I got a pair of Onkyo SKF-550F speakers for just under $20 today.
They work, however I wanted to install a proper 12 dB/octave two way crossover int hem to improve the sound.
The cap in series with the tweeter is 1.5uF and just as I suspected either the crossover is much higher than the 4kHz the manual lists or they used the cap value to lower the tweeter output by crossing it over real high either to better match it with the woofers or to reduce the increase in output near the crossover frequency caused by no inductor in series with the woofer.
I also found using a sinewave that the speaker seemed to play lower if the port was covered.
The specs of the speakers are as follows:
Impedance 8 ohms
Maximum input power 130 watts, but the individual woofers have 25 watts stamped on them so max power I'd feed them is 50 watts.
SPL 87.5dB 1W/m
Frequency response 60Hz-50kHz. There's no way these are only 3dB down at 60Hz unless mounting on the wall helps or the port is closed off.
Here's the woofers in series not in the cabinet.
Here's one woofer not int he cabinet
Here's the tweeter by itself.
What is the impedance of the woofers and tweeter?
The DATS V3 provides Re, minimum impedance and maximum impedance, but not nominal impedance.
In the past I've used Re for crossovers when the impedance was not known and that seemed to work ok.
The plan is to build a quick and dirty 6dB/octave crossover at 4.5kHz just to see if the tweeter is too loud. If it is not then I'll order or build a proper 12 dB/octave crossover.
They work, however I wanted to install a proper 12 dB/octave two way crossover int hem to improve the sound.
The cap in series with the tweeter is 1.5uF and just as I suspected either the crossover is much higher than the 4kHz the manual lists or they used the cap value to lower the tweeter output by crossing it over real high either to better match it with the woofers or to reduce the increase in output near the crossover frequency caused by no inductor in series with the woofer.
I also found using a sinewave that the speaker seemed to play lower if the port was covered.
The specs of the speakers are as follows:
Impedance 8 ohms
Maximum input power 130 watts, but the individual woofers have 25 watts stamped on them so max power I'd feed them is 50 watts.
SPL 87.5dB 1W/m
Frequency response 60Hz-50kHz. There's no way these are only 3dB down at 60Hz unless mounting on the wall helps or the port is closed off.
Here's the woofers in series not in the cabinet.
Here's one woofer not int he cabinet
Here's the tweeter by itself.
What is the impedance of the woofers and tweeter?
The DATS V3 provides Re, minimum impedance and maximum impedance, but not nominal impedance.
In the past I've used Re for crossovers when the impedance was not known and that seemed to work ok.
The plan is to build a quick and dirty 6dB/octave crossover at 4.5kHz just to see if the tweeter is too loud. If it is not then I'll order or build a proper 12 dB/octave crossover.
Last edited:
try plugging 6R into it but I may well be wrong, I often am. Personally I'd be more inclined to an asymmetric 1st/ 2nd
In order to make a 1.5uF cap produce a 4500Hz crossover I'd need a tweeter impedance of 23 ohms.
Now the real crossover frequency of the tweeter with 1.5uF cap is 26kHz.
What better way to attenuate the tweeter than by using a smaller value cap. Saves a resistor and the speaker still sounds ok for home theatre use.
Admittedly when I tested the speakers using a small amp they did sound decent.
Here's the measurement with the 1.5uF cap in series.
Also do I use Re, the minimum impedance or some average of the impedance over a specified range as the nominal driver impedance when calculating the needed parts for a crossover?
Now the real crossover frequency of the tweeter with 1.5uF cap is 26kHz.
What better way to attenuate the tweeter than by using a smaller value cap. Saves a resistor and the speaker still sounds ok for home theatre use.
Admittedly when I tested the speakers using a small amp they did sound decent.
Here's the measurement with the 1.5uF cap in series.
Also do I use Re, the minimum impedance or some average of the impedance over a specified range as the nominal driver impedance when calculating the needed parts for a crossover?
Last edited:
Now if I use a high pass filter calculator here's the result of 1.5uF and 4 ohms impedance.
http://sim.okawa-denshi.jp/en/CRhikeisan.htm
Now if I do 1.5uF and 23 ohms I get the following.
If I do 4 ohms and 8.83uF I get this.
So my guess is to save money they chose to not use an inductor in series with the woofers and instead used a lower value of tweeter coupling cap to sort of make the output at the listed crossover frequency be reasonably flat.
Provided the series woofer impedance is 8 ohms and the tweeter impedance is 4 ohms (not sure why they do that unless it's to keep the impedance down at the crossover frequency due to the lack of a zobel network on the woofer) I can go with either of these two crossover frequencies.
4kHz
https://www.parts-express.com/Dayto...ver-4-000-Hz-12-dB-Octave-260-1898?quantity=2
https://www.parts-express.com/Dayto...ver-4-000-Hz-12-dB-Octave-260-1908?quantity=2
4.5kHz
https://www.parts-express.com/Dayto...ver-4-500-Hz-12-dB-Octave-260-1920?quantity=2
https://www.parts-express.com/Dayto...ver-4-500-Hz-12-dB-Octave-260-1910?quantity=2
For the woofer zobel network I'll need a 4.3uF cap and an 8 ohm resistor.
https://www.parts-express.com/Dayto...0V-Polypropylene-Capacitor-027-419?quantity=2
https://www.parts-express.com/Dayto...ision-1-Audio-Grade-Resistor-006-8?quantity=2
http://sim.okawa-denshi.jp/en/CRhikeisan.htm
Now if I do 1.5uF and 23 ohms I get the following.
If I do 4 ohms and 8.83uF I get this.
So my guess is to save money they chose to not use an inductor in series with the woofers and instead used a lower value of tweeter coupling cap to sort of make the output at the listed crossover frequency be reasonably flat.
Provided the series woofer impedance is 8 ohms and the tweeter impedance is 4 ohms (not sure why they do that unless it's to keep the impedance down at the crossover frequency due to the lack of a zobel network on the woofer) I can go with either of these two crossover frequencies.
4kHz
https://www.parts-express.com/Dayto...ver-4-000-Hz-12-dB-Octave-260-1898?quantity=2
https://www.parts-express.com/Dayto...ver-4-000-Hz-12-dB-Octave-260-1908?quantity=2
4.5kHz
https://www.parts-express.com/Dayto...ver-4-500-Hz-12-dB-Octave-260-1920?quantity=2
https://www.parts-express.com/Dayto...ver-4-500-Hz-12-dB-Octave-260-1910?quantity=2
For the woofer zobel network I'll need a 4.3uF cap and an 8 ohm resistor.
https://www.parts-express.com/Dayto...0V-Polypropylene-Capacitor-027-419?quantity=2
https://www.parts-express.com/Dayto...ision-1-Audio-Grade-Resistor-006-8?quantity=2
So for a temporary 6dB/octave crossover just to see how the speakers sound before going with a full 12 dB/octave crossover, I'd need a 10uF cap for the tweeter and a .32mH inductor.
The idea being to see how balanced the tweeter actually is with the woofers.
If they're not balanced I will have to decide if it's worth trying to get the l-pad right.
These aren't really going to be used for critical listening unless I can get them to have a -3dB point of 60Hz. Even then based on the power listed on each woofer they are only rated at 50 watts so would be practically unusable for the kind of listening I do at times.
The idea being to see how balanced the tweeter actually is with the woofers.
If they're not balanced I will have to decide if it's worth trying to get the l-pad right.
These aren't really going to be used for critical listening unless I can get them to have a -3dB point of 60Hz. Even then based on the power listed on each woofer they are only rated at 50 watts so would be practically unusable for the kind of listening I do at times.
Last edited:
Here's the measurements of the stock speaker and the one modified with a 6dB/octave crossover and zobel.
Stock speaker port open
Stock speaker port closed.
Modified speaker port open.
Modified speaker port closed.
I think I see the difference between small signal and large signal testing.
With the signal generator of the DATS V3 the speaker played lower with the port closed, however on a proper amplifier the speaker seemed to play lower with the port open.
Listening to each speaker individually in my main system I can hear the improvement that a somewhat proper crossover and zobel make. I know the 12dB/octave crossover with zobel will make these sound even better, however the stock speaker integrated with the sub better given the woofer is in phase with the sub, whereas with the 6dB/octave crossover the woofer is 90 degrees out of phase with the woofer.
Thinking about it, the woofers without a coil in series do play higher than 4.5kHz so perhaps the 1.5uF cap on the tweeter was to keep the output of the tweeter more equal to the woofers and where the outputs were equal possibly just happened to be 4.5kHz.
That's about all I expected from a company who rated their home theatre receiver these speakers went with with one rating for just two channels driven and a lower rating for all channels driven.
The idea was to use these at work for a PC used to show a video presentation, however depending on how these sound with the 12dB/octave crossover I may rethink that idea for now until I get my Advent Maestros redone properly.
Stock speaker port open
Stock speaker port closed.
Modified speaker port open.
Modified speaker port closed.
I think I see the difference between small signal and large signal testing.
With the signal generator of the DATS V3 the speaker played lower with the port closed, however on a proper amplifier the speaker seemed to play lower with the port open.
Listening to each speaker individually in my main system I can hear the improvement that a somewhat proper crossover and zobel make. I know the 12dB/octave crossover with zobel will make these sound even better, however the stock speaker integrated with the sub better given the woofer is in phase with the sub, whereas with the 6dB/octave crossover the woofer is 90 degrees out of phase with the woofer.
Thinking about it, the woofers without a coil in series do play higher than 4.5kHz so perhaps the 1.5uF cap on the tweeter was to keep the output of the tweeter more equal to the woofers and where the outputs were equal possibly just happened to be 4.5kHz.
That's about all I expected from a company who rated their home theatre receiver these speakers went with with one rating for just two channels driven and a lower rating for all channels driven.
The idea was to use these at work for a PC used to show a video presentation, however depending on how these sound with the 12dB/octave crossover I may rethink that idea for now until I get my Advent Maestros redone properly.
Here's the crossover.
Parts are ordered.
I'm gonna use these with this amp
https://www.parts-express.com/Fosi-...mplifier-165W-x-2-350W-x-1-235-258?quantity=1
and a Sony SAWP780 sub with a built in amp that doesn't work which because the amp doesn't have a built in low pass, I decided not to fix it.
Only thing I don't like about the Fosi amp is it has no high pass filter for the main speakers which seems to be the case for most 2.1 channel amps.
I looked into adding a 12dB/octave high pass, but that required a larger inductor value than what is made for speakers.
That said, I don't want to use a capacitor in series with the speaker as the 6dB/octave roll off would cause audible issues with the roll off (assume it's 12dB/octave as the manual doesn't say) of the low pass.
It sounded ok when I tested it with the Sony sub, but it didn't seem to integrate very good which is likely due to the simple crossover the Onkyos have and I took one and did a proper 6dB/octave crossover for testing purpose so it's likely that with a 12dB/octave crossover the speakers and sub may integrate properly without needing the high pass on the main speakers.
Parts are ordered.
I'm gonna use these with this amp
https://www.parts-express.com/Fosi-...mplifier-165W-x-2-350W-x-1-235-258?quantity=1
and a Sony SAWP780 sub with a built in amp that doesn't work which because the amp doesn't have a built in low pass, I decided not to fix it.
Only thing I don't like about the Fosi amp is it has no high pass filter for the main speakers which seems to be the case for most 2.1 channel amps.
I looked into adding a 12dB/octave high pass, but that required a larger inductor value than what is made for speakers.
That said, I don't want to use a capacitor in series with the speaker as the 6dB/octave roll off would cause audible issues with the roll off (assume it's 12dB/octave as the manual doesn't say) of the low pass.
It sounded ok when I tested it with the Sony sub, but it didn't seem to integrate very good which is likely due to the simple crossover the Onkyos have and I took one and did a proper 6dB/octave crossover for testing purpose so it's likely that with a 12dB/octave crossover the speakers and sub may integrate properly without needing the high pass on the main speakers.
Got the speakers done.
Tried one last night as the glue on it to hold the crossover components in place had set up enough to where I could mount the drivers.
The integration with the sub was much better.
Gonna try the whole system in a bit.
Here's what DATS shows for the speaker.
Port open
Port closed
I noticed the Fs is different. How does that affect the -3dB point of the speaker?
Also should I try some polyfill in the box or leave it as is?
Tested the system and it sounds good.
I don't like the amp not having a high pass on the main speakers though as I know that I could make it sound better with a variable high pass for the mains.
Tried one last night as the glue on it to hold the crossover components in place had set up enough to where I could mount the drivers.
The integration with the sub was much better.
Gonna try the whole system in a bit.
Here's what DATS shows for the speaker.
Port open
Port closed
I noticed the Fs is different. How does that affect the -3dB point of the speaker?
Also should I try some polyfill in the box or leave it as is?
Tested the system and it sounds good.
I don't like the amp not having a high pass on the main speakers though as I know that I could make it sound better with a variable high pass for the mains.