KS-223

IMG_1082.JPG

Introduction
Enclosure
Crossover
Simulations
Measurements
Sound
Summary


Introduction

This is my first and maybe the last 2.5-way design. Good feature in theory is that there is only one active mid-woofer below crossover frequency. Normal WTW 2-way may have excessive vertical directivity below X/O causing power step towards lower treble when X/O slopes are deep (>2nd order) and tweeter has not a waveguide/horn and on-axis response is flat. 2.5-way has phase difference between mid-woofers which increases power level (compared to on-axis) and directs highest pressure downwards at mid-range (with small cones). So, this is one kind of compromise - like any other speaker.

The speakers will be located in a small room (10 m2) with a couch, TV, DVD/BD player etc. Design requirements were small footprint, narrow cabinets, no speaker stands and color compatibility with TV stand (white Lundia).

W15LY001 is good looking and almost full-range with damped cone resonances. SPL capacity and non-linear distortion are not the best, but I'm sure these are easily acceptable in our secondary stereo set.
Scan-Speak R2604/832000 is close to Vifa XT25TG30-04. Slightly stuffy - at least off-axis - but clean sound without irritating coloration is expected.

Technical specifications

General 2.5-way, vented
Dimensions 168 x 1000 x 264 mm (W x H x D)
Volume 22 liters
Tuning 46 Hz, vent Ø60-70 x 163 mm
Intertechnik BR/HP 70
X/O frequencies 3300 Hz
Sensitivity 89 dB/2.83V/1m
Drivers Seas W15LY001 x2
Scan-Speak R2604/832000
Impedance 4 ohm, min. 2.9 ohm @ 205 Hz


Enclosure

Drawing KS-223.pdf

IMG_1062.JPG IMG_1067.JPG IMG_1064.JPG

IMG_1069.JPG IMG_1076.JPG

Enclosure material is 22 mm MDF. There is internal beveling (about 14x14 mm) for mid-woofers. Tweeter and upper mid-woofer are flush mounted. Damping material is located on the back wall behind mid-woofers, top and bottom - thickest on the back and bottom.


Crossover

Filter schematic, version 2 (05.05.2013):

KS-223_XO_v2.PNG

Coils are air cored, made by Graditech. Caps are SCR MKP 250 VDC.
(version 1)

IMG_1087.JPG

Simulations

Filter version 2. Response of W15LY001 is simulated with box diffraction below 400 Hz.

Frequency response

KS-223_FR-simu.png

Filter response

Woofer, lower
KS-223_FR_Net1.png

Woofer, upper
KS-223_FR_Net2.png

Tweeter
KS-223_FR_Net3.png

Phase response

KS-223_PR-simu.png

Impedance response

KS-223_ZR-simu.png

Measurements

Crossover version 1.

Frequency response

On-axis response
- measuring distance 100 cm, on tweeter axis
- time window 3.33 ms (valid >300 Hz)
- input 2.83 V
KS-223_FR_100x92cm_0deg_3p33ms_Clio-Mic01.png

Near field responses
- below 400 Hz (near field) at 5 cm, time window 1000 ms, port=green, lower W15=red, upper W15=orange, W15 signals amplified by 6 dB
KS-223_FR_5+5+5+100cm_0deg_1000+3p33ms.png

Directivity pattern horizontal 0...180°, steps 5°
- measuring distance 100 cm
- time window 3.50 ms
KS-223_DP_100x92cm_0-180deg_3p50ms_Clio.png

On-axis response, normalized to 0 dB
KS-223_DP_100x92cm_0-180deg_3p50ms_norm_Clio.png

Power response approximation blue, directivity index red, reference (on-axis/max) black
- measuring distance 120 cm
- time window 6.00 ms
- includes 36 horizontal & 36 vertical measurements
KS-223_Pow+Ref+DI_120cm_6ms_hor+ver.png

Room response
- measuring distance 220 cm, on-axis
- time window 1000 ms
KS-223_FR_220cm_0deg_1s_1p6oct.png

Step response

100 cm on-axis:
KS-223_SR_100cm.png

Harmonic distortion

Measuring distance 100 cm, on-axis
- SPL 90 dB
KS-223_HD_90dB-1m.png

SPL 100 dB
KS-223_HD_100dB-1m.png

Excess group delay

Room measurement at 220 cm, on-axis
- time window 100 ms
KS-223_GDx_220cm_100ms.png

Burst decay

Room measurement at 220 cm, on-axis
KS-223_BD_220cm.png

STI

Room measurement at 220 cm, on-axis
KS-223_STI_220cm.png

STI was measured in a living room without special room acoustics treatment producing the following T30 and EDT:
KS-223_T30_220cm.png KS-223_EDT_220cm.png

Impedance

KS-223_ZR.png

Sound

Overall balance is quite neutral but usually lower mid-range sounds a bit forward or emphasized. This may be related to 2.5-way design or Helmholtz-resonator with relatively low Q, or both. In a large room (~50 m2) bass is mild and sound tight, kicking and slightly thin. In a small room (~10 m2) mid-bass is clearly emphasized if listening place is close to back wall. Some resistive material may be needed in the ports to reduce exaggerated authority and bass extension.
Treble is generally very nice. On-axis balance may be hot close to speakers but flat power response supports also casual off-axis listening. Sound won't be too stuffy nowhere without heavy acoustical damping. Small 10°..15° toe-out is helpful at short listening distances (<2.5 m) decreasing possible biting and roughness at mid-treble. It also helps speakers to "disappear" if listening triangle is (too) wide.
Range of listening distance is wide. Decent subjective resolution/clarity and tonal balance is maintained from less than 2 meters until 3 meters if early delay time of room is tolerable, for example 300 ms like above. Resolution at bass range naturally drops at long listening distances but resolution at treble won't sound artificial (=relatively too high) at any distance.
Sound stage is stable and accurate but not pinpoint accurate.

SPL capacity is limited below 100 dB at 2 meters. That is maximum if music does not have too much pressure at the lowest octave. Cone excursion will end with extreme bass much below 100 dB. Intermodulation between bass and mid-range is audible before excursion limits are reached.


Summary

There is no need to sell or demolish these speakers. They are good all-around speakers for both intensive and background listening. Phase difference between mid-woofers won't destroy the sound, but engineering seems to be more complex than conventional 2- or 3-way. Phase aligned is easier to measure and BSC with shelving filter may enable more free and accurate response shaping.

Page created 26.07.2012
Updated 06.12.2013