Could somebody check this XO Please?

[Lojzek]

You'll have to help me out and explain what the measurements are.

I'll assume you did the loopback test and the gear works.

Put the speaker away from any obstacles as much as you can and make
separate measurements of the bare drivers without moving the mic.
Mic in the 0 deg horizontally, between the drivers vertically at 1 m
distance.

Use the log sine sweep and try measuring the impedance(ARTA) of the
drivers in the box. If you can't do that, I will simulate impedance with the
manufacturer specs. Just make sure you export the Impulse response data
of the drivers as .txt files here and then we will try simulating and see how
it goes. Any box standing waves should be damped before working on the
XO filter.

 

[ODougbo]

Moved the mic out to one meter - about the same results: 10db rise at 5K
To check set-up; turned tone treble control far left - lower green line.

Measure woofer/tweeter independently - Ok will need to move the xo outside.

 

[Lojzek]

The measuring window is too short. Move it to the left. You want your
measurements to show at least from 100 Hz and above.

Please, check the loopback linearity by measuring separately:

1. only sound card with its natural sampling frequency
2. sound card + preamp
3. soundcard + preamp+amp

Turn off any EQ settings of your electronics.
Gotta be sure it's totally flat throughout 20Hz-20 kHz.

 

[ODougbo]

Lojzek,
I found a few threads on HOLM so going back through them and trying connections/settings on the computer. There is still a flat line from 20 - 200 so something is not right.
ARTA seems to be working and there are some similarities.

Mission creep I don't think I'm going to like this morel 20 tweeter no matter what, but have drop in replacement XT25's - so the mission is to swap tweeter with a new second order xo.

Secondly - need to work on measuring

 

[Lojzek]

Try zooming in the impulse response and drag the right side of the
window further to the right side of the impulse so your gating marking
travels to the left on the FR graph. Check the measurements with different
sampling frequencies. The measurement of the 2nd harmonic can go until
the half of the sampling frequency, the 3rd to 1/3 of it etc.

 

[ODougbo]

Hard to believe this and retested many times with and without woofer.

Tweeter is in the box with wires out the port:

The RED line is going though the 3rd order xo.
The BLUE line is the control; a 6.8uf cap bypassing the board completely.

 

[ODougbo]

Used the same testing process with the woofer in box (only)
Blue line is no filter
Red line is a 1.0mh coil in series
Green line is the LEAP crossover

 

[Lojzek]

Ok, now repeat the measurement of the woofer in the box
without the filter and export the measurement as a .txt file
and attach it here. Do the same for the tweeter and rememeber
to leave the mic in the same spot. Otherwise we will never do
the job.

 

[ODougbo]

Will do Lojzek (never did move anything)

Still checking here. Un soldier xo parts (tweeter side only) so I could check them. They all measured ok.
Is there any chance I wired this up wrong?
[the board was pre-wired as you see it]

 

[ODougbo]

Here's the mid-woofer zip, had to change signal length to 17.

I'll have to use a filter cap so not to damage the tweeter - 6.2uf ok?

 

[ODougbo]

Morel MDT 20 tweeter with the 6.2uf protection cap (only)

 

[ODougbo]

The last download(s) did not have the correct info.
Hopefully this will do it.

 

[ODougbo]

Ok a few start up problems, but getting closer.

Re-opened this before sending, hopefully the response is close enough to zero. Any more input and a clipping window pops up.

This is the mid-woofer in the box.

 

[ODougbo]

This is the tweeter with out moving anything.
The start frequency was set to 150.

 

[ODougbo]

For now just clipped a xo together using several combinations.

It was tough to move the 8ohm woofer only (red line); ended up with:
L2 = 2.0mh
C2 = 8uf

Once that had a reasonable curve added a 5.1uf (only) to the 8ohm tweeter.
C1 = 5.1uf

Did load up XSim...there's a huge learning curve here

 

[ODougbo]

one more time, data file transfer (only)

 

[Lojzek]

Measurements of post #36 prepared for simulation.

 

[ODougbo]

How'd I miss this post on the Troels design; woofer also on sale!
https://www.madisoundspeakerstore.com/approx-5-woofers/peerless-ppb-830860-5.25-poly-cone-woofer/

The xo board? going to send it back; it must be missing a tab or something - very strange.

For 8 ohm, this Troels design is good. It even offers the XT25 or a regular DX25TG soft dome for choice. How flat your woofers are remains a variable. The 830860 is very flat. 0.7mH Le IIRC.
Peerless HDS PPB 830860

 

[ODougbo]

Going back to OP - a resistor lead lifted and needed soldered.
How did I miss that
But the good news is - ordered new MiniDSP mic from Madisound, plug and play, very nice.

Blue line is the 2-way box with 5.25 Peerless and Morel 30 Tweeter.
(Small rise at 4.75K)

Red line is MTM, 2 x 5.25 Peerless and Vifa XT25.
(Small dip at 4K)