Three way crossover question

diyAudio Moderator
Joined 2008
Paid Member
Yes if you want.

Sometimes when you make consolidations like this you can affect the sound just slightly. I'd expect to try something like 10uF and then trim the value by adding a small extra capacitor, or the other way. It's quite common to use a base value and a trimmer component.
 
Last edited:
I'm not familiar with the term bypassing.The XO mid range filter will just have the two caps in parallel.
So I’ve been playing with a bunch of Dayton precision caps and now have three groups of caps in parallel, a Jantzen superior Z 5.6, a combination of Dayton precision caps = 5.2 and another group of Dayton caps = 5.2 The sound has never been better. It seems the louder upper midrange was masking much of the tweeter. The tweeter is now is much more revealing of cymbals, ect.. I have a few 0.47 and 0.22 caps I’m going to play with for a final tweaking. This is fun.
 
Last edited:
  • Like
Reactions: 1 user
So now, considering the system as a whole, including the room, I think the sound is as good as it can get. The 3 parallel groups of caps are: 1. Jantzen Superior Z 5.6, grp2 Dayton precision 2.7, 1.5, 1.0, grp3 Dayton precision 27, 1.5, 1.0 and 2 0.47s. My question is will replacing 3 caps in grp2 a with 5.2 and 5 caps in grp3 with a 6.14 potentially make a difference. I realize a 6.14 probably doesn’t exist so I would use something close or possibly two caps.
 
I must have OCD. So now, after listening for two days, I wanted to very, very, very slightly reduce the upper midrange just a smidge. So I added my last two Dayton Audio precision 0.22 caps, one to grp2 and one to grp3 and I like it. And again would replacing multiple caps, in grp2 and grp3, with one cap, in each grp, have any effect on the sound? Now I need a sedative.
 
I've been listening to one channel at a time because the ten caps, comprising grp2 and grp3, are in one XO. So I connect that XO to the right speaker, then connect it to the left and repeat, repeat... The total of 10 caps = 11.8uF so I plan to buy two 6.8uF and two 5.1uF caps. I think
+0.1 is close enough. Thoughts?
 
I just finished the day listening to Keith Jarrrett, Charles Lloyd, Miles Davis, Tiger Okoshi and Mahler's 5th. And 17.4uF in the midrange filter, split over three caps in parallel, is the magic for me. and I aslo listened to Django Reinhardt, just to hear if guitar sounded right. It did. And I may have mentioned, in an earlier post, that I'm a classically trained flute player and have played in jazz clubes in NYNY. My ears are as important as my techincal skill.
 
Last edited:
  • Like
Reactions: 1 user
So the 10 caps, with a combined value of 11.8uF, were replaced with two caps, with a combined value of 11.9uF, thinking the sound would be the same. That didn’t happen and the upper piano notes were again overly predominate. So I added a 2.7uF and 1.5uF, and now the overly predominate upper piano notes are back to where I like them. Is this typical?

P6 mid range filter.jpg
 
diyAudio Moderator
Joined 2008
Paid Member
I'm guessing, but it's possible your new combination has a different parasitic resistance. Sometimes a small change in resistance has a very similar effect to a small change in capacitance, in the places where it matters. So, yes it is plausible.
 
One plus is the tweeter is revealing more detail. Now looking back, what surprises me is that about ten years ago, when I didn't know what I was doing, it took about three iterartions to complete my first outboard XO and to this day I woulldn't change anything. Now, after at least ten iterations on my second XO, I think I may be very close.
 
One plus is the tweeter is revealing more detail.
Henry,

That is one way to look at attenuating the bass response with the upper woofer's 4R7 resistor ;).
The added resistor also changes the shape of the 2.5 way acoustic crossover, as well as the addition of the capacitors (an additional 90 degree delay) has also changed the phase response between the upper woofer and tweeter.

Normally, one would change the value of the series coil (1.5mH) when going from first to second order, though without measurement, hard to tell what the acoustic crossover has ended up as.
The Haas precedence effect does make the leading tweeter sound louder...

Reversing the polarity of the tweeter to hear whether amplitude response in the acoustic crossover region increases or decreases would be another iteration to pursue if it wasn't in the first ten. Of course, a polarity reverse may also require a different capacitor value..

Anyway, you are having fun!

Art
 
Yesterday the sound was very good. Today the left channel tweeter is missing a lot of cymbal and the right channel tweeter sounds great. So I tried all the obvious tests. I reversed outputs on the amplifier, no change and swapped XOs and no change. Maybe the ferro fluid in the left channel tweeter has dried up. Any ideas?
 
Henry,

Dried up ferrofluid can cause a tweeter to play louder, though gummed up (semi-evaporated hydrocarbon carrier oils) ferrofluid can cause it to be quieter.
The changes usually occur so slowly that one becomes adjusted to it, unless listening closely for certain specific ranges:
"So I connect that XO to the right speaker, then connect it to the left and repeat, repeat..."

It is possible to clean out the ferrofluid completely on most tweeters, then either refill, or go without.
Going without would reduce power handling slightly, probably not a problem for you unless you decide to start listening to Metallica's "Death Magnetic"..

It is common to find "brand new" replacement ferrofluid filled tweeters not matching each other, or ones of an older vintage.

I used to have eight stage monitors with ferrofluid filled tweeters, no two were identical, and deviations between the extremes were as much as 6dB.
I numbered them so the closest matches were always paired, then equalized each pair separately.
More recently, a pair of ferrofluid filled Celestion CDX14-3050, also with quite different response between the pair.

Used to change the oil in my vehicles myself, now I take them to be serviced.
I'm also not a fan of ferrofluid unless it is properly and regularly maintained by someone other than me ;)

Cheers,
Art