Wow this is too much to comprehend this late at night. I though I knew a little bit about electronics and speaker design but your crossover designs are beyond me. At least tonight. I'll have another crack at it tomorrow. But I'm bound and determined to build the best studio monitors I can so keep the info comming if you don't mind teaching me a thing or two.
brsanko said:
If you mean my Usher 2-way design, yes, it does look complicated at first. But it simply
)) consists of an electrical third order low pass filter plus a CLR notch filter in the woofer network and an electrical second order high pass filter plus a CLR notch filter in the tweeter network.Perhaps, you haven't seen how complicated some of very respected speaker designer's crossover work is. For example, look at:
http://www.htguide.com/forum/showthread.php4?t=11321
http://www.htguide.com/forum/showthread.php4?t=13154
http://www.zaphaudio.com/ZD5.html
Compared to these, mine should be much easier to understand.
Yeah wow that's some intense stuff. I'm reading Zeph's site it's some good stuff. I think I might just make one of his designs, not sure yet. I've been reading up on speaker and amplifier design for about 7 ort 8 years now and I still have so much to learn I didn't have internet for a while so I've been out of it for a while, but it is facinating. I've found that people who are really passionate about one design or another are typically ignorant of the advantages of other designs and the short falls of their own. I have the distict disadvatage of not having that much money so must of what I know is from reading not from listening. but hopefully soon that will all change.
another new design
Ok I spent several hours looking at Zaph's site and his reviews and tests of drivers and have come up with another design all my own. I'll start with a PE premade curved .25 cu ft cab and, leave it sealed. The woofer I liked for my purposes was the Seas P18RNXP, and for tweeters I like the North D2506-S a crossover point somewhere between 2500Hz and 3000Hz using either a first order crossover, or an electronic active crossover and a five channel amp, with the subwoofer I described above. Now please feel free to tear this design apart. I've learned alot in the last couple days
Ok I spent several hours looking at Zaph's site and his reviews and tests of drivers and have come up with another design all my own. I'll start with a PE premade curved .25 cu ft cab and, leave it sealed. The woofer I liked for my purposes was the Seas P18RNXP, and for tweeters I like the North D2506-S a crossover point somewhere between 2500Hz and 3000Hz using either a first order crossover, or an electronic active crossover and a five channel amp, with the subwoofer I described above. Now please feel free to tear this design apart. I've learned alot in the last couple days
First, the PE 2.5 cu ft cabinet's baffle is only 7.5" wide and its usable width excluding the edge roundover is only 6.5". So you cannot use it for a 7" driver.
Second, the Seas P18RNX is not the best Seas Prestige 7" w.r.t. distortion performance. It is the new ER18RNX with its improved motor design.
Third, the North tweeter has been discontinued. No longer available
Fourth, if you want to have the same setup as mine (small sealed or vented standmounters plus a subwoofer), look no further than the four designs at my web pages. You're not going to find better driver combinations for 2-ways within a reasonable budget. I know this because I've already done a lot of homework. For possible 7" driver options I've found, see this post: http://www.diyaudio.com/forums/showthread.php?postid=1320168#post1320168
If you read Zaph and Mark K's driver test pages over and over again, you'll agree with my choices.
BTW, when you look at frequency response plot in Zaph's 6.5" driver test page, ignore response below 500 Hz. He used a very short gating for measuring the FR of this group of drivers.
Second, the Seas P18RNX is not the best Seas Prestige 7" w.r.t. distortion performance. It is the new ER18RNX with its improved motor design.
Third, the North tweeter has been discontinued. No longer available
Fourth, if you want to have the same setup as mine (small sealed or vented standmounters plus a subwoofer), look no further than the four designs at my web pages. You're not going to find better driver combinations for 2-ways within a reasonable budget. I know this because I've already done a lot of homework.
For possible 7" driver options I've found, see this post: http://www.diyaudio.com/forums/showthread.php?postid=1320168#post1320168If you read Zaph and Mark K's driver test pages over and over again, you'll agree with my choices.
BTW, when you look at frequency response plot in Zaph's 6.5" driver test page, ignore response below 500 Hz. He used a very short gating for measuring the FR of this group of drivers.
well according to this page you can still get the tweeter http://www.northcreekmusic.com/WhatsNew.htm I guess I'll have to have my cabinets custom made. Is the woofer you mentioned on Zaph's site? I'm sure your designs are incredible and outperform anything I'll come up with but comming up with the design myself is half the fun. Heck it's about 80% of the fun. I've still got a lot to learn about crossovers and I think I'd like to understand a design completely before I build it so my goal right now was to find drivers that would lend themselves well to a simple first order design. I'm not quite sure yet if I've done that. I still want to take a closer look at the 5.25" drivers and see what they have to offer. Thanks for all the advice.
You need to know that the electrical first order (acoustic LR2, to be more precise) is really not simple to design. To obtain good phase tracking, you'll need 1) slanted or stepped baffle, 2) delay network, or 3) waveguided tweeter. Also, there are not many choices of low-distortion drivers that satisfy the requirement of LR2 design: flat response at least 2 octaves above and below the XO frequency. This is why the acoustic LR4 design is more popular; easier to implement.
To understand this better, read Zaph's ZD5 and Waveguide TMM design pages.
Also, if you want to have fun, how about using one of the driver combinations of my designs and designing your own crossover?
There are not many 5.5" drivers that outperform the 6.5" drivers in their distortion performance even in midrange---I honestly don't know why, but that's the fact. You'll know if you research more.
All the drivers I mentioned are in Zaph's page. Also, don't forget to look at Mark K's test pages. His test method is very thorough.
To understand this better, read Zaph's ZD5 and Waveguide TMM design pages.
Also, if you want to have fun, how about using one of the driver combinations of my designs and designing your own crossover?
There are not many 5.5" drivers that outperform the 6.5" drivers in their distortion performance even in midrange---I honestly don't know why, but that's the fact. You'll know if you research more.
All the drivers I mentioned are in Zaph's page. Also, don't forget to look at Mark K's test pages. His test method is very thorough.
Before I forget, some more things to consider.
When you look at harmonic distortion measurements, remember that 3rd and 5th order harmonics are much more important than the 2nd order one. They are more audible and have worse effects.
When you read tweeters' distortion measurements, distortions above 4 to 5 kHz are not so much meaningful. For example, a high 5th order harmonic at 4 kHz is not important since it is at 20 kHz---not audible. You need to focus on their distortion levels between 1.5 kHz and 3 kHz.
When you look at harmonic distortion measurements, remember that 3rd and 5th order harmonics are much more important than the 2nd order one. They are more audible and have worse effects.
When you read tweeters' distortion measurements, distortions above 4 to 5 kHz are not so much meaningful. For example, a high 5th order harmonic at 4 kHz is not important since it is at 20 kHz---not audible. You need to focus on their distortion levels between 1.5 kHz and 3 kHz.
Ah I see that explains why the distortion graphs only go up to 10khz. I had assumed the graph showed the distortion at the frequncy where it was heard not at the frequency that produced it, critical bit of information. It's going to take me a while to understand all the complexities of crossovers but that is ok because it's going to be a few months at least before I have the money to impliment the designs in the first place (expecting a large settlement soon). Maybe I should forget about comming up with design ideas and concentrate on learning for a while. Of course this isn't quite as much fun. Now I've heard it's i'ts important to keep crossover frequencies out of the critical midrange but I guess I'm not sure what frequencies ore considered critical. Thanks again for all your help
300 Hz to 3 kHz is said to be the ciritical band. But crossing a 5.5" to 7" midwoofer at 3 kHz has some disadvantages:
1) Worse off-axis performance due to the woofer's beaming
2) High system distortions around your crossover frequency especially when your tweeter has lower distortions than the woofer around the xo point---you don't use the strength of your tweeter.
About 1.2 kHz is soprano C, which is about the highest note in usual music passage---higher notes above this is rare. So, as long as the XO point is above 1 kHz, I think we're okay.
1) Worse off-axis performance due to the woofer's beaming
2) High system distortions around your crossover frequency especially when your tweeter has lower distortions than the woofer around the xo point---you don't use the strength of your tweeter.
About 1.2 kHz is soprano C, which is about the highest note in usual music passage---higher notes above this is rare. So, as long as the XO point is above 1 kHz, I think we're okay.
tinitus,
Do you know how high 1 kHz vibration is in music we usually listen to?
Fundamental vibration of the high C, which only talented tenors can sing smoothly in some operas, is at 523 Hz. Soprano C is at 1047 Hz (I was a little inaccurate above), and soprano E, which is about the highest note a soprano sings in some operas (e.g. sung by the Queen of Night in Mozart's Magic Flute), is at 1319 Hz. Violin's very high pitch note is usually between 1 and 2 kHz. Of course, some instruments can go up 4 kHz, but it's very, very rare. In most of music passages, you listen to fundamental vibrations below 1 kHz.
Having a good tweeter in a speaker is important not because it produces fundamental vibrations of musical instruments, but because it produces delicate harmonics and sibilants that determines the unique timbre of each instrument.
The 300 Hz to 3 kHz band is not an absolute guideline. It's just conventional cutoffs. How about 300 Hz? It's also somewhat arbitrary. Male bass voice can go down to 90 Hz, and it is really used in some operas.
Do you know how high 1 kHz vibration is in music we usually listen to?
Fundamental vibration of the high C, which only talented tenors can sing smoothly in some operas, is at 523 Hz. Soprano C is at 1047 Hz (I was a little inaccurate above), and soprano E, which is about the highest note a soprano sings in some operas (e.g. sung by the Queen of Night in Mozart's Magic Flute), is at 1319 Hz. Violin's very high pitch note is usually between 1 and 2 kHz. Of course, some instruments can go up 4 kHz, but it's very, very rare. In most of music passages, you listen to fundamental vibrations below 1 kHz.
Having a good tweeter in a speaker is important not because it produces fundamental vibrations of musical instruments, but because it produces delicate harmonics and sibilants that determines the unique timbre of each instrument.
The 300 Hz to 3 kHz band is not an absolute guideline. It's just conventional cutoffs. How about 300 Hz? It's also somewhat arbitrary. Male bass voice can go down to 90 Hz, and it is really used in some operas.
You make it sound very simple I wish I had your strong conviction
A good 5" will have an easy job playing high nodes, where a tweeter will be under hard load playing low nodes
It is my experience that a tweeter crossed high sound much softer
But I do agree that with a low xo point you will probably get a more clean sound, if thats the goal ... where I believe that a higher xo point makes the sound softer, you could say more "tubelike" ... in a positive sense
But I really dont think there is any easy way or "standard" to it all
I dont think the goal should be lowest possible distortion ... I would rather like to see some understanding of the relationship between the various distortion figures
I still think the worst distortion is caused by the crossover itself
My major concern at the moment in my 3way is bass distortion, it sounds a bit too soft ... I think less than 1% is considerd low
I wish I had your strong convictionA good 5" will have an easy job playing high nodes, where a tweeter will be under hard load playing low nodes
It is my experience that a tweeter crossed high sound much softer
But I do agree that with a low xo point you will probably get a more clean sound, if thats the goal ... where I believe that a higher xo point makes the sound softer, you could say more "tubelike" ... in a positive sense
But I really dont think there is any easy way or "standard" to it all
I dont think the goal should be lowest possible distortion ... I would rather like to see some understanding of the relationship between the various distortion figures
I still think the worst distortion is caused by the crossover itself
My major concern at the moment in my 3way is bass distortion, it sounds a bit too soft ... I think less than 1% is considerd low
Choice of a low XO frequency should be depend on driver choice. You may think I always prefer to use a low crossover freq. Not at all. If I used, for example, the Dayton RS52, then I'd not use a 1.5 kHz XO point, but probably 3 kHz. Because the 2" dome has good off-axis responses and very low linear/nonlinear distortions beyond 3 kHz.
And don't jump to the conclusion that the reason why you obtain soft sound is your use of a higher XO frequency. It is more likely due to a low power response around your xo freq, which is caused by your midrange driver's beaming at higher frequencies. Even with a low xo frequency, you can produce virtually the same effect by adjusting your tweeter's low rolloff.
And don't jump to the conclusion that the reason why you obtain soft sound is your use of a higher XO frequency. It is more likely due to a low power response around your xo freq, which is caused by your midrange driver's beaming at higher frequencies. Even with a low xo frequency, you can produce virtually the same effect by adjusting your tweeter's low rolloff.
If you don't know how to design a passive crossover (like I don't), and have a way to obtain amplification for cheap, or have lots of money to burn, try active. Your amplifiers will love it. And you never know...after a lot of experimenting and tinkering...your ears might too! If you have access to some good pro amplifiers with XLR outputs like I do, they really come in handy for applications like these. You'll have more headroom than you can possibly fathom. Dynamics should increase a bit.
On the other hand, if you're very experienced in crossover design (unlike me) and don't care for any of the above, you might prefer passive.
As for beaming... usually, the larger the radiating surface area of the driver, the worse the off-axis response, or rather, the lower in frequency at which the off-axis frequency response will begin to fall off. Strictly looking at on axis and off axis frequency response, a smaller midrange will usually be easier to cross higher than a larger one. I think this is part of the reason why many people go for dome mids, since tend to look better on paper. Keep in mind, a response graph doesn't mean everything in the world.
For example, on one extreme... my 63" tall Eminent-Technology LFT-8A planar magnetic hybrids (for sale) are crossed at 180Hz from a critically damped 8" woofer in a sealed cabinet to a large array of midrange planar drivers, which then cross over to about an 18" long x 1" wide tweeter (just glanced at it and guessed the size). Anyway, off-axis frequency response is terrible, and requires that I sit with my head in a vice, commonly referred to as a small "sweet spot". They're also very difficult to position correctly in the room to obtain the best overall listening experience. As previously mentioned...move your head 2" either direction, and the sound completely changes. Stand up from the listening postion with your eyes closed, and it's as if someone mysteriously threw blankets over the speakers.
At the other extreme, take a look at the new little 2" Peerless 4 ohm "full range" driver at Madisound for $17...
http://www.madisound.com/catalog/product_info.php?products_id=8267
http://www.madisound.com/catalog/PDF/830970.pdf
...Frequency response at 60 degrees off center doesn't even begin to think about dropping off until 3 KHz, and doesn't get too extremely severe until around 10 KHz, at which point it falls drastically at around 11-12 KHz to like -23 dB or something, and then quickly rises back up again another 12dB or so. Crossed to a small woofer around 300+ Hz, you could also cross this thing to a cheapish tweeter at 3-5 KHz or more and be able to move around anywhere in the room and still get decent sound...at very low volume levels, of course. And it probably wouldn't sound as great as one would initially think. Creates more possibilities for something small and fun to play around with. Being 4 ohm though, I'm not sure how they would work in a small line array.
Basically, beaming occurs when the sound at higher frequencies is literally beaming at you from the driver. In that case, it's either crossed way too high, or isn't crossed over at all. If I remember correctly, the loudspeaker will begin to beam when the wavelength approaches the diaphragm size.
On the other hand, if you're very experienced in crossover design (unlike me) and don't care for any of the above, you might prefer passive.
As for beaming... usually, the larger the radiating surface area of the driver, the worse the off-axis response, or rather, the lower in frequency at which the off-axis frequency response will begin to fall off. Strictly looking at on axis and off axis frequency response, a smaller midrange will usually be easier to cross higher than a larger one. I think this is part of the reason why many people go for dome mids, since tend to look better on paper. Keep in mind, a response graph doesn't mean everything in the world.
For example, on one extreme... my 63" tall Eminent-Technology LFT-8A planar magnetic hybrids (for sale) are crossed at 180Hz from a critically damped 8" woofer in a sealed cabinet to a large array of midrange planar drivers, which then cross over to about an 18" long x 1" wide tweeter (just glanced at it and guessed the size). Anyway, off-axis frequency response is terrible, and requires that I sit with my head in a vice, commonly referred to as a small "sweet spot". They're also very difficult to position correctly in the room to obtain the best overall listening experience. As previously mentioned...move your head 2" either direction, and the sound completely changes. Stand up from the listening postion with your eyes closed, and it's as if someone mysteriously threw blankets over the speakers.
At the other extreme, take a look at the new little 2" Peerless 4 ohm "full range" driver at Madisound for $17...
http://www.madisound.com/catalog/product_info.php?products_id=8267
http://www.madisound.com/catalog/PDF/830970.pdf
...Frequency response at 60 degrees off center doesn't even begin to think about dropping off until 3 KHz, and doesn't get too extremely severe until around 10 KHz, at which point it falls drastically at around 11-12 KHz to like -23 dB or something, and then quickly rises back up again another 12dB or so. Crossed to a small woofer around 300+ Hz, you could also cross this thing to a cheapish tweeter at 3-5 KHz or more and be able to move around anywhere in the room and still get decent sound...at very low volume levels, of course. And it probably wouldn't sound as great as one would initially think. Creates more possibilities for something small and fun to play around with. Being 4 ohm though, I'm not sure how they would work in a small line array.
Basically, beaming occurs when the sound at higher frequencies is literally beaming at you from the driver. In that case, it's either crossed way too high, or isn't crossed over at all. If I remember correctly, the loudspeaker will begin to beam when the wavelength approaches the diaphragm size.
Avantages of an active system
-More dynamic
-Ability to modify the signal towards a desired response in a smaller volume
-Variable x-over, possible adjustments
-More efficient use of amp power
Drawbacks:
-Needs many amps
-Sound quality not as high
-Amps must have matching entry impedance else integration will be difficult
-More dynamic
-Ability to modify the signal towards a desired response in a smaller volume
-Variable x-over, possible adjustments
-More efficient use of amp power
Drawbacks:
-Needs many amps
-Sound quality not as high
-Amps must have matching entry impedance else integration will be difficult
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.