I have a problem with my two way monitor speakers.
the treble seems a bit harsh and hashy so to speak. How do I set about resolving the issue.
I am using a 40litre speaker cabinet that is reflex loaded with a tube.
The drive units are:
Seas Excel Millenium (tweeter)
Seas W22 EX (bass unit)
I am using Hovland musicap capacitors.
the damping is a combination of the rubber deflex damping sheets opposed by contoured foam.
I am wondering if lambswool will smooth out the sound without making it too woolly.
I also want to say that my tweeters are not soldered to the tabs - they are currently on little electric circuit clips - would soldering have much of an effect.
In addition to all this, my crossovers are in a seperate box from the loudspeaker, and am wondering whther point to point wiring would improve the sound, or would sticking to the PCB be the way to go?
Any suggestions will be very much appreciated
Lohan
the treble seems a bit harsh and hashy so to speak. How do I set about resolving the issue.
I am using a 40litre speaker cabinet that is reflex loaded with a tube.
The drive units are:
Seas Excel Millenium (tweeter)
Seas W22 EX (bass unit)
I am using Hovland musicap capacitors.
the damping is a combination of the rubber deflex damping sheets opposed by contoured foam.
I am wondering if lambswool will smooth out the sound without making it too woolly.
I also want to say that my tweeters are not soldered to the tabs - they are currently on little electric circuit clips - would soldering have much of an effect.
In addition to all this, my crossovers are in a seperate box from the loudspeaker, and am wondering whther point to point wiring would improve the sound, or would sticking to the PCB be the way to go?
Any suggestions will be very much appreciated
Lohan
Speaker of this caliber (expensive) should not sound harsh! I think the crossover is not professionally made. Instead of "fixing" the problem, I think it is better to redesign the crossover if it were not properly made.
I guess that during development, for the woofer to match with the tweeter, it has to be crossed high, so that the magnesium cone resonance become critical. May be it is this cone break up that is not well damped. You have to check where this break up is and check whether the LC-trap already in place and is with correct values.
If, if the problem is with the tweeter (which it should not), if you want to pad down more, you can increase the series resistor, then (re)calculate the parallel resistor to keep the original impedance
As for point-to-point or using PCB, I don't think you can hear the difference. Orientation of the coils (one to each other) is more critical.
Soldering the tweeter pads will have no audible effect, but for long term reliability and avoiding bad effect of cheap material clip, better you solder them.
For the damping/stuffing, it will affect the subjective volume of the enclosure, so you have to listen when you need to add and when you need to reduce.
Soldering the tweeter pads will have no audible effect, but for long term reliability and avoiding bad effect of cheap material clip, better you solder them.
For the damping/stuffing, it will affect the subjective volume of the enclosure, so you have to listen when you need to add and when you need to reduce.
Tweeter roll in point set too low frequency can cause harshness by allowing tweeter resonance exitation.
Cure is to shift tweeter xo point higher in frequency, another help is to fit zobel network right at tweeter terminals....be prepared to use higher value polystyrenes paralleled with polypropylenes to make it work right
Eric..
Cure is to shift tweeter xo point higher in frequency, another help is to fit zobel network right at tweeter terminals....be prepared to use higher value polystyrenes paralleled with polypropylenes to make it work right
Eric..
Hello gentleman
I got the crossover made and designed by wilmslow audio for me - so I assume it was reasonably well done.
I only use vinyl as a source, and my amplifier is a Lavardin IT - which is very clean and smooth, but not cold. My other speakers are YAmaha NS 1000m's, and despite their berrylium mid and tweeter, they are a lot cleaner and have less hash.
I am gonna fish out the schematics of my cross-over, and the loudspeaker enclosure design, and I'll post it asap
Lohan
I got the crossover made and designed by wilmslow audio for me - so I assume it was reasonably well done.
I only use vinyl as a source, and my amplifier is a Lavardin IT - which is very clean and smooth, but not cold. My other speakers are YAmaha NS 1000m's, and despite their berrylium mid and tweeter, they are a lot cleaner and have less hash.
I am gonna fish out the schematics of my cross-over, and the loudspeaker enclosure design, and I'll post it asap
Lohan
Unlikely the amp.
Possibly the nasty brealup of the midbass... no matter hoe well suppressed will still leave an imprint.
dave
What crossover frequency and slopes do you use?
The W22EX has a magnesium cone. It has a lot of strengths (I use similar drivers myself!), but the main challenge is a rather nasty breakup at 3,5 - 4 kHz. It needs to be suppressed with a steep crossover slope and perhaps a notch filter centered on the breakup frequency. In addition, the breakup will cause increasing distortion levels well below the breakup frequency, starting at some 1600 Hz and getting increasingly worse as you approach the breakup frequency. My guess is that you hear the breakup harshness in some way.
Personally, I wouldn't use the W22 in a passive two-way design. Either use two slightly smaller Excels, for instance 2 x W18EX, with steep crossover slopes, or build a three-way with W22, W15 and the Millennium. Either way, you will get some more space for the crossover slopes. I have 2 x W17EX in my speakers, and I use a DEQX HDP3 to get 60 dB/oct crossover slopes. No harshness there.
The W22EX has a magnesium cone. It has a lot of strengths (I use similar drivers myself!), but the main challenge is a rather nasty breakup at 3,5 - 4 kHz. It needs to be suppressed with a steep crossover slope and perhaps a notch filter centered on the breakup frequency. In addition, the breakup will cause increasing distortion levels well below the breakup frequency, starting at some 1600 Hz and getting increasingly worse as you approach the breakup frequency. My guess is that you hear the breakup harshness in some way.
Personally, I wouldn't use the W22 in a passive two-way design. Either use two slightly smaller Excels, for instance 2 x W18EX, with steep crossover slopes, or build a three-way with W22, W15 and the Millennium. Either way, you will get some more space for the crossover slopes. I have 2 x W17EX in my speakers, and I use a DEQX HDP3 to get 60 dB/oct crossover slopes. No harshness there.
Wire a capacitor directly across the tweeter voice coil. Start with 1 mfd. If this isn't enough, start increasing it. If it's too much, back off to 1/2 mfd. This will roll off the treble at 6 db per octave. The larger the cap, the lower the inflection point.
The best solution will depend on what frequency range the harshness comes from. Often a peak in the 4 khz range will make treble sound hard and bright. If this is the problem you need to increase the crossover point to the tweeter by inserting a capacitor in series with the tweeter. The smaller the value, the higher the LF cutoff. Start with say 20 in series with what you have now and then get progressively smaller.
Or you could just try an equalizer.
Last edited:
On second thought, I definitely think that making these two drivers work together would be problematic. I wouldn't go anything less than fourth order at 1.4khz, preferably 16th order acoustic around 1khz if your crossover is linear phase active. Even then the directivity mismatch here is likely going from an 8inch woofer to a 1.25 in tweeter ... you would want an ultra 'treated', and large room, and very particular placement.
Yes, but the Millennium tweeter is not recommended for use below 2 kHz by Seas. You start to lose power handling and get distortion from the tweeter instead if it gets pushed too hard/low. In the extreme, you have a dead Millennium.
"I also want to say that my tweeters are not soldered to the tabs - they are currently on little electric circuit clips - would soldering have much of an effect.
In addition to all this, my crossovers are in a seperate box from the loudspeaker,"
Could You provide the schematic of Your crossover?
And specify the lenght of the cables, and the thicknes of the wires You use?
Not sure, but maybe this could introduce problems, and alter the crossover function.
It should be a fairly complex crossover, as suggested by others too, as the midbass unit has a brutal peak, and the recommended freqvency ranges are on the blade's edge btween the tweeter and the woofer. Actualy I might try to convert this setup to a 3 way one. Not impossible to harmonise the 2 drivers, just seems quite a bit of challange, and probably demands ecessive use of sharp slope filters, and in order for those to work properly impedance has to be smooted precisely, it may well happen that the wires from the crossover to the speaker present enough series impedance to mess up this sensitive state.
In addition to all this, my crossovers are in a seperate box from the loudspeaker,"
Could You provide the schematic of Your crossover?
And specify the lenght of the cables, and the thicknes of the wires You use?
Not sure, but maybe this could introduce problems, and alter the crossover function.
It should be a fairly complex crossover, as suggested by others too, as the midbass unit has a brutal peak, and the recommended freqvency ranges are on the blade's edge btween the tweeter and the woofer. Actualy I might try to convert this setup to a 3 way one. Not impossible to harmonise the 2 drivers, just seems quite a bit of challange, and probably demands ecessive use of sharp slope filters, and in order for those to work properly impedance has to be smooted precisely, it may well happen that the wires from the crossover to the speaker present enough series impedance to mess up this sensitive state.
if they forgot to give it some BSC it might sound way more harsh than need be
xo schematic and pictures of actual layout is needed
I'm sure something can be done
though a big question if it will be enough
schematic may show obvious flaws
it will certainly never be perfect
but just a small improvement might make it 'less unbareable'
maybe, maybe not
not that I think its the problem
but I'm not too keen on external passive xo with long wires
its good practice to place components close to drivers
especially the 'correction' curcuits
xo schematic and pictures of actual layout is needed
I'm sure something can be done
though a big question if it will be enough
schematic may show obvious flaws
it will certainly never be perfect
but just a small improvement might make it 'less unbareable'
maybe, maybe not
not that I think its the problem
but I'm not too keen on external passive xo with long wires
its good practice to place components close to drivers
especially the 'correction' curcuits
The millenium excel is one of the finest dome tweeters in the world. While power handling improves with a higher crossover point, you don't make $200 low fs tweeters that can't handle lower frequencies. Otherwise there's barely any advantage over a lesser tweeter. Why pay so much if you're not using the high end driver to its strengths?
Woofer (seas w22ex001.LSP)
L 1 1.00 mH
C1 8.00 uF
Tweeter (seas T25cf002.LSP)
C2 12.00 uF
CL2 0.40mH
Here's my speaker circuit info, and loudspeaker calculations
BRAND/TYPE
Seas IT
Excel W22EX001
CONFIGURATION
1 single unit
Rg [Ohm] 0.00
TS PARAMETER
Enclosure
System parameter
FS [Hz] 27.00
Vented Box
F3 [Hz] 38.99
Vas [1] 75.0
Alignment: vented
Peak [dB] 0.03
Qes 0.39
Vbox [1] 33.0
Spl [dB] 87.78
Qms 2.60
Fb [Hz] 30.09
Eta [%] 0.38
Qts 0.34
Vent 1
Dv_n [cm] 9.81
Dd [cm] 18.88
Dv [cm] 8.30
Alpha 2.27
Sd [cm2] 220.0
Lv [cm] 48.51
Vd [cm3] 132
Sv [cm2] 54.11
Re [Ohm] 6.00
Q1 7.0
Mmd [g] 30.59
Eta [%] 2.65
REF [dB] 87.78
Pe [w] 120.0
L 1 1.00 mH
C1 8.00 uF
Tweeter (seas T25cf002.LSP)
C2 12.00 uF
CL2 0.40mH
Here's my speaker circuit info, and loudspeaker calculations
BRAND/TYPE
Seas IT
Excel W22EX001
CONFIGURATION
1 single unit
Rg [Ohm] 0.00
TS PARAMETER
Enclosure
System parameter
FS [Hz] 27.00
Vented Box
F3 [Hz] 38.99
Vas [1] 75.0
Alignment: vented
Peak [dB] 0.03
Qes 0.39
Vbox [1] 33.0
Spl [dB] 87.78
Qms 2.60
Fb [Hz] 30.09
Eta [%] 0.38
Qts 0.34
Vent 1
Dv_n [cm] 9.81
Dd [cm] 18.88
Dv [cm] 8.30
Alpha 2.27
Sd [cm2] 220.0
Lv [cm] 48.51
Vd [cm3] 132
Sv [cm2] 54.11
Re [Ohm] 6.00
Q1 7.0
Mmd [g] 30.59
Eta [%] 2.65
REF [dB] 87.78
Pe [w] 120.0
do You mean that Your crossover consists of only the following 4 parts?
"
L 1 1.00 mH
C1 8.00 uF
C2 12.00 uF
CL2 0.40mH"
if so, it is not a magic trick that it sounds harsh. If possible, provide a picture of the actual built crossover, and a schematic too -just in case it has more components than those-.
In case this is all there is, then You will need to rework it quite a bit, actualy make one from scratch. But surely a bit of effort and passion can do lightyears better than that , witch is a good news actualy. Since it means the harshness can be reduced/elliminated, specialy compared to current state.
"
L 1 1.00 mH
C1 8.00 uF
C2 12.00 uF
CL2 0.40mH"
if so, it is not a magic trick that it sounds harsh. If possible, provide a picture of the actual built crossover, and a schematic too -just in case it has more components than those-.
In case this is all there is, then You will need to rework it quite a bit, actualy make one from scratch. But surely a bit of effort and passion can do lightyears better than that , witch is a good news actualy. Since it means the harshness can be reduced/elliminated, specialy compared to current state.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.