test crossovers or capacitors with frequency sweep

OP:

In this case your best bet is going to be to do close mic driver to driver measurements. Use a 60Hz reference tone and set output to 2.8V. Put your measurement mic at 1/4" in front of a driver, and measure. Repeat for each driver.  For cone woofers/mids measure the mic distance in front of the dust cap.  Use a plastic ruler to make sure you don't have the magnet pull it through the cone. :)

This part is relatively simple, assuming you don’t move your volume control.

Your drivers should match up. The complicated thing is putting it together into whole speaker measurements. Too much go go into hear.

Two places to go for help is the Room EQ Wizard forums as well as DIYaudio.

@rodman99999 - I don't know how many have actual anechoic chambers.  Close-microphone, quasi-anechoic measurements are a lot more convenient.  For the definitive resource see Joseph D'Appolito's book, "Testing Loudspeakers"

https://amzn.to/3xxf6ml

Thanks to you all for your advice. I will try this simply for the fun of it, as a mic is not that expensive. I guess an important conclusion from what you are saying is that capacitor changes that can be heard and matter will not make a big difference in the response curve, or will be smaller than the variance in any case. I will compare near and far from the speaker to minimize effect of the room. If I am really fired up, might take the system outside :)

Unless you are very experienced with using measurement mics this is a bad idea. 

The variance in results will just drive you nuts.

                                                     +1

     Typically: manufacturers curve their systems, using closely calibrated mics and equipment, in an anechoic chamber, which most (normal) folks don't have in the home.

Yep, the DATS is the way to go. 

thanks- my main issue is that it is really hard to get to the crossover for one of the speakers that I also want to work on (B and W 803 matrix series 2).  

Unless you are very experienced with using measurement mics this is a bad idea.  The variance in results will just drive you nuts.

The best way to do this is to use an impedance tester. Dayton Audio DATS is a great unit, even seen it used at a major high end manufacturer. Around $200 I think. You could also roll your own with Room EQ Wizard.