Replace a capacitor when the value is still good?

[arken]

Is there any reason to change a capacitor when it's value it still within 10%. 

[RickB]

Well couple  of questions. What kind of cap and how and what did you use to test them?

[RTally]

There are two parameters of importance for capacitors. The obvious one is capacitance value. The other is equivalent series resistance (ESR). A change in capacitance value will impact the crossover frequency. For a first order filter, it takes a large capacitance value change to create an audible null or peak in the frequency response. 

A change in ESR will attenuate the tweeter volume for a high pass filter. Generally, changes in ESR will be more audible than capacitance value changes. Too high of an ESR causes the speaker to sound muffled, as if a blanket is thrown over the speaker. If ESR is to be tested, the test meter must use a frequency in the audio range for the test to be meaningful.

To answer your question, it is possible to have a cap that test within spec for capacitance value but out of spec for ESR. Such a cap needs to be replaced.

To test a cap, it needs to be removed from the circuit first. At that point, why waste time testing except for intellectual curiosity. Once the cap is removed from the circuit, you might just as well put in a new one.

If you want to test, a better test is done with a real time analyzer (RTA) to check the FR of the speaker. I play noise and use a phone app to check the FR of my speakers. For simple checks, the phone app is fine, but a proper RTA and mic would be best. Using a phone or RTA is a simple, objective way to determine if there are any unusual nulls or peaks or attenuation. It also allows for easy verification of operation of any switches or pots in the crossover circuit.

[DavidR]

For NPE caps YES. I recapped my AR90s that had been sounding dull. Most of the upper range caps tested well within the +/- 10%. The ESR was quite high. The change was remarkable.

[arken]

 The reason I asked is because I pulled a few caps from an ar11 that were still within 10%. . I have never checked the ESR but I am going to give it a try. There is so many variables to capacitors , what type , size , brand. It can be a bit confusing especially given that people don't even agree on what type to use in any given application. Though it sounds like changing them is a pretty safe bet. RTally is right if your gonna take the time to pull it out so it can be properly tested you might as well just change it. Thanks for all the great information!!

[DavidR]

To help you with the capacitor selection I used NPE on my 10Pi and all film on my 91 speakers. There is very little difference if any at all. Perhaps a tad more detail with the film caps.

However, on the 10Pi series caps (tweeter 10Uf and mid 40uF) I used a Vishay 0.01uF film cap in parallel. The 10Pi has Mundorf ECap and Bennic.

The 91's have Clarity CSA 3.9uF with an AudioCap Theta 0.1uF bypass cap and on the midrange a Clarity 25uF CSA with a 0.01uF Cornell-Dublier 940C bypass cap. All the shunt caps are Dayton.