when to use high grade caps

Hi

I'm about to build some high quality speakers and have to buy capacitors for the crossovers. The question that rises is which ones will give a good result. There are some realy expensive capacitors out there and I wonder if these are worth the money. I read some threads on this forum, and I see that the opinions vary. Some claim you might just as well buy the cheapest ones around and others are willing to spend hundreds of $$$ on a crossover. I have no clue what to choose.

I have the following questions.
- Under which conditions do high grade capacitors matter?
- I want to build a three way speaker. Which crossover parts (tweeter, mid , bass) benefit from better caps.
- If there is a resistor or a coil in series with the capacitor does it still matter?
- which ones have a good price quality ratio
- which commercial speaker brands use what type. (from what I've seen, they use entry level "audiophile" capacitors like mundorf MCap or just elco's)
 
Last edited:

SY

diyAudio Moderator Emeritus
2002-10-24 10:19 pm
Chicagoland
www.SYclotron.com
I'll come out in the middle. You want a high quality cap with low microphonics, ESR commensurate with the crossover requirements, and relatively low inductance. That means encapsulated, machine-wound caps, high tension windings, with polypropylene being the preferred dielectric. Metal foil would be preferable to metallized.

Avoid any cap labelled "audio" or "audiophile" or any cap boasting of exotic materials- that usually degrades performance, which makes some audiophiles happy because of the added distortion and microphonics.

Stick with well-engineered mass-produced brands. Wima, Vishay, Panasonic, that sort of thing.
 

opensource

Member
2005-08-23 11:46 pm
Hello.

About which capacitor could have a better sounding than another: try this well known site http://www.humblehomemadehifi.com/Cap.html
You should put a better quality capacitor where the "signal" goes through. Typically (but depends on your filter) on the highs and some of the mids. Standard quality should ne OK for the others.
I put Obligato as "high quality" in my filter, MCap as standard quality. Also: bass are a little more punchy with high quality air coils (lower resistance is better). True for coils on the signal path (bass and maybe on the mids).

Philippe
 
Hello.

About which capacitor could have a better sounding than another: try this well known site Humble Homemade Hifi
You should put a better quality capacitor where the "signal" goes through. Typically (but depends on your filter) on the highs and some of the mids. Standard quality should ne OK for the others.
Philippe

The article you mention starts with the equivalent circuit diagram of a capcitor. There is a resistor and a coil in series. This is the reason why I wonder if there is anything to gain (if there is any to begin with) with an expensive capcitor if the series resistance/inductance is far lower than the parasitic components "inside the capcitor".

In my understanding of electronics all parts in the filter are part of the signal path. Whether these are in series of in parallel is not relevant. Or
do I not I not understand you correctly?
 

SY

diyAudio Moderator Emeritus
2002-10-24 10:19 pm
Chicagoland
www.SYclotron.com
This is the reason why I wonder if there is anything to gain (if there is any to begin with) with an expensive capcitor if the series resistance/inductance is far lower than the parasitic components "inside the capcitor".

It won't be. If you get a good, name-brand non-audiophile foil/film cap, the inductance will be negligible for this application. The linked site is... imaginative, lots of story-telling to market their services.

edit: I notice they also sell "audiophile" capacitors as well as a variety of audio placebos. Add several large grains of salt to anything you read there.
 

opensource

Member
2005-08-23 11:46 pm
In my understanding of electronics all parts in the filter are part of the signal path. Whether these are in series of in parallel is not relevant. Or
do I not I not understand you correctly?
I was meaning components in series. Beware: I didn't told you that you'll have to buy some Duelund capacitors to build good speakers. Also: you'll not transform "standard" speakers into high quality one with just some good components. You only may expect to have a slight/moderate improvements (more "punchy" bass with low resistance air coils on bass, some better definition with the highs and no idea for the mids. I only built a pair of 2-way floorstanding speaker) so don't spend too much with components in your crossover.
10%/15% of the total amount of your speakers+cabinet would be reasonable.

SY said:
The linked site is... imaginative, lots of story-telling to market their services.
Ok. Your point of view. The "capacitor test page" is an old page. There wasn't shop on the web site some years ago and I agree it's not a good idea to mix "tests" and "shopping".
Guess what you may find in the crossover of some B&W speakers? Mundorf capacitors...
600 Series | Technologies - Bowers & Wilkins | B&W Speakers
 
I'll come out in the middle. You want a high quality cap with low microphonics, ESR commensurate with the crossover requirements, and relatively low inductance. That means encapsulated, machine-wound caps, high tension windings, with polypropylene being the preferred dielectric. Metal foil would be preferable to metallized.

Avoid any cap labelled "audio" or "audiophile" or any cap boasting of exotic materials- that usually degrades performance, which makes some audiophiles happy because of the added distortion and microphonics.

Stick with well-engineered mass-produced brands. Wima, Vishay, Panasonic, that sort of thing.
I am sorry but "low microphonics" in a crossover, that has no gain?
Low microphonics should be used in any sort of amplifier that is subject to vibration. Not a loudspeaker crossover.
 

SY

diyAudio Moderator Emeritus
2002-10-24 10:19 pm
Chicagoland
www.SYclotron.com
I am sorry but "low microphonics" in a crossover, that has no gain?

Yes. Do you want the transfer function of the filter to wiggle around with the acoustic output? We call that effect "distortion." Since crossovers are generally inside of a cabinet, it becomes double-plus-ungood because of the very high SPLs to which the cap is subjected.
 
I wouldn't build speakers with a passive crossover anymore. I'd either build using a quality FR driver or multi-amp with an active crossover.

I won't go into the arguments as to why this is preferable, but it's arguable that it is.

That's not what you asked, but I think it's in context.

Electronic crossovers are the choice of us professionals. I have worked with PA systems and home audio systems and passive crossovers give rise to vibration issues. In that, they tend to fall to bits. Never ever have I in 45years had a microphonic issue, just capacitors shedding their wires because the builder didn't allow for vibration.
 
dre,
Sy is correct about just using a name brand capacitor and forget the audiophile nonsense. I actually know a capacitor manufacturer and I have seen more than one audiophile capacitor that is nothing more than a standard cap with a fancy cover. What I have done with passive crossovers in the past was to use either metalized film or foil/film polypropylene caps. You could use polystyrene but they are not quite a good as the polypro. Don't try and find high value capacitors, use multiple caps of lessor value and add them together to get the values you want. This has the advantage of being able to use lower cost capacitors and hand select the caps that give you the final values you are after. A bunch of 100mfd caps are much cheaper than say one 300mfd cap. Most caps are going to be 10% values and the multicap approach allows you to get values much closer. I usually get within 1% by using multiple caps and I usually also have some very low value caps to help get the numbers this close. Another thing to look for is the voltage rating. I use all 100volt capacitors unless they may be 250volt but that is just because I got some of those cheap, they are a little larger but they are better than many of the 25volt caps in a passive crossover.

As for air-core coils I only use these not any with ferrite cores. But though you will hear about the magic of air core dimensional ratios there is another method that you can use. You can if you wind your own coils use a solenoid winding instead. Take a small wooden dowel and wind the coil on the wooden dowel. You can leave the dowel in place and use it to help mount the coil if you leave a portion of the dowel sticking out. The wood is non inductive and can not saturate. Use a large gauge wire like 16 gauge wire. Because the diameter of the wood dowel can be as small as 1/4 inch the actual length of wire necessary to wind the coil will be less as the inside coils will have a small diameter and this takes much less length. This will also lower the Dcr value of the inductor as the wire length will be much less.

Just my 2cents worth of knowledge.
 
That is an amusing thread. If a capacitor is linear and well-behaved then a DC bias does nothing. If a capacitor is non-linear (all caps are slightly non-linear, but usually far too small to notice) then the main effect of a DC bias is to change a little odd-order distortion into a greater amount of even-order distortion (plus probably some remaining odd-order). If this change is audible then the solution is to use a better dielectric (more linear) or better cap construction (e.g. tighter winding, which is normal for ordinary commercial caps anyway).

If there is a zero-crossing issue with current, as some claim, then DC bias does not change this at all - the AC current still goes through zero as before. It would change a voltage zero-crossing issue, but there is no evidence for this as far as I am aware. Dielectrics don't do anything strange at zero electric field.

Sorry to go a bit OT, but it is surprising what some people believe.
 
DF96,
This is no different than those who truly believe in the boutique capacitors. As I have stated here and in other threads I have actually seen some of the high end brand names wound on the exact same machines with the same films as the standard capacitors. The only differences were as simple as a nice pretty wrapper and at best I saw some that used a braided lead-out wire instead of a solid lead. They looked pretty but the electrical properties were identical. The only thing that some of these manufacturers had to pay for was the different film wrapper cover and a simple pad printer artwork change. It was a fraud to say the least, but it wasn't the actual capacitor manufacturer making the claims just the sellers of the parts.
 
You could argue that morally the manufacturer is an accessory to the fraud, but I guess most just turn a blind eye to it. Legally the blame rests with whoever makes the silly claims. However, if they are sufficiently extravagant that a normal sensible person would know that they are false then a court might find that they were merely 'advertising puffery' and not fraudulent! If you have to tell a lie, make it a big one.