Just thought I'd post a few speaker systems I built this year based on this thread where System7 Steve shared his Phase aligned Series crossover:
http://www.diyaudio.com/forums/mult...tor-parallel-inductor-low-pass-crossover.html
The 1st was the straight up Visaton based system Steve shared the design of.
I used 1" thick oak stair tread for the front.
I then built a set of 5" woofer satellite speakers using the same Visaton G20SC tweeter but with an SB Acoustics 5" woofer (SB 13PFC25-8). I used a standard PE box for these - needed to route out the inside of the grill frame to make the woofer fit.
I used REW software and a home built microphone (from a Panasonic Electret microphone capsule) to measure the SPL and acoustic Phase, Impendence and electrical Phase to input into Visaton's Boxsim program. (I used a ~3'x6' piece of MDF as a "semi infinite' baffle outside to make the SPL measurements).
Finally, I built a SEAS driver based version that shows substantially lower 3rd order harmonic distortion in the 5Khz region than the Visaton version. I had to measure both the SEAS drivers with REW and port the info into Boxsim. Of course since the SEAS 22TFF has a much more pronounced impendence hump at 1Khz than the Visaton G20SC it required some modification to the crossover to compensate.
Measurements are pretty close to the design values - especially considering the mic is not calibrated.
I am contemplating a next version with best in class tweeter distortion (Scan Speak 6600 or SEAS 27TBFC/G) paired with the SEAS Read paper 7" woofer or the Scan Speak Revelator. Any thoughts from the experts?
http://www.diyaudio.com/forums/mult...tor-parallel-inductor-low-pass-crossover.html
The 1st was the straight up Visaton based system Steve shared the design of.
I used 1" thick oak stair tread for the front.
I then built a set of 5" woofer satellite speakers using the same Visaton G20SC tweeter but with an SB Acoustics 5" woofer (SB 13PFC25-8). I used a standard PE box for these - needed to route out the inside of the grill frame to make the woofer fit.
I used REW software and a home built microphone (from a Panasonic Electret microphone capsule) to measure the SPL and acoustic Phase, Impendence and electrical Phase to input into Visaton's Boxsim program. (I used a ~3'x6' piece of MDF as a "semi infinite' baffle outside to make the SPL measurements).
Finally, I built a SEAS driver based version that shows substantially lower 3rd order harmonic distortion in the 5Khz region than the Visaton version. I had to measure both the SEAS drivers with REW and port the info into Boxsim. Of course since the SEAS 22TFF has a much more pronounced impendence hump at 1Khz than the Visaton G20SC it required some modification to the crossover to compensate.
Measurements are pretty close to the design values - especially considering the mic is not calibrated.
I am contemplating a next version with best in class tweeter distortion (Scan Speak 6600 or SEAS 27TBFC/G) paired with the SEAS Read paper 7" woofer or the Scan Speak Revelator. Any thoughts from the experts?
Attachments
Here are some of the details from the SEAS version (plus one more of the SB version)
I needed to get 4cm of set back to get phase alignment - so I increased the angle from ~ 10deg to ~ 15 deg. The box is a bit smaller at ~ 17L. Same Oak stair tread front, this time I used oak veneer on the rest of the enclosure.
I needed to get 4cm of set back to get phase alignment - so I increased the angle from ~ 10deg to ~ 15 deg. The box is a bit smaller at ~ 17L. Same Oak stair tread front, this time I used oak veneer on the rest of the enclosure.
Attachments
Aw, that is SO GOOD, my friend! I am amazed. 😀
That was a bonkers thread we did. It went up to the stratosphere.
http://www.diyaudio.com/forums/multi-way/233544-why-resistor-parallel-inductor-low-pass-crossover.html
I tend to get more interested in the theory than the practise, having limited woodwork skills. But that was a serial/series crossover circuit that I liked at every level.
For those who are interested, a shallow 5" driver is quite well time aligned with a tweeter anyway, so you can use a regular flat baffle.
With a 6" driver, you need some offset:
The whole thing was a mess until I separated out the bafflestep compensation from the main filter, and quit the stupid first order approach. Then it all works and impedance is good.
How does it sound? I'd expect a very good image and integrated sort of sound. You must have liked something to do so much work. IMO, you shouldn't be thinking a better soft dome tweeter will help. Go the whole hog and try a cone tweeter with an attenuator instead of a Zobel. This is better/cleaner sounding straight off. Trust me. Visaton TW70, Visaton TW6NG, Monacor HT 22/8, Parts Express Phenolic Ring tweeter. Take your pick.
When you align phase very well, you start to hear the sound of the drivers clearly, and their faults. With bad phase, it's all a muddle. I could explain the difference between a series and a parallel crossover with two images, but that would pander to the lazy people here, eh? 🙂
That was a bonkers thread we did. It went up to the stratosphere.
http://www.diyaudio.com/forums/multi-way/233544-why-resistor-parallel-inductor-low-pass-crossover.html
I tend to get more interested in the theory than the practise, having limited woodwork skills. But that was a serial/series crossover circuit that I liked at every level.
For those who are interested, a shallow 5" driver is quite well time aligned with a tweeter anyway, so you can use a regular flat baffle.
An externally hosted image should be here but it was not working when we last tested it.
With a 6" driver, you need some offset:
An externally hosted image should be here but it was not working when we last tested it.
The whole thing was a mess until I separated out the bafflestep compensation from the main filter, and quit the stupid first order approach. Then it all works and impedance is good.
How does it sound? I'd expect a very good image and integrated sort of sound. You must have liked something to do so much work. IMO, you shouldn't be thinking a better soft dome tweeter will help. Go the whole hog and try a cone tweeter with an attenuator instead of a Zobel. This is better/cleaner sounding straight off. Trust me. Visaton TW70, Visaton TW6NG, Monacor HT 22/8, Parts Express Phenolic Ring tweeter. Take your pick.
When you align phase very well, you start to hear the sound of the drivers clearly, and their faults. With bad phase, it's all a muddle. I could explain the difference between a series and a parallel crossover with two images, but that would pander to the lazy people here, eh? 🙂
- Status
- Not open for further replies.