Running the Scan-Speak 15w8530k00 full-range with no x-over

Here is my latest crossover for the 2-way. I want to add the 18w-4531-00 for bass help only. Ill use a 35litre box as a stand for the 2-ways.
 

Attachments

  • xover.png
    xover.png
    138.1 KB · Views: 237
I am considering this driver for a two way. I like Gringo's approach to the XO. If you're still active on here? Are you still happy with your latest XO?
Hi there. I barely come here anymore but it feels like home. What I have now for a crossover is essentially the same as before but on the tweeter I have padded the tweeter with an L-pad voltage divider and after that a 2.2 µF series cap. to the tweeter. Voltage divider is a 2ohm in series with the driver and 15 ohms to the negative terminal. 2.8db drop.
The tone of these speakers is just that much better than my Sonus Fabers. Especially on piano and female voices. I have a complete Sonus Faber Venere surround system and I’m thinking of selling it and using my 4531s for a centre channel. Again with the same tweeter. I have any number of possible candidates for surround rear and backs. They don’t contribute much so I don’t think they need to be Scanspeak drivers as well.
I no longer need the large room filling capacity of tower speakers.
 
Last edited:
Mcspack:
Sorry frankly I can’t remember why I did any of this. Hahaha! I wish I had an XSim simulation of the present crossover with the 2.2 µF capacitor. But I had a computer meltdown and things not properly backed up. So I have no FR or ZR data for the Scanspeak drivers anymore. The reduction of the tweeter capacitor to 2.2 µF seemed to do the trick. But it likely leaves a big dip in the response curve as the crossover pole frequency for the two drivers are not the same! Putting the 1mH coil back on the woofer might help eh?
Seat of the pants ‘voicing’! I know I know ….. Like I said I’d love to see the Xsim simulation curve of it but I don’t have the ability to do that anymore. I had planned to get a test microphone and see what the curve looked like in real life. It sounds pretty good.
I’ve kind of lost interest in building stuff and just wanna listen to some music after years of tweaking.
But if I sell my Italian towers, center and surrounds I guess I’ll need to build a centre channel with those two 4531s. I have never heard the 4531s.
 
Last edited:
Isn't that input shunt going to be dumping a fair bit of power in the LF?
Huh? The amp presents the same voltage across both woofer and tweeter circuits. The impedance in the tweeter circuit and the woofer circuits will determine what current flows through each of these two circuits. Power is voltage x current. What happens in the tweeter circuit has absolutely nothing to do with the current flowing through the woofer circuit.
The amplifier is theoretically an ideal current source capable of infinite current. We know that is not the case with real amplifiers but even if the voltage sags at the amplifier output terminals it simply means less current goes through both the woofer and tweeter circuits.
 
Last edited:
A reference to the input shunt resistor handling more current / power than if it were located as it would normally be in an HF leg (i.e. in this case 'after' the cap), so it should be specified accordingly to avoid any potential for failure. Notwithstanding the slightly strange drawing, as far as I can see this is just a standard parallel 1st order electrical filter, but with a fixed series-shunt L-pad inserted 'ahead' of the HP cap. Since this is ultimately a summed circuit with an AC signal, that 15ohm R2 input shunt therefore affects the total system impedance, not just that of the tweeter (you can see that for yourself in seconds with XSim by simply flipping the series cap location relative to the resistors), and will thus need to handle more current et al.

For example (different midbass but near enough for illustration). Grey on the impedance plot = with the input shunt as-is, blue with the cap location shifted 'before' the damping resistors. Current plots 01 for the shunt resistor as is, 02 with shifted cap position.
 

Attachments

  • Impedance01.png
    Impedance01.png
    11.3 KB · Views: 38
  • C01.png
    C01.png
    9.8 KB · Views: 36
  • C02.png
    C02.png
    4 KB · Views: 42
Last edited:
A reference to the input shunt resistor handling more current / power than if it were located as it would normally be in an HF leg (i.e. in this case 'after' the cap), so it should be specified accordingly to avoid any potential for failure. Notwithstanding the slightly strange drawing, as far as I can see this is just a standard parallel 1st order electrical filter, but with a fixed series-shunt L-pad inserted 'ahead' of the HP cap. Since this is ultimately a summed circuit with an AC signal, that 15ohm R2 input shunt therefore affects the total system impedance, not just that of the tweeter (you can see that for yourself in seconds with XSim by simply flipping the series cap location relative to the resistors), and will thus need to handle more current et al.

For example (different midbass but near enough for illustration). Grey on the impedance plot = with the input shunt as-is, blue with the cap location shifted 'before' the damping resistors. Current plots 01 for the shunt resistor as is, 02 with shifted cap position.
Thanks I will look more carefully at this.
 
Interesting,

So Gringo, I'm guessing in your original XO you had too much overlap between the midbass driver and the tweeter? Is that why it maybe sounded too bright? These Scanspeak Rev 15w are reportedly the best sounding midbass drivers, that can also go fairly deep. I'm thinking an XO point between 3k and 5k? For a tweeter that is happy with a simple capacitor-resistor in series for 6db 1st order slope?

For those 4531s Gringo, what about a Quad setup? maybe put them in front since the uncoated versions are supposed to have better midrange and the 4531s are a cpl DB more sensitive? Maybe figure a way to get a good phantom center from them? I'd try that 1mH coil in there first to see if that sounds better. If you do try that 1mH coil let me know what you think about the sound...
 
Hmm,

After reading this thread again. Gringo I wonder if the 1.5mH coil might be taming some of that upper midrange rise mentioned earlier. Maybe that caused part of the brightness? I guess you'll find out by swapping/listening. Maybe a proper notch filter on the midbass to tame that rise?
 
Narry a problem. As a broad observation, damping the impedance by accident or design if you've a current source amplifier is no bad thing in & of itself, & it's been done for that & other reasons -just make sure you spec. it so it's well within tolerance.
I put the voltage divider after the capacitor in the tweeter circuit. That should be better correct Scottmoose? As I had a PC crash and do not have the FR and ZR curves anymore for these drivers I cannot see the changes in XSim. I did just acquire a measurement mic and will be using it and REW to see what kind of response I am actually getting. I will post results soon.