Speaker crossover capacitor & inductors values: schematic & implementation?

Electrolytic caps have relatively high ESR and wide tolerances.  A good designer will calculate the ESR and the expected over (or under) capacitance into the crossover design. Any changes to ESR and capacitance may degrade the speaker performance. That said:

1) Perhaps two 47uF caps provide for the capacitance and ESR the designer is shooting for. After all, 47uF caps are more abundant in the market than 50uF caps and therefore more choices to zero in on a desired performance spec.

2) Film caps have less ESR and tighter tolerances than electrolytic caps... replace with caution for reasons above.

3) There are 11nh ferrite core SMD chips available (Murata) but why are you replacing it? The size does make sense because it provides a corner frequency in the low Ghz, thereby making it the first line of tweeter protection from HF energy. I don't know enough about crossover design to make a suggestion without knowing the self-resonant frequency of that inductor. The DCR should not be a factor since a well behaved amplifier will not pass DC to the speaker.

Thanks.
2) The main thing that concerns me is the change in ESR.  I've heard varying opinions on the extent of the issue.  It has been suggested that when switching to low ESR film caps, to add some compensating resistance in series at the driver.  But as you suggest knowing what the difference in ESR is, is important.  Measuring a 30 year old cap, may not provide the ESR value factored in to the design when the cap was new.

3) I am contemplating replacing the old 11nH inductors following the concept that ferrite core inductors are problematic, particularly in series with tweeters. This is what the 11nH inductor looks like:

It has been suggested that when switching to low ESR film caps, to add some compensating resistance in series at the driver

Well, the importance of this really depends.  If this is in series with the driver, and the difference is < 0.3 Ohms it’s probably OK, but you might still hear a difference exactly because you’ve changed the level of that driver relative to everything else.   Modders do this and go "OMG!! I’m a genius this sounds so much better..."  Well, it’s different because you’ve now altered the crossover.  It’s not necessarily that your $150 cap is so much better.

When ESR or DCR (for coils) really matters is when they are parallel (go to ground) to the driver.  In some cases you may even see small value resistors in series.  Those very small values should be a clue that the total R is a critical value.   Very small total R values in this leg can really throw off the entire speaker’s impedance curve.   When I pick coils and caps to use in a speaker I take these DCR and ESR values into account.  They aren't an accident.  I did not randomly pick a small gauge coil because I could not afford bigger, so assume the original designers did the same. 

These situations are where doing a complete electrical analysis  and using a crossover simulation like XSim or VituixCAD can really help you evaluate what you are doing. 

This is a discussion best held in DIYaudio where a lot of builders and experts hang out. 

Let me add a little more... if I have a choice between a smaller coil with more DCR or a big coil and separate R I'll pick the smaller coil because coils handle high power better. 

BTW, if you are doing this, get something like the Dayton Audio DATS.  It's immensely useful for speaker diagnosis, part measurement and creating accurate schematics.   

Thanks, Erik.

Will repost this in DIYaudio and get a Dayton Audio DATS.   I appreciate the help.

If you do replace that inductor with an air coil Jantzen, it will be the equivalent of a longer length of wire from the terminal to the first node of the tweeter crossover. That may or may not have an effect on HF roll off. When I spoke with a tech at Thiel Audio years ago about replacing a tweeter and mentioned snipping the wires, he said to de-solder so as not to change the length of the wires. Another thing to consider is that a ferrite core does an excellent job of absorbing EMI whereas an air coil is an EMI antenna. That may be a tradeoff that negates the better linearity of an air coil. 

Another thing to consider is that a ferrite core does an excellent job of absorbing EMI whereas an air coil is an EMI antenna. That may be a tradeoff that negates the better linearity of an air coil. 

Interesting.  I thought air core inductors were better at EMI rejection and do not t produce significant EMI like a magnetic core might.
Of course, the discussion might be moot as Jantzen, Solen, Dayton, etc. don't make inductors in such small values.

"47s" are used because they are a much more common value to obtain = <$$$.