Understanding a couple crossover components...

Recently built a DIY speaker kit.....my 2nd after putting together a different one nearly a decade ago.....So far I'm not a huge fan of how it's voiced and how it sounds in my particular application. (I liked my Classix II better......) I've found measurements for this kit that confirmed my suspicions of what I was hearing.....
So I'm looking at altering/experimenting with the crossover, but I'm new to this, and learning about crossovers.... Been reading through some of the massive threads on these, and trying to understand the basics of how they work, etc.

So, I've drawn up the supplied crossover for this kit...... I can understand that that L2 and C3 are a filter for the woofer, basically...and that C1 and L1 do the same for the tweeter....and R1 is probably there to bring the tweeter down to the desired level... But I can't figure out the purpose of the 2nd capacitor in series after the first capacitor/inductor combo on the tweeter....What would this do? And the purpose of the R2 parallel resistor...I assume for impedance matching? Or do the 2nd capacitor and parallel resistor somehow work in combo? Or is there something else I'm not understanding?

Thanks for any insight!



  • Stock-Xover.jpg
    108.7 KB · Views: 13
This way you might tell which problems are simply due to voicing and which are actual crossover problems.

What's the difference? As in, how is the speaker voiced outside of the crossover? Are you referring to the enclosure, the raw drivers, and how they interact separately from the crossover?

Certainly something I've considered trying before I actually mess with the crossover...as I want to also help confirm how MUCH I want to change things...and a test EQ setup would allow me to try before I "buy." :)
Joined 2008
Paid Member
Sure, I understand there is a grey area between the crossover for crossing, and the crossover for voicing. The difference is that unless you take control of one, you could mess it up when doing the other one. Using an equaliser keeps them separate, at least for now and at least until you properly measure or simulate as needed.
  • Like
Reactions: 1 user
C1, C2 & L1 form a 3rd order high pass filter. R1 and R2 form an Lpad

Ahhh, so C2 is working with C1 and L1 as a filter.... I guess I had only looked at 2nd order stuff......thinking that C2 was separate...Now I google 3rd order high pass and see an example like this:
(from https://www.the12volt.com/caraudio/cross18db.asp#hp ) Didn't realize that's what a 3rd order would look like!

Got it, now I can study up on 3rd order Xovers (and understand L-pads a bit better) and maybe further try and understand what the designer was trying to accomplish. Thanks!
The kit was a set of S2000s.
What I was hearing was : 1. boomy-ness caused by rear firing ports being inside a cabinet/box and close to a wall (plugged the ports for now to solve that...might make new enclosures to move the ports to the front or bottom of the phonograph cabinet I'm using, and evenutally will have a sub anyway...)
2.: A hollow sound, like a lack of midrange..... Or too much roll off above bass frequencies for my ears... (Some of this might be my room, but again, heard some of this in my garage upon initial listening to confirm function.)
3. definite lack of highs... they were there (nearfield I could hear them for sure...,) but quieter than I like. This was true immediately upon initial testing, and with my kit speakers inside an old phonograph cabinet as I'll use them permanently. (So a factor of the speaker itself, not the listening area/enclosure)

I had a Classix II prior, which also had less highs than I'd like, but not as pronounced as this.
I took a set of old Walnut Minimus 7s I have, putting them in the same positions inside this phonograph cabinet, and I like the sound balance much better (outside of the lack of low ends on those...) but that confirmed to me that improving the treble and midrange volume..I'd enjoy more.
The Minimus 7s have a nearly flat slope from what I've found, whereas the S2000s have more of a downward slope in FR regression than many other speakers. So I suspect I'd like to shift that slope to not be down 10db, but more like 3-5, or even less than that possibly to compensate for my listening setup.

The measurements: https://www.audiosciencereview.com/...s/carmody-s2000-diy-speaker-kit-review.17507/
Amir points out a 10db regression from 200 hz (Where these peak, and might be even worse /higher than that in my room at 200 hz) to 20 khz... Most other speakers he measures are flatter than this. Paul voiced these how he wanted, I get that. I just didn't realize my own personal preference was that different when I bought them. Otherwise I would've looked harder at C-Notes...but here I am, have these now. :p

As far as the S2000s.. The drivers... Tweeter is a Wavecor TW022WA02. http://www.wavecor.com/html/tw022wa02.html
The woofer is a "Denovo Audio" (DIYSG) branded driver of their own creation. I think I stumbled across what they used as a base... the Wavecor WF120BD05 has the same frame casting from what I can tell... and Wavecor also makes a stamped frame version with a fiberglass cone. I suspect DIY SG used the cast 05 as a base with a carbon fiber cone on it. I'm using that driver as my woofer in Xsim, and the predicted FR comes out VERY CLOSE to what ASR measured. http://www.wavecor.com/html/wf120bd05_06_09_10.html
The boxes are the same size as the Overnight Sensations.
Last edited by a moderator:
To fill in a little more here:
I'm using the S2000s inside a phonograph cabinet...SAF and all.... (And the amplifier will be under the cabinet on brackets soon... not on the floor.)


So I have them set up like this now:
IMG-20230513-WA0006 (1).jpg

But, the cabinet grille frame is thick (1/2" ish) AND at an angle:

Cut out the holes some (tricky without removing the old grille fabric!):


But now that I'm looking at possibly changing the crossover....I'm considering also building a single MDF box for these drivers, optimized for this enclosure/cabinet. As in, I could tilt the front to match the grille angle, and maybe overlap the tweeters a bit in the middle (Too close to get any real separation anyway) or maybe leave the tweeters towards the outside but squeeze the ports in front........and/or make the box a tad thicker and make the ports downward firing. (Stock they are rear firing into this cabinet, which did NOT sound good.)
Yes, measurements clearly show lack of highs. It seems like response is customized to such taste. As quick fix, you can remove R2 and reduce the R1 to 3,3 Ohms. That will boost the HF by approx 3dB.

Bass and mids measurements looks ok, balanced within 3dB (or so).
So, my guess is boomy sound is probably some room interactions.
Hollow sound/midrange issues might be caused by mids coming out of vent, measurements do show some high vent output around 1kHz area.
And sorry to say, but position inside that phonograph cabinet is not optimum, that may create some issues too.
  • Like
Reactions: 1 user
And sorry to say, but position inside that phonograph cabinet is not optimum, that may create some issues too.
As in, just being inside of that cabinet? Or do you think there's something that could be better this way? As I contemplate building a new enclosure/box for these, thinking maybe of putting the tweeters together (WTTW instead of TWWT) to get them away from baffle/cabinet edges, and to minimize comb filtering or anything like that between them. Sadly, due to SAF, the drivers must remain behind the grille fabric, so I do have that constraint....which I accept. Just trying to make the best of it.
Do an experiment by moving them out of the box and upright on stands into the room, away from walls.
See if they sound better.

If so, then it's the current location and position that's making them sound bad.

Understood, that's why I mentioned I observed this when I first tested the speaker in a different environment, not enclosed. Also why I tried some Minimus 7s in the cabinet to see how they sounded in that same space. So, I did the experiment 2 ways, basically....and I found I would like to change the voicing of the S2000s I have.
Speakers are laid on the sides, no big deal but the vertical response is now the horizontal response.
And the vertical in those measurements looks average to not so great.

Most the issue is way to much padding on the tweeter.
Far as lack of highs.

Rear ports i will never like for this reason.

Seem like decent drivers, a better baffle front port and crossover would help.
Reducing the padding for now would bring up the highs.
The crossover impedance curve is garbage though. big 4 ohm dip and huge rise all through the
mids. mud bucket
  • Like
Reactions: 1 user

As the consensus is that the tweeter level is set too low (and would be easy to adjust in the circuit), a simple test would be to use a laptop or somesuch where you can eq the playback signal (foobar, jriver, equalizer apo); adjust different frequency bands up and down e.g. from 1k up, and see if this improves the voicing.