CHR-70 gen. 3 in folded MLTL

[onni]

And here's impulse/step-response and CSD:

/Anton

 

[onni]

I found a few inductors (salvaged from other speakers) and tried with some resistors and this combination was the one that gave the nicest response:

It's (3.9+1.8 Ohm) 4.7 Ohm and an unknown inductor (seems like an 1 mH unit).

This is the response:

Now I'm trying them in my office, sound full and nice. Looks decent too:

I'll build a box for the amplifier with the BSC, a power switch and a volume control (right now 1 % windows volume is loud enough).

/Anton

 

[onni]

The rising impedance of the driver causes the top octave to be a little too bright. So I've ordered parts for making a notch filter and at the same time a slightly larger inductor for the BSC. Here is XSim result:

Filter construction:

Cost: About 15 € (I have capacitors already).

/Anton

 

[xrk971]

Nice work Onni!

1mH and 5R is the usual 1st trial I use on a full range speaker like this for BSC. Sometimes just adding more R, like 10R can reduce that peak too rather than a notch. However, your notch sim looks great.

 

[onni]

Nice work Onni!

1mH and 5R is the usual 1st trial I use on a full range speaker like this for BSC. Sometimes just adding more R, like 10R can reduce that peak too rather than a notch. However, your notch sim looks great.

Thanks X!

I tried that as a first guess as well, but maybe the chr-70 requires slightly different values? I'm quite new to designing filters, I've mostly done DSP and PLLXO:s so far.

My last post was with a ZMA-file for the RS100-8 (they looked very similar). Now I've learned to trace the figure, so the curves look slightly different.

Here's a comparison of:
Black: No filter
Red: My proposed filter (1.6 mH+6.8 Ohm BSC + notch at ~13 kHz)
Blue: BSC consisting of 1 mH+10 Ohm components

The difference between using a 1 or 1.6 mH coil is not big, but it helps to reduce the slightly hot area around 1 kHz. I'll need to experiment with the notch values so that I don't make the sound dull off-axis.

/Anton

 

[onni]

Result with notch and new BSC

Now I've finalized the filter. The off axis (15 deg) response caused me to make the notch a little shallower to reduce the risk of making the sound dull. Instead of a 3.3 Ohm I chose a 5.6 Ohm resistor. Here is a comparison on axis, 1 m distance, ~2 m above ground:

And here 15 deg off axis:

Here is the impulse/step response 15 deg off axis:

And distortion (80 dB):

and at 85 dB:

This (the sweep) sounds fine to my ears, but when I run a sweep at 90 dB there is some obvious noise below 100 Hz.

I also made a ground plane measurement to see how the LF area behaves (in half space):

It definitely reaches below 50 Hz, I'm satisfied with that.

I've built a small box for the filter and amplifier out of a scrap piece from my media console build and some plywood:

Tomorrow is the big test when the complete setup will play background music for my 30th birthday 🙂

/Anton

 

[xrk971]

Nice amp (what was inside?)!

Nice response!

 

[onni]

Nice amp (what was inside?)!

Nice response!

Thanks!

The amp is a TPA3116 like this:

The linear pot is modified according to sound.westhost.com to make it near logarithmic.

Air coil (Jantzen 20 AWG) inductors, ceramic resistors, WIMA mkp caps.

/Anton

 

[Zoonar]

I had the pleasure of hearingen them, impressive.
The amplifier looks aly nice.

 

[zman01]

Nice work Onni, and hope the Birthday celebrations went well! 🙂