall, you might hear a difference, but maybe not, as the EQ is less effective due to being cathode-driven AND the high IM of Marshall's wiring arrangement could easily mask the cap tone change.

Makinrose: The Ys fall between the Cs and Xs grit-wise. The differences are small, and as I said above, value-dependent - more noticeable with high values than with low.

In my own amps, I generally use plastic caps everywhere in the signal path, and even in the kits. The REV Reverb kit may be the last one with a ceramic in it, but it is <100pF so not too disconcerting

[makinrose]

It's been a while since this thread was started. I did quite a bit of experimenting with ceramic caps v.s plastic caps since then. In the end I came the conclusion that I prefer to use polyester or polypropylene. Better clarity is what I like. To my ear the brand and construction of plastic caps doesn't make much difference as long as it's the same dielectric. I know there are people who claim foil/film is better than a metalized cap but I cannot hear the difference. Overall, the experience has reenforced the notion that you need to be using quality parts, but the circuit design is where most of the tweaking really needs to occur. Over time I've become far less precious about copying old amp circuits and more concerned about building amps that are a like the versions of those amps that players have in their heads. Just my two cents.

[nauta]

Hey makinrose

I think thats a really valuable 2-cents

[makinrose]

Thanks Nauta!

[Resistor001]

Hi Guys,

I've a question about obtaining 270pF polyester caps rated for 400V. I'm sourcing parts for the Tot Reverb Project.

One of the coupling caps is stated as being the value previously stated. I'm in Australia and have no intention of sourcing a single cap from overseas - I would if I knew I'd need a lot, but in this case, I doubt I would in the future. I cannot find a polyester cap of this value at the specified rating. What alternatives do I have? I see mica caps at this value, which I assume would be fine for stopping AC (not AC, DC... duh). However, they may add more distortion to the reverb signal (if so, is it significant???).

Another issue is that I see lots of Cornell Dubilier caps and I know they bought out Illinois Capacitors. I buy from DigiKey and DK seem to be labelling caps made out of the IC factory as CD capacitors. So, now I have an aversion to purchasing any caps branded as CD. Did Cornell make decent caps before they bought IC and if so, how do I differentiate when DK does not? Are all CD capacitors made in the IC factory? 

[makinrose]

Hi Guys,

I've a question about obtaining 270pF polyester caps rated for 400V. I'm sourcing parts for the Tot Reverb Project.

One of the coupling caps is stated as being the value previously stated. I'm in Australia and have no intention of sourcing a single cap from overseas - I would if I knew I'd need a lot, but in this case, I doubt I would in the future. I cannot find a polyester cap of this value at the specified rating. What alternatives do I have? I see mica caps at this value, which I assume would be fine for stopping AC (not AC, DC... duh). However, they may add more distortion to the reverb signal (if so, is it significant???).

Another issue is that I see lots of Cornell Dubilier caps and I know they bought out Illinois Capacitors. I buy from DigiKey and DK seem to be labelling caps made out of the IC factory as CD capacitors. So, now I have an aversion to purchasing any caps branded as CD. Did Cornell make decent caps before they bought IC and if so, how do I differentiate when DK does not? Are all CD capacitors made in the IC factory? 

Generally speaking Silver Mica or Ceramic Disc Caps are used for the small values like 270 pf.  Silver Mica is very hifi and the value is low enough that ceramic is fine too...I would only worry about the quality of the IC's electrolytics. I've never had any trouble with their film caps or ceramic caps in the multitude of amps I've serviced. The electrolytics are another story: those really are junk.

[Resistor001]

Thank you Makinrose,

After doing some more investigating, Cornell Dubillier do have a good name when it comes to making capacitors. I'm not sure if they ever made electrolytics, but for the time being I'll steer clear of electrolytics with their name as the producer, since there is a chance it may be an IC electrolytic.

[makinrose]

Thank you Makinrose,

After doing some more investigating, Cornell Dubillier do have a good name when it comes to making capacitors. I'm not sure if they ever made electrolytics, but for the time being I'll steer clear of electrolytics with their name as the producer, since there is a chance it may be an IC electrolytic.

Sure thing!! Glad to help.

[K O'Connor]

Hi Guys

CD has made elecrolytics for many decades and are well known for their quality. They tend to be a bit pricey.

Were I writing TUT3 today I would specify snap-mount main filters, as they are compact and have high values up to 600V. They mount on a clear space of the mounting card, with two holes drilled 10mm apart to accept the leads, which should be snug in the holes. On the underside, use buss wire to tie to eyelets, turrets, etc.

Many of the cap tone assessments - and indeed assessment of any component's tone - can be swamped by the effects of poor layout and wiring. With Galactic Grounding you will more readily hear the impact each component type has.

Going back to the first post of this thread: Voodoo parts are a waste of money unless it makes you feel good. Other brands of the same dielectric, for caps, will sound the same and cost less.

Have fun