I see very often that even very expensive speakers do not have a flat frequency response and really bad lobing. Also i see in this speakers very expensive parts.
When i talk to the designers they have of cause chosen the parts by listening.
I think that this is simply bad design.
Ones the fundamental balance of the speaker is right it makes sense to use good parts.
I do that too. To put the sound of caps on an absolute scale is dangerous though.
When i talk to the designers they have of cause chosen the parts by listening.
I think that this is simply bad design.
Ones the fundamental balance of the speaker is right it makes sense to use good parts.
I do that too. To put the sound of caps on an absolute scale is dangerous though.
ESR of a capacitor is particularly important with bass filter capacitor shunts, since a tiny resistance has a big effect. Recall those designs that add a 0.5R resistor to the shunt arm like Klipsch designs. In the SEAS Embla kit example below, with so much resistance in both capacitor arms, I think you could use even cheapie non-polars to good effect. Non-polars have an ESR around 0.5 ohm. 
Attachments
Last edited:
I'm interested in how the esr conclusion was made , was it by measurement , controlled listening or was it conjecture , how was it determined an el cheapo electrolytic sounded better than a PP...
If it wasnt a controlled test as outlined in Linear audio , then all kind of brain farts could have happened....
Just saying
If it wasnt a controlled test as outlined in Linear audio , then all kind of brain farts could have happened....
Just saying
Last edited:
Nobody doubts that non-polars are imperfect components, particularly since the wet electrolyte dries out over 20 years. Electrolytic is also significantly non-linear with voltage, since they incorporate two polar electrolytic capacitors back to back. But as mentioned, they can work well.
The thing to understand about capacitor ESR is that it IS often significant in crossover design. If you replace non-polars with better polypropylene or polyester, you should take the changed ESR and consequent tonal level changes into account. Often this is done by adding a 0.5R wirewound in series to the film type. That's all.
I can certainly hear the difference between non-polars and plastic film types, it's a small but a significant improvement in detail and a reduction in sibilance, but I'm less certain about the difference between 250V and 630V types.
Given how expensive and big exotic crossover components are, you certainly have to balance that against spending more on better drivers. Good engineering is all about optimising everything to match quality.
What's the point of a $50 capacitor with a $5 cone tweeter? That's not good engineering.
The thing to understand about capacitor ESR is that it IS often significant in crossover design. If you replace non-polars with better polypropylene or polyester, you should take the changed ESR and consequent tonal level changes into account. Often this is done by adding a 0.5R wirewound in series to the film type. That's all.
I can certainly hear the difference between non-polars and plastic film types, it's a small but a significant improvement in detail and a reduction in sibilance, but I'm less certain about the difference between 250V and 630V types.
Given how expensive and big exotic crossover components are, you certainly have to balance that against spending more on better drivers. Good engineering is all about optimising everything to match quality.
What's the point of a $50 capacitor with a $5 cone tweeter? That's not good engineering.
Last edited:
The different parameters of the different types will lead to different performance.
That is why it is important to select the correct type for the intended duty. The parameters matter.
Select an inappropriate type and the parameters will affect performance.
Some, using their ears, may be able to hear this. Part of that listening group will like the change, others in that listening group will not like the change.
The before and after types may both be wrongly selected. Both sets of "test" listeners may be listening to a crippled implementation.
That is why it is important to select the correct type for the intended duty. The parameters matter.
Select an inappropriate type and the parameters will affect performance.
Some, using their ears, may be able to hear this. Part of that listening group will like the change, others in that listening group will not like the change.
The before and after types may both be wrongly selected. Both sets of "test" listeners may be listening to a crippled implementation.
7,
Have you ever measured them, if those are the resistors you are using no wonder you cant hear the the difference the caps are making ...
Oh right, I can't hear acoustic differences with boutique capacitors because I'm not using boutique resistors!
This is some sort of annulment of the laws of physics and the science of engineering IMO. Occam's razor says that if you can't measure an effect, it doesn't exist.
Take my advice, and put more effort into making an easy amplifier load, especially at the top end. Also think about what source impedance the drivers are seeing, and hence how they are electrically damped. You CAN hear that.
Last edited:
Is the assertion of this thread that different caps do not have different sounds or that there is no correlation between brand/price and perceived performance?
It makes perfect sense that a nonlinear driver has a greater effect on the performance of a system, but does this negate the possibility that a cap can alter the sound of non linear driver...or circuit.
Arguing that cap A is not worth X is reasonable and understandable, but is it being disingenuous to say that there is no difference?
It makes perfect sense that a nonlinear driver has a greater effect on the performance of a system, but does this negate the possibility that a cap can alter the sound of non linear driver...or circuit.
Arguing that cap A is not worth X is reasonable and understandable, but is it being disingenuous to say that there is no difference?
It does not look to me as Mr.Wayne would have any trouble
with his amp handling high frequency impedance or
similar, but what do I know, I'm just making wild guesses.
My Rotel amp is a bit of a cheapie, and is doubtless barely good enough. It is sensitive to load. But I see that as a challenge rather than a problem. It's also about bang for buck.
I'm sorry that I seem to be hogging the thread a bit today, but I see no real way that a film capacitor introduces significant non-linearity into a filter.
Different grades may have different lossiness and brightness, but you'd fix that with the attenuation in the crossover.
My own gut feeling is that there are far more important design considerations with other aspects. Amplifiers and drivers and cabinets are far more imperfect.
You should make an effort where it counts, and capacitors pretty much work as advertised IMO.
I'm sorry that I seem to be hogging the thread a bit today, but I see no real way that a film capacitor introduces significant non-linearity into a filter.
Different grades may have different lossiness and brightness, but you'd fix that with the attenuation in the crossover.
My own gut feeling is that there are far more important design considerations with other aspects. Amplifiers and drivers and cabinets are far more imperfect.
You should make an effort where it counts, and capacitors pretty much work as advertised IMO.
I think this is correct. First, it is necessary to get the technical data as good as possible, then determine if it makes sense to see how quality components effect the listening experience. Once it is determined better parts make a difference, then use that as reference to see if a lower cost solution is possible to achieve the same experience or near experience to evaluat cost effectiveness. Of course, if cost is not an issue, then just keep the expensive components in.
Generally, you can model the effect of that.ESR of a capacitor is particularly important with bass filter capacitor shunts, since a tiny resistance has a big effect. Recall those designs that add a 0.5R resistor to the shunt arm like Klipsch designs. In the SEAS Embla kit example below, with so much resistance in both capacitor arms, I think you could use even cheapie non-polars to good effect. Non-polars have an ESR around 0.5 ohm.
Expensive caps have their place and that is in the products of the
same price range. If one wants to build a high priced speaker for
sale, he needs to do everything targeted group of buyers appreciates
and these people like it expensive,shiny, bulky... Mundorf Supreme
sounds better than Mundorf Classic. Mentioning silver or gold
can only help.
The same applies to DIYers. The ones having no economic issues
will appreciate somewhat differently than the ones always on the
budget.
I can't agree.
Once it is determined that different parameters of the parts make a difference, determine which parameters need optimising, in which direction at each of the locations of particular components.
Price and expensive have nothing to do with selecting and optimising the component parameters.
Ooh, Andrew, some selective quoting there! You just broke the rules of honest journalism!
Lojzek said that the high-end brigade would sniff at anything less than Gold plated and Diamond dusted. Even if it might all be irrelevant.
Trouble with component upgrades is it is quite endless. Should I replace all the electrolytics and opamps in my Rotel RA-931? Perhaps I need to do some work on my RCD-965BX Special Edition CD player, which was known for its harsh midrange?
Oh, I won't sleep tonight at all...
Oh wait. It sounds pretty good really.
Lojzek said that the high-end brigade would sniff at anything less than Gold plated and Diamond dusted. Even if it might all be irrelevant.
Trouble with component upgrades is it is quite endless. Should I replace all the electrolytics and opamps in my Rotel RA-931? Perhaps I need to do some work on my RCD-965BX Special Edition CD player, which was known for its harsh midrange?
An externally hosted image should be here but it was not working when we last tested it.
Oh, I won't sleep tonight at all...
Oh wait. It sounds pretty good really.
Last edited:
- Status
- Not open for further replies.