Modifying Crossovers

A quick story:

I was a audio dealer, speaker designer, and still currently involved in audio service and performance upgrades. Years ago I attended a convention and had a heart-to-heart with one of the most successful speaker manufactures of the past 30 years, He had recently introduced their new "flagship" model to rave reviews, and was an overnighht success. He was polite enough to give this guy some of his time, and we got deep into speaker "nurdystuff". After sharing my thoughts with him, he paused for a few seconds, looked me in the eyeballs and said: "I believe in all that.. I just don’t want to fight my engineers over it."

A few weeks later I decided to do a "what if?" experiment and did a "high end" top to bottom approach to the speaker, addressing all the things that stood out to me that could use a little "TLC." A month later, the regional manager for the company visited my store and I convinced him to spend some quality time with the OEM vs "modded" versions of the speaker. He sit there for a while listening, and I could tell the gears were churning. After the music stopped, I asked him how he liked them. His answer: "They sould better. Okay, what did you do?" My answer: "Everything your engineers told you wouldn’t make a difference." I think this may have perculated to the top.

I agree with @erik_squires that I wouldn’t mess with a high end speaker -- UNLESS they can be reverted to origninal condition. a) you might not like the changes, b) it may effect the resale value, and c) you might have just voided whatever remains of the factory warranty (on the raw drivers).

OEMs choose crossover designs based on a number of factors. Cost, manufacturing efficiencies, ease of service in the field, etc. But, there also has to be "buy in" to the design and components by all iinfluencers of the speaker design/manufacturing process. This is where philosophies (sonics, sonics ,engineering, financial, etc) enter in and present a very complex situation.

I have a very simple philosophy as I approach speaker "mods". Get things out of the way that make the speakers sound worse. These are usually not hard to find in speakers under $25k.

A quick note on "better" parts. If the values of the replacement parts are the same as original, the measured frequency response should be within 1db (non detectable) of the original. However, when the music starts playing, things become different in a hurry -- in a good way. Dynamic compresion is one of those stock OEM attributes that doesn’t sound "wrong" when you hear it. Until you realize how much headroom was compromised and micodynamics were squashed when the speaker is upgraded by better internals. This includes modern AND vintage designs.

Just my 2 sense.

@carlsbad2 

Lookin’ good Jerry.  Let us know how they sound.

My cautionary tale is to be careful in doing too much.  Tweeter and mid-cap upgrades are simple and easy. 

If you have a 1980's - 1990's vintage speaker though, it's worth doing a complete analysis before doing much work.  The issue is the tools and effort it took to get a good crossover back then to work across both frequency and impedance.  There's plenty of examples of speakers in this vintage that need different crossover designs. 

Be extremely careful replacing caps that go to ground, their effects on the impedance curve is unexpected, so measure and compensate for ESR changes. 

Best,

E

Thanks everyone for this very interesting and informative discussion. Let’s close this out. 

Cheers

This used to be a VERY sore point for me. I got so tired of seeing SO Called High End Brand Names using something similar to Cat5 wire and a single off the shelf Radio Shack quality capacitor inside. I sincerely hope that this is a thing of the past. I might say that I do believe that most of the really quality speaker manufacturers have real science backing them up when they design their crossovers. I still try to have faith in the educated human mind.
I must remind the folks that there are so many components to chose from and different ideas behind how they go about creating a particular crossover. A Particular driver may have some inherent oddity that requires something that you are not aware of and can’t predict.
Also lets just look at one component, the capacitor. And even at that, One Brand, MUNDORF. They make so many different caps for any given value that it would be insurmountable for the average user to pick one and say that just because it is a Mundorf; it will give my speaker a better sound. You might hit on the right one but if you miss yo just wasted a bunch of spare change and time. and at the same time probably gained a slanted opinion of Mundorf when they really don’t disserve it.
Looking at Inductors, I would love to brag about continuous ribbon wound inductors over wire wound inductors but even then this could blow up in my fact just because I don’t know something about the coil in the speaker’s driver that would necessitate the wire wound over the ribbon wound.
And just to show you that I can screw up a speaker as well as the next guy. I took one of my own designs where I originally used Chartreuse caps in and always liked them, but thought I could do better so I spent a couple hundred dollars on some Mundorf caps. I have regretted that decision every day since.