Need help with a xover please...

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello

I would like to make a passive xover for my Visaton KE25SC tweeter and my Morel HU621 drivers...

I've measured them in the box (15L ported tuned to Fs of the Morel) and the graphs are below...specs for the Morel too...not much needed for the Visaton I'm guessing...8 ohm...Rdc 5.9 ohms...89db...

Any help would be appreciated!!

Thanks!!
 

Attachments

  • Visaton Small Box.jpg
    Visaton Small Box.jpg
    488.1 KB · Views: 344
  • Morel Small Box.jpg
    Morel Small Box.jpg
    462.1 KB · Views: 329
  • HU621 Specs.jpg
    HU621 Specs.jpg
    179.2 KB · Views: 324
Last edited:
The Visaton tweeter has built in LCR correction of Fs resonance, whic is handy.

The Morel looks pretty smooth.

I wonder if you fancy a serial crossover? Reckoned to be good with the right drivers:
http://www.diyaudio.com/forums/multi-way/206843-sreten-speakerman-go-series-xos-56.html#post3046046

An externally hosted image should be here but it was not working when we last tested it.


Your woofer must be quite similar to the Morel CAW 638 though I'd download the specs for comparison. You can check Zobel values here:
Impedance Equalization (L-Pad) Circuit Designer / Calculator
 
I just ran it up the flagpole on a serial crossover. It doesn't work because the tweeter is too flat on frequency response to the Fs resonance so you get too much driver overlap and a midrange peak. :mad:

Back to the drawing board. For a standmounter with bafflestep correction on a parallel crossover, it's gonna be about 1 to 2 mH slope coil as you are thinking. :)
 
I was just looking at this 6" morel HU 621 driver. How do they get the low rolloff? :confused:

Morel Loudspeakers - Raw Drivers - Woofers

Last time I saw a frequency response like that, it was an 8" or 10" SEAS Excel and Exotic woofer! :D

Yes, low order could be the order of the day. You could give it a bash with just a coil I reckon. Will mull this over and remind myself what SEAS did with the X2-08 W8 Exotic.
X2-08 EXOTIC W8

Ah, here it is:
An externally hosted image should be here but it was not working when we last tested it.


Uploaded with ImageShack.us
 

Attachments

  • Morel_HU_621_frequency_Response.JPG
    Morel_HU_621_frequency_Response.JPG
    45.6 KB · Views: 109
  • Morel_CAW_638_Frequency_Response.JPG
    Morel_CAW_638_Frequency_Response.JPG
    52.6 KB · Views: 64
Last edited:
Well...that was the DEQX measurement in the box that I am using...

The Visaton is good down to 1.5k - 1.8k...would like to use somewhere in there for a xover point...

They both seem to be natural together which is why I wanted to try a passive for giggles :D

First order seems like it could work!!

I mean there are lots of speakers that use the internal workings of the driver for their natural rolloff point and a cap for the tweeter...Reference 3a comes to mind!!

The Morel runs a bit too high though for that...therefore a first order maybe?? 10uF cap and a .47mH coil??
 
Last edited:
I think you have to state at the outset whether this is a bookshelf, or a standmounter away from the wall. It makes a big difference to your options. :)

A bookshelf can get away with a small bass coil, maybe 0.6mH here. You then need to line up the tweeter to a -6dB point around 3kHz with this woofer. Traditionally you'd then use a second order (around 4uF/0.4mH) on the tweeter. This is like the old KEF Coda with the B110 bass unit.

The standmounter works pretty much as SEAS advertise in that "Exotic" 2.1 kHz crossover. Phase and levels are a problem with this sort of guess work. First order tweeter could be tricky, but see what you can do.

The W130S I used here is nowhere near as efficient as the Morel, but it's interesting to sim it for a closed box standmounter anyway. Phase wasn't too bad.
 

Attachments

  • SEAS_Exotic_Crossover.JPG
    SEAS_Exotic_Crossover.JPG
    34 KB · Views: 155
  • Seas_Crossover_Sim_FR.JPG
    Seas_Crossover_Sim_FR.JPG
    60 KB · Views: 156
  • Seas_crossover_Sim.JPG
    Seas_crossover_Sim.JPG
    23.7 KB · Views: 156
Stand mount...front ported...sorry!!

You know Steve...thank you so much for your time on this...I really do appreciate all your help and suggestions!!

K...looks like that's a good as any place to start...coil on the driver and a simple network on the tweeter as you suggested!!

L-pad?? Would that help any??
 
I'm interested in Morel drivers, as it goes. People here slag them off for reasons I really don't understand. This one looks fun. :)

Let's reiterate what the Morel HU 621 frequency response is below. Your bass coil will tilt that down essentially and do the bafflestep correction with any luck. It's gonna be between 1mH and 2mH IMO, but it's trial and error without accurate modelling.

You will then be left with a sharpish rolloff at about -6dB @ 3kHz as mentioned. Your second order filter needs to come in at this point. You can model that bit in Boxsim to a large extent. With bafflestep correction, inevitably the tweeter needs attenuating so some sort of padding is essential. Not necessarily an L-pad. Sometimes a resistor in front of the treble filter works well too.

Try a fist order treble by all means, but I think it will be messy. At which point, I must duck out for now. Quite a tiring day at the forums, and the NFL football kicks off soon...:D
 

Attachments

  • Morel_HU_621_frequency_Response.JPG
    Morel_HU_621_frequency_Response.JPG
    45.6 KB · Views: 154
See...I don't get that either...I have not heard a bad Morel driver...they are great!!

K...gonna order some parts...a couple of coils...a set of L-pad's for giggles...caps I can get at work...will play around as soon as I can get some stuff in...will keep you posted!!

Go Bears...Go Pats!!! My SB picks :p
 
Last edited:
sideways, reference your query on good test values to try on the crossover, I include a sim. Your Morel is rolling off around 3kHz, so the sim isn't showing what is going to happen exactly and it will be better than this I reckon. Surprisingly the tweeter is wired out of phase here, but who knows how it'll turn out with the Morel.

4.7uF for sure. As the adjustable input resistor on the treble filter gets smaller, the treble coil gets bigger to approach 0.4mH, but I think 0.2mH is OK down to about 3.3 ohm input resistor so lots of leeway. If I had to guess the (red) input resistor, I'd go for 4.7 ohms initially given the woofer is efficient. It really doesn't work so nicely with an L-pad.

The bass coil is another moveable feast. Hard to guess, but 1.5mH will be less bassy than 2mH and might be realistic since no-one has speakers in the middle of the room. There's always a little wall reinforcement.
 

Attachments

  • Morel_Test_Crossover.JPG
    Morel_Test_Crossover.JPG
    22.6 KB · Views: 129
  • Morel_Test_Circuit.JPG
    Morel_Test_Circuit.JPG
    64.1 KB · Views: 60
Hey Steve

You know if I can wrap my head around some of these programs...Boxism...PCD...I might have been ok but this old brain of mine can't seem to get them to do what I want them to do LOL!!

Anyhoo...parts list...hopefully I get this right...

Woofer: 1.5mH coil

Tweeter: 4.7uF cap, 4.7 ohm resister inline with the cap, .2mH coil across the tweeter

What's the 8.2ohm resister across the tweeter for??

The 4.7 ohm resister inline with the tweeter can be changed in value to attenuate the tweeter correct??

Tweeter is out of phase because it is a 2nd order xover correct??

I hope I'm getting this all LOL!!

:)

BTW...I've heard the 166 and I think Zaph is unfair...specs don't always tell you the whole story!!
 
... some of these programs...Boxism...PCD...
If you do not put measuremnts of your drives in some CAD which can help to design crossover you will get some result, but definetly it will not be correct one, just some aproximations, and your part list can be absolutely wrong.

Correct way of crossover design is to measure you drivers in-cabinet (both - phase and freq.ressponse), add there impedance measurements for each driver and then play with schematics in CAD to get flat response from the system.

Boxsim is good for Visaton drivers and constructions, as driver parameters are there, but it is also approximation and real life responce is different from the model.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.