Hello,
I'm new to this forum, but I've been reading these posts for a while. I appreciate very much the information I have gathered from your writings.
I'm starting my first loudspeaker project soon, and I would like to hear if there is something wrong with my thoughts, or something I should remember to take into account. I'm seeking for clear improvement compared to my current speakers, Kef Q65, which are not great but quite good.
My speaker will be floorstanding 3-way and my design philosophy is to have crossover frequencies as far from critical midrange as possible. So I plan to use midrange with very wide range, and good fullrange driver is therefore good choice. I'm thinking of following drivers:
Woofer: Seas L22RN4X/P in closed box (about 20 l volume)
Midrange: Jordan JX92S in closed box (about 3 l volume)
Tweeter: Visaton MHT12
I've heard JX92S in speakers that had great sound quality to my taste. L22RN4X/P is chosen because it has good parameters for small closed box. MHT12 is chosen for it's good reputation among DIYers, I've never heard negative comments of it. Of course prices and availability at local vendors are taken into account in choice of drivers.
Upper crossover will be passive at 8.5 kHz. MHT12 can go much lower than that and JX92S doesn't have very nasty cone break-ups above that, so 2nd order will propable be enough.
Lower crossover is more difficult because I don't have separable pre and main amplifier in my integrated amp and passive crossover would require huge component values at low frequencies. I will try using JX92S without lower crossover, and hope that small cabinet volume will provide enough mechanical stiffnes to prevent driver from going beyond it's linear excursion (I don't use very high sound levels). With Seas I'll use some plate amp, propably Hypex DS1.2, that has only 12dB/octave lowpass but its easy to add 6dB/octave in speaker level to line level conversion. Crossover will be around 90 Hz.
I'll use quite wide baffle (about 35cm) because I can't place these speakers very far from back wall at my current room. With wide baffle and approximity of back wall I think I won't need very much BSC (maybe about 2dB will do).
Some notes and issues:
- Phase alignment of tweeter and midrange needs to be done with proper measurements. (I'll build small test boxes so that I can change relative offset of drivers and seek minimum amplitude at XO frequency while polarity of other driver is reversed)
- Phase alignment of woofer and midrange can't be done with sensible driver offset because XO is low. I'll rely on good luck and possibility to reverse polarity of woofer.
- Some felt on front of baffle will reduce baffle edge difractions.
- Tweeter has less horizontal directivity at XO frequency than midrange. Sudden change of directivity might sound bad. I will try using some kind of absorbing foam pads on both sides of tweeter to reduce radiation to sides. Waveguide would be other possibility, but since I don't need more sensitivity for tweeter I don't see that approach very practical.
Well, this text is getting quite long so I won't go further with details. Just let me know if this project seems sensible to you.
I'm new to this forum, but I've been reading these posts for a while. I appreciate very much the information I have gathered from your writings.
I'm starting my first loudspeaker project soon, and I would like to hear if there is something wrong with my thoughts, or something I should remember to take into account. I'm seeking for clear improvement compared to my current speakers, Kef Q65, which are not great but quite good.
My speaker will be floorstanding 3-way and my design philosophy is to have crossover frequencies as far from critical midrange as possible. So I plan to use midrange with very wide range, and good fullrange driver is therefore good choice. I'm thinking of following drivers:
Woofer: Seas L22RN4X/P in closed box (about 20 l volume)
Midrange: Jordan JX92S in closed box (about 3 l volume)
Tweeter: Visaton MHT12
I've heard JX92S in speakers that had great sound quality to my taste. L22RN4X/P is chosen because it has good parameters for small closed box. MHT12 is chosen for it's good reputation among DIYers, I've never heard negative comments of it. Of course prices and availability at local vendors are taken into account in choice of drivers.
Upper crossover will be passive at 8.5 kHz. MHT12 can go much lower than that and JX92S doesn't have very nasty cone break-ups above that, so 2nd order will propable be enough.
Lower crossover is more difficult because I don't have separable pre and main amplifier in my integrated amp and passive crossover would require huge component values at low frequencies. I will try using JX92S without lower crossover, and hope that small cabinet volume will provide enough mechanical stiffnes to prevent driver from going beyond it's linear excursion (I don't use very high sound levels). With Seas I'll use some plate amp, propably Hypex DS1.2, that has only 12dB/octave lowpass but its easy to add 6dB/octave in speaker level to line level conversion. Crossover will be around 90 Hz.
I'll use quite wide baffle (about 35cm) because I can't place these speakers very far from back wall at my current room. With wide baffle and approximity of back wall I think I won't need very much BSC (maybe about 2dB will do).
Some notes and issues:
- Phase alignment of tweeter and midrange needs to be done with proper measurements. (I'll build small test boxes so that I can change relative offset of drivers and seek minimum amplitude at XO frequency while polarity of other driver is reversed)
- Phase alignment of woofer and midrange can't be done with sensible driver offset because XO is low. I'll rely on good luck and possibility to reverse polarity of woofer.
- Some felt on front of baffle will reduce baffle edge difractions.
- Tweeter has less horizontal directivity at XO frequency than midrange. Sudden change of directivity might sound bad. I will try using some kind of absorbing foam pads on both sides of tweeter to reduce radiation to sides. Waveguide would be other possibility, but since I don't need more sensitivity for tweeter I don't see that approach very practical.
Well, this text is getting quite long so I won't go further with details. Just let me know if this project seems sensible to you.
All seems sensible enough to me. The Jordans should work well in this situation. One other idea -why stop at just a 35cm wide baffle? As you're going this far, you could go for it, and build a genuine wide-baffle design, a la the Sonus Faber Stradivari Homage -wide baffles work beautifully, believe me, and they don't need felt on the edges either. I've tried it, and I like them a lot. Here's some more information on the subject if you haven't already seen it: http://www.troelsgravesen.dk/diy_loudspeaker/IBL.pdf
Makes for interesting reading at any rate!
All the best
Scott
Makes for interesting reading at any rate!
All the best
Scott
Thanks for hints.
Yes, I've seen that link before. Using wider baffle with edges tilted backwards would be my choice if there were no constraints on floor space, wood working skills and tools and WAF. Size and shape of current design is a compromise.
But when I get my more traditional box enclosure done, I can allways try out separate wings on the sides of the speaker to find out what they do and maybe build modified enclosure later. Getting baffle step down to crossover frequency would be cool. I think my current design serves as a good starting point and DIYer can allways go forward with improvements.
Yes, I've seen that link before. Using wider baffle with edges tilted backwards would be my choice if there were no constraints on floor space, wood working skills and tools and WAF. Size and shape of current design is a compromise.
But when I get my more traditional box enclosure done, I can allways try out separate wings on the sides of the speaker to find out what they do and maybe build modified enclosure later. Getting baffle step down to crossover frequency would be cool. I think my current design serves as a good starting point and DIYer can allways go forward with improvements.
Sounds like you have a good grasp of the issues here.
3L seems a bit small for the jordan ... 7.5L gets a Q of .707 and an F3 of around 80 Hz
You could then cross at this point by just putting a 2nd order LP on the woofer, however, it will have the same F3 as the mid in a sealed box!!!
This isn't going to work well in a sealed box with a passive eq. You may want to consider going active and/or using a different driver which is a bit more efficient (or just using two of them). A 2nd driver could deal with BSC.
If you made the baffle ~0.5 wide then the -3db freq for BSC would be at 200 Hz. If you cross here to a pair of the Seas driver then you will have BSC automatically built in.
I did a quick sim in WinISD and if you use a sealed box and an xo point of 200 Hz, you get an F3 @ 60 Hz or 40 Hz for a vented version. This keeps excursion low for the mid
3L seems a bit small for the jordan ... 7.5L gets a Q of .707 and an F3 of around 80 Hz
You could then cross at this point by just putting a 2nd order LP on the woofer, however, it will have the same F3 as the mid in a sealed box!!!
This isn't going to work well in a sealed box with a passive eq. You may want to consider going active and/or using a different driver which is a bit more efficient (or just using two of them). A 2nd driver could deal with BSC.
If you made the baffle ~0.5 wide then the -3db freq for BSC would be at 200 Hz. If you cross here to a pair of the Seas driver then you will have BSC automatically built in.
I did a quick sim in WinISD and if you use a sealed box and an xo point of 200 Hz, you get an F3 @ 60 Hz or 40 Hz for a vented version. This keeps excursion low for the mid
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.