Also i was wondering, for the mid, should i make a bandpass filter?
lets say the sub woofer is with a 100 hz low pass filter, and the tweeter starts at 4000 hz
should i make a 100hz - 4000hz bandpass for the mid?
or i could make a low pass cutting at 4000hz(or whateve the tweeter filter cuts at.
I could calculate the mid's thiele-small params.
lets say the sub woofer is with a 100 hz low pass filter, and the tweeter starts at 4000 hz
should i make a 100hz - 4000hz bandpass for the mid?
or i could make a low pass cutting at 4000hz(or whateve the tweeter filter cuts at.
I could calculate the mid's thiele-small params.
ok, well the laquer on the box is dry.... i have to do the XO now... what should i do?
cut it at 3k? im sure its higher that the tweeter's fs.
and my mid goes to 3k easily..
and for the mid? should i do a bandpass from 200 to 3000, or a low pass at 3000?
the woofer takes care to > 200
cut it at 3k? im sure its higher that the tweeter's fs.
and my mid goes to 3k easily..
and for the mid? should i do a bandpass from 200 to 3000, or a low pass at 3000?
the woofer takes care to > 200
I hope you haven't cut the holes yet, at least for the tweeter. Spacing is important.
For the crossover to properly work, the centers of the tweeter and the woofer should be one wavelength apart at the crossover frequency. If you want to cross over at 4K, then that means the centers should be about 3.25" apart.
Judging from the picture you gave of the tweeter, it appears to be about 2" across. So a 4" speaker will have a radius of 2"-there should be very little space between the tweeter and the woofer if the centers are 3.25" apart.
If the tweeter hole is already cut, tell me how far the center of the tweeter hole is from the center of the woofer. The wavelength of a 1000 Hz tone is 13 inches.
For the crossover to properly work, the centers of the tweeter and the woofer should be one wavelength apart at the crossover frequency. If you want to cross over at 4K, then that means the centers should be about 3.25" apart.
Judging from the picture you gave of the tweeter, it appears to be about 2" across. So a 4" speaker will have a radius of 2"-there should be very little space between the tweeter and the woofer if the centers are 3.25" apart.
If the tweeter hole is already cut, tell me how far the center of the tweeter hole is from the center of the woofer. The wavelength of a 1000 Hz tone is 13 inches.
Oh yes, one more thing. Relative sensitivity.
You say you have no tweeter of known efficiency to compare loudness levels with. Okay, how about your mid?
A 4 inch mid should be pretty linear up to 4K or 5K at least. Try running a tone through the mid at 3.5K, 4K, 5K and compare loudness levels. Use your computer's mic-it does not have to be totally accurate. You don't need a mic with regular response to play an identical tone through two speakers and to record which is louder. Just remember to keep the microphone the exact distance from each speaker-say 3 or 4 inches.
If you want to measure the tweeter's frequency response, you can run a series of tones through it that can go quite low-just keep the amplitude very low. Heck, you can run a 200 Hz tone through a tweeter if it is only one tenth of a volt. You might want to move your mic real close if the voltage is that low.
Even half a volt will pose no danger.
You say you have no tweeter of known efficiency to compare loudness levels with. Okay, how about your mid?
A 4 inch mid should be pretty linear up to 4K or 5K at least. Try running a tone through the mid at 3.5K, 4K, 5K and compare loudness levels. Use your computer's mic-it does not have to be totally accurate. You don't need a mic with regular response to play an identical tone through two speakers and to record which is louder. Just remember to keep the microphone the exact distance from each speaker-say 3 or 4 inches.
If you want to measure the tweeter's frequency response, you can run a series of tones through it that can go quite low-just keep the amplitude very low. Heck, you can run a 200 Hz tone through a tweeter if it is only one tenth of a volt. You might want to move your mic real close if the voltage is that low.
Even half a volt will pose no danger.
Thanks for the reply
Unfortunely the hole are already cut... heres a pic of the thing.
the distance between the mid's center and the tweeter's center are 3.8 inch, and the box's front is slanted so both voice coils are at same distance form the listener.
oh and... even on the mid, im not sure ot its sensitivity, i guess i'll have to listen to it alot... and arenge that with the equalizer, then once i like it, and if needed make some attenuation circuit. How about that?
Unfortunely the hole are already cut... heres a pic of the thing.
the distance between the mid's center and the tweeter's center are 3.8 inch, and the box's front is slanted so both voice coils are at same distance form the listener.
oh and... even on the mid, im not sure ot its sensitivity, i guess i'll have to listen to it alot... and arenge that with the equalizer, then once i like it, and if needed make some attenuation circuit. How about that?
Attachments
Nice lacquer job on the speaker 🙂
You should change that quote to read, "Fortunately, the holes were cut so the centers are 3.8 inches apart, which will give us one wavelength at 3500 Hz, where we will cross over-very close to the 4000 Hz we originally planned.
Tha slanted board might give us some problem. Can you give me the dimensions of the speaker, including the different dimensions of the top and bottom? From there, I can figure the angle and offset-not that I am an expert.
Also, with your test equipment, do you think that you can do some tests for an impedance curve and possibly a phase curve for your tweeter? We can guess without them, but it would be better if we knew.
Ilianh said:
You should change that quote to read, "Fortunately, the holes were cut so the centers are 3.8 inches apart, which will give us one wavelength at 3500 Hz, where we will cross over-very close to the 4000 Hz we originally planned.
Tha slanted board might give us some problem. Can you give me the dimensions of the speaker, including the different dimensions of the top and bottom? From there, I can figure the angle and offset-not that I am an expert.
Also, with your test equipment, do you think that you can do some tests for an impedance curve and possibly a phase curve for your tweeter? We can guess without them, but it would be better if we knew.
You want an LC 12 dB/octave filter, right? The point I am driving at is the books will tell you that the polarity of one of the drivers-mid or tweeter-must be reversed in such an arrangement. Otherwise, the response is supposed to be 6 dB down at the crossover.
For some time, I have been wondering about the phenomenon of reinforcement when two speakers have their centers one wavelength apart. I ran into some problems with that once.
The thought occurred to me, (and to Roddyama, in another thread), that if you keep the polarity straight and place the speakers so the centers are one wavelength apart at the crossover frequency, the two effects will cancel and you can have both correct polarity and a 12 dB/octave network.
After examining several speakers and crossovers, including some supplied by Softman in another thread, (thanks, Softman), it can be seen that this is precisely what happens.
So you can use the values for a 12 dB/octave network and enjoy correct polarity.
This is my plan for your crossover.
For some time, I have been wondering about the phenomenon of reinforcement when two speakers have their centers one wavelength apart. I ran into some problems with that once.
The thought occurred to me, (and to Roddyama, in another thread), that if you keep the polarity straight and place the speakers so the centers are one wavelength apart at the crossover frequency, the two effects will cancel and you can have both correct polarity and a 12 dB/octave network.
After examining several speakers and crossovers, including some supplied by Softman in another thread, (thanks, Softman), it can be seen that this is precisely what happens.
So you can use the values for a 12 dB/octave network and enjoy correct polarity.
This is my plan for your crossover.
Okay, I will calculate the offset of the tweeter from the woofer from your drawing.
Actually, if the voice coils are better lined up because of the slant of your baffle, maybe the results will more accurately follow the formulas instead of throwing them off! We shall see.
YO said you had an oscilloscope. YOu should be able to get phase from there, if you can hook a mic up to it. Instead of a curve, perhaps you should just check above 2K
Such as:
2K
2.5K
3K
3.5K
4K
4.5K
5K
6K
8K
10K
13K
16K
Just record the values for those frequencies.
Impedance curve-or impedance values-can be done with a resistor and meter. I'll get back to yo on that later or tomorrow.
Actually, if the voice coils are better lined up because of the slant of your baffle, maybe the results will more accurately follow the formulas instead of throwing them off! We shall see.
YO said you had an oscilloscope. YOu should be able to get phase from there, if you can hook a mic up to it. Instead of a curve, perhaps you should just check above 2K
Such as:
2K
2.5K
3K
3.5K
4K
4.5K
5K
6K
8K
10K
13K
16K
Just record the values for those frequencies.
Impedance curve-or impedance values-can be done with a resistor and meter. I'll get back to yo on that later or tomorrow.
Nelson Pass said:
Should i just try until i like the sound? 😉
also for a second order butterworth, i get these values on 3.5khz,
4.02 uF and 0.51 mH (the cap would be 10% accuracy and the coil just cant be that precise, even if i wind my own) , and im not sure i can find parts this those exact values
What if the thing cuts at lets say, 3400 or 3600 hz?
Then the space between the tweeter's center and the mid's center wont match the cutting frequency wavelenght.
another thing, on second order crossovers, i see that the tweeters polarity is inversed, when i make tests using a 8 ohm resistance i get no 180 degree dephasation on the high pass filter. All I get is like a 10 degree dephasation (i tried first and second orders and they have that too)
I may try the microphone thing as you said, but i wonder if id need a soundproof room to get somehow precise results.
- Status
- Not open for further replies.