and before we forget:
I have been noticed that in ready made crossovers , caps used in notch filters are different from those found in the signal path
mmm, all the caps except C6 in the tweeter have resistors which negate the need for a film cap there, so you can back to back 2 polar capacitors to create a non polar or use non polars, it is up to your choice, I would not use a film cap,
The exception is if the cap has no resistor in series, then it becomes sensitive to DF and ESR.
To recapitulate, notch filter with a capacitor requires a film cap, except if there is a resistor in series with the said cap.
I have been noticed that in ready made crossovers , caps used in notch filters are different from those found in the signal path
mmm, all the caps except C6 in the tweeter have resistors which negate the need for a film cap there, so you can back to back 2 polar capacitors to create a non polar or use non polars, it is up to your choice, I would not use a film cap,
The exception is if the cap has no resistor in series, then it becomes sensitive to DF and ESR.
To recapitulate, notch filter with a capacitor requires a film cap, except if there is a resistor in series with the said cap.
Well... I think this is it, trying an ultimate way to correct the gaps in 500 700 hz, narrowing it down, etc it is not as good as this.
the XO above actually let the drivers sound their best, the Xoings
is very clean if you look the chart in the picture. from 600hz the mid should pick up and renforced by a peak of the woofer it should not sound bad at all...
I think the actual response should be better, Xsim doesn't resolve the full complexity of this XO.
the high-midrange is essentially reduced by -3db, you should hear 2x more bass relative to high frqs.
the Major aspects: the woofer has been cleaned up and is playing up to 450hz
In your design the woofer was screaming up to the 1000hz range ! this is very bad and could beam at you high frequencies.
the midrange takes over at 800hz
in your design the mid was screaming over the top of the woofer, creating lot of problems, focus, beaming, uneven sounds, renforcement, cancelation etc, down to 300hz.
the 450hz to 800hz is the critical XO region and is down 6 db ! from 500-700 hz , you confirmed you didn't hear a dip, so I think it is created by room modes and canceling at the microphone.
So, in any case this region is down but should be not that bad, maybe from 500-600 is bad... nothing you can do about it.
In the high frequencies from 1000 + , everything is cleaned up , you get at least 15db gain of resolution at the XO point, you should hear everything much clearer, with clear timing and speech precision.
You had also a nasty peak from 7khz to 9khz , this is gone,
So try it!
the XO above actually let the drivers sound their best, the Xoings
is very clean if you look the chart in the picture. from 600hz the mid should pick up and renforced by a peak of the woofer it should not sound bad at all...I think the actual response should be better, Xsim doesn't resolve the full complexity of this XO.
the high-midrange is essentially reduced by -3db, you should hear 2x more bass relative to high frqs.
the Major aspects: the woofer has been cleaned up and is playing up to 450hz
In your design the woofer was screaming up to the 1000hz range ! this is very bad and could beam at you high frequencies.
the midrange takes over at 800hz
in your design the mid was screaming over the top of the woofer, creating lot of problems, focus, beaming, uneven sounds, renforcement, cancelation etc, down to 300hz.
the 450hz to 800hz is the critical XO region and is down 6 db ! from 500-700 hz , you confirmed you didn't hear a dip, so I think it is created by room modes and canceling at the microphone.
So, in any case this region is down but should be not that bad, maybe from 500-600 is bad... nothing you can do about it.
In the high frequencies from 1000 + , everything is cleaned up , you get at least 15db gain of resolution at the XO point, you should hear everything much clearer, with clear timing and speech precision.
You had also a nasty peak from 7khz to 9khz , this is gone,
So try it!
thanks gabdx for your great job. This surely has taken you a great number of hours to fine tune. will you kindly post the final xover schematics to compare how different it is from the one in post 60. unfortunately I shall not be able to out it into practice now as I live on "an island lost at sea" as sung by Sting and inductors and audio caps are strange creatures to people selling crap chinese loudspeakers drivers here in Mauritius, so I shall have to be patient until I travel to europe which will be around late december to get my stock of components and a good lcr meter. In the meantime for tweaking purposes I shall have to resort to approximate diy coils and back to back caps. Once again many tks for precious contribution.
Hi to all , after much tweaking I finally have lost all my patience and am still not satisfied .Finally I am thinking of acquiring in a very near future either a Peerless/tymphany P830668 or a Scanspeak 26W/8534G00 to replace the monacor woofer . Any advice to help me finalize my choice taking into consideration good bass sound and the easier one as regards crossover.
In fact a friend of mine tested the SPH255 using an active xover associated with his 2 way DEAN loudspeaker ( old vifa and Audax tw) and the bass result was not so appealing. I have not been able to test your xover cause components are not availble in MRU and I shall be in Europe late november for purchase . As a precaution I think I will building the Tarkus project ( P Carmody) using the Peerless 830668, which has been highly appraised.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.