Jump to content

Capacitor Rabbit Hole. Just How Do You Measure?


Dave A

Recommended Posts

A topic that has no end to it or so it seems. Reading some capacitor stuff tonight got me to thinking about how people claim to hear audible differences in using different caps. I have measured ESR and Capacitance but something I have never done is to take something like TrueRTA or REW and a calibrated mike and see it there are any measurable differences between caps in audio output. I do not want this to become a capacitor war over this one or that one is best. What I want to know is has anyone who really knows what they are doing tried measuring in this fashion. I have the tools but when it comes to this not the knowledge and I would like to know.

  • Like 1
Link to comment
Share on other sites

Anything wrong doing it this way ?

 

1. expose capacitor

First of all, completely remove the capacitor to be checked from the circuit. All contacts to the circuit must be removed and the two poles of the capacitor must be made freely accessible.

 

2. check capacitor optically

Before measuring the capacitance with the multimeter, the capacitor should be checked for damage. If the surface shows small dents, fine cracks or even leaking liquids, then this may indicate a defective capacitor.

 

3. discharge capacitor

The next step is to ensure that the capacitor is completely discharged. To get all residual current out of the capacitor, it can be connected to a load.  A light bulb, for example, is suitable for completely discharging the energy from the capacitor.

 

4. adjust the multimeter

Now the multimeter should be set to the function capacitance measurement (measured values in Farad). The measuring range is usually adjusted automatically by the instrument .

 

5. measure capacitance of capacitor with multimeter

Now both test leads can be connected to the poles of the capacitor. The display of the multimeter should now show a measured value that corresponds approximately to the value indicated on the capacitor. If both values are very similar, then the capacitor is in good condition. If the measured value determined is significantly lower than the value indicated on the capacitor or if no measured value is displayed at all, then the capacitor is defective and must be replaced.

 

Link to comment
Share on other sites

  I agree with both of you. Sonic differences between caps of similar construction is overstated in my mind. 

  One tip for checking the “sound” of a film cap is to perform the tests stated by MicroMara. Then thump them with a finger to hear the physical sound that comes from the capacitor. Teflon and polycarbonate film caps are higher pitch. Polyethylene, kapton, and polystyrene are lower pitch. PET, and other soft plastics lower still. 

    Electrolytic caps are considered a bad choice for signal coupling and filtering in crossovers. But have owned amplifiers that were AC coupled output with electrolytic caps. They sounded fine. OTL, single ended solid state, and single voltage rail solid state topologies are examples. 

Link to comment
Share on other sites

The trouble is that measurements and test with electrical equipment only tell part of the story. The ability to hear is far more complex than we can ever understand fully. If we, people, understood all there is to know about hearing there would be no deaf people. I have tired many different capacitors in audio gear and x-over networks and I can obviously hear differences in same quality capacitors. Capacitors are such a simple component, just foil separated by an insulator, you would not expect there to be a significant difference if quality control is much the same from one brand to another. I have found there is not a big difference. Just subtle at best. If one sounds too bright with your speakers it may be just to accurate and your high end needs to be rolled off till you find what you like. Multiple ways this can be done. Most technicians I know consider a cap is just a cap and buy what they like for the price they want to pay. Only when you get to audiophiles is the discussion of different capacitors is there a debate. Just like audiophiles debate the differences in speaker cable. I just consider 18 gauge copper good enough. So called audiophiles sometime just want to improve or buy something else in their system. Most visitors to this site just buy Klipsch speakers and enjoy what they have. 

Link to comment
Share on other sites

I have no problem measuring capacitors and use a B&K 885 capacitance/ESR/inductance meter which is more accurate than any Multimeter I have had. My current multimeter is a good Fluke but it is no where as precise as the B&K 885 I also have for crossover work. What many claim is there are differences in the audio output from the construction of the caps beyond just the ESR and capacitance. So what I am looking for is someone who has measured the output of a crossover using different caps by measuring the actual audio output. I want to know if they found any differences and how they did it.

 

  As far as measuring capacitors I know how and am equipped to do so better than most for capacitance and ESR. I keep running across the idea that a $500 capacitor sounds better than a 1% Dayton capacitor and both measure the same on my meter. People spend tons of money on this stuff and I want to know if there really is a difference besides golden ears and price tag bragging rights. I want to see measurable evidence and the idea advanced at times that science has not advanced far enough to measure the secret sauce in these fancy caps is, to me, preposterous. So either measuring a complete crossover and analyzing that or using actual audio output, like perhaps pink noise, and analyzing that is what I am really looking for. Someone who has done that and can say yes or no to caps make a difference.

 

  I am not sticking electrolytics into this by the way.

  • Like 2
Link to comment
Share on other sites

In the experiment I did with my EV's the Daytons and Audyns measured similar yet the difference was definitely audible to me and my daughter.  It was obvious to both of us.  I swapped them twice and never told her what I did.  she had the exact same impression of the sound that I did and it wasn't that subtle.  There has to be something that measuring capacitance and ESR doesn't show.  If you hear a difference in 2 components and they measure the same then you're not measuring the right thing.  I'm not smart enough to tell you what you should be measuring though.

  • Like 3
Link to comment
Share on other sites

41 minutes ago, henry4841 said:

The trouble is that measurements and test with electrical equipment only tell part of the story. The ability to hear is far more complex than we can ever understand fully. If we, people, understood all there is to know about hearing there would be no deaf people. I have tired many different capacitors in audio gear and x-over networks and I can obviously hear differences in same quality capacitors. Capacitors are such a simple component, just foil separated by an insulator, you would not expect there to be a significant difference if quality control is much the same from one brand to another. I have found there is not a big difference. Just subtle at best. If one sounds too bright with your speakers it may be just to accurate and your high end needs to be rolled off till you find what you like. Multiple ways this can be done. Most technicians I know consider a cap is just a cap and buy what they like for the price they want to pay. Only when you get to audiophiles is the discussion of different capacitors is there a debate. Just like audiophiles debate the differences in speaker cable. I just consider 18 gauge copper good enough. So called audiophiles sometime just want to improve or buy something else in their system. Most visitors to this site just buy Klipsch speakers and enjoy what they have. 

HF is a big problem and I end up using L-Pads more and more if I am not setting something up with active DSP. I expect HF is the far more egregious offender here.

  • Like 1
Link to comment
Share on other sites

It should be easy enough to measure the audible differences of different caps in crossover networks using a spectrum analyzer but then what do you do with that data? I think an interesting use would be to develop a standardized curve of what the ideal stock crossover for a given speaker should sound like and then use DSP EQ to recreate the correct sound signature just as the  original designers intended it to sound. You could also use it to emulate the sound of any capacitor or crossover you like. Too bad such data doesn't exist it could be valuable to a lot of people. Might be able to capitalize on an idea like that. 

Link to comment
Share on other sites

2 minutes ago, CECAA850 said:

In the experiment I did with my EV's the Daytons and Audyns measured similar yet the difference was definitely audible to me and my daughter.  It was obvious to both of us.  I swapped them twice and never told her what I did.  she had the exact same impression of the sound that I did and it wasn't that subtle.  There has to be something that measuring capacitance and ESR doesn't show.  If you hear a difference in 2 components and they measure the same then you're not measuring the right thing.  I'm not smart enough to tell you what you should be measuring though.

You are one of the people saying this that really has me interested in how to measure such a thing. Hopefully there will be a real answer to this in this thread.

Link to comment
Share on other sites

2 minutes ago, Dave A said:

You are one of the people saying this that really has me interested in how to measure such a thing. Hopefully there will be a real answer to this in this thread.

The Bonehead obviously believes this as well and I certainly am in no position to argue with him.  The caps he endorses has the "Klipsch" sound and others don't so there has to be an audible difference to him as well.

Link to comment
Share on other sites

4 minutes ago, wetowne said:

It should be easy enough to measure the audible differences of different caps in crossover networks using a spectrum analyzer but then what do you do with that data? I think an interesting use would be to develop a standardized curve of what the ideal stock crossover for a given speaker should sound like and then use DSP EQ to recreate the correct sound signature just as the  original designers intended it to sound. You could also use it to emulate the sound of any capacitor or crossover you like. Too bad such data doesn't exist it could be valuable to a lot of people. Might be able to capitalize on an idea like that. 

It's mostly an academic interest on my part over this topic since I use a Xilica now for my personal speakers or speaker builds. Finding and fixing up old speakers has gotten to hard to do so my interest in crossovers and capacitors is mainly just curiosity. For a topic that has so many die hard advocates there is a huge paucity of reliable data to go by.

Link to comment
Share on other sites

1 minute ago, jimjimbo said:

I count the number of forum  posts relative to a specific brand/value, and then make my purchase decisions from there.  Dave, surely I would have expected nothing less from you......

Right. So Audiogon or Klipsch?😁

 

3 minutes ago, CECAA850 said:

The Bonehead obviously believes this as well and I certainly am in no position to argue with him.  The caps he endorses has the "Klipsch" sound and others don't so there has to be an audible difference to him as well.

Roy said "It’s about the WHOLE network and voltage transfer curves" and he is considering many things at one time. I am mostly referring to tinkerers who make claims about components and I want to know. It's like the Audyn's you used over the Dayton's which I figure measured the same with the only method I have to measure by. The vast majority of the  caps claims are made for go into recapping of existing crossovers and are not part of a comprehensive balanced system plan like Roy does with his work.

Link to comment
Share on other sites

1 minute ago, Dave A said:

Roy said "It’s about the WHOLE network and voltage transfer curves" and he is considering many things at one time.

So different caps transfer voltages differently?  All you need to do is figure out how to measure that!

Link to comment
Share on other sites

2 minutes ago, CECAA850 said:

So different caps transfer voltages differently?  All you need to do is figure out how to measure that!

I got a couple of 535 crossovers with the last pile of gear I bought. It's a pc board that is somewhere around 10" x 12" + in size and LOADED with coils and caps and resistors. There is a couple dozen components on there. I am guessing here but I don't think the specific values of those new caps play more than a minor role in the output. It all adds up of course and enough incremental improvement make for a large over all improvement. The whole systems approach he talked about is clearly evident on this board. This is a complete different world than the one I am discussing with recapping old crossovers and claims made. The crossover you had for instance where you found the difference is primitive compared to this new stuff. Maybe there is a voltage transfer difference in the caps, I don't know. Hoping someone who does know will comment.

Link to comment
Share on other sites

13 minutes ago, CECAA850 said:

The Bonehead obviously believes this as well and I certainly am in no position to argue with him.  The caps he endorses has the "Klipsch" sound and others don't so there has to be an audible difference to him as well.

 

That post got me thinking as well, not sure how often he weighs in but don't think he would waste his time unless he felt it was important.

 

I wonder if that is more in reference to the overall culture of modding or as specific as Sonicap (as an example) do not sound the same as "Klipsch approved" capacitors.

 

Shocked that I just found out about "Klipsch approved" caps now.

Link to comment
Share on other sites

1 minute ago, CECAA850 said:

I believe that this is something new.

Curious why Klipsch doesn't sell approved recap kits, decent potential revenue stream if you can sell a $100 cap kit every 15-20 years.

 

Was watching a video by the Chief and he made a comment about 'going for the 20% other people don't' or something but it made me stop and think about how much testing and refinement goes into a heritage speaker until they say 'this sounds exactly the way we want it to.' And clearly they know what good sounds like.

 

I can also add a jet kit and new suspension to my motorcycle and the original designer might say 'that's not how we intended the bike to ride' but 99% of people would say it's faster and handles better, how is that a bad thing?

  • Like 2
Link to comment
Share on other sites

  • dtel locked this topic
Guest
This topic is now closed to further replies.
×
×
  • Create New...