Seems to be the idea that works, nice job!
2.5 dB for BSC indicates that you use speakers quite near wall? (as I do)
Have low XO you used as 3-way? 5 octaves do ~100-3200 Hz.
I think you use 1st slope for 0.5 way? Doesn't that make hole in freq response in the area below baffle step frequency if you cross 0,3-0,5 times lower than baffle step?
Bass driver don't compensate baffle step loss or does it? I think 18W is too sensitive in mids..
I believe it is very good design. Maybe suitable for large mid driver so it won't suffer that much of lacking high pass.
One trick could be using low Qts driver (and not very low Fs) for midwoofer so that its natural rolloff starts at ~200-250 Hz. So you can cross basswoofer bit high, more load for bigger woofer and smaller coil for XO.
I would say it dosen't matter aslong as you don't have a mid-bass driver. Main issue is intermodulation distortion caused by the bass driver moving as it produces midrange. One way around this for example is to use subwoofers with full range drivers like I do. Another very large bass drivers not playing high like gedlee. A third way around this is to filter the midrange from the bass driver making a 3 way. The traditional 2 way with a 7" woofer and 2/3K crossover is rubbish at loud levels though.
My rational for considering a 4-Way is to clean up the mid bass, see the Frankenstein becomes Cinderella thread for an example
7" woofer, 2K crossover
I'm a fan of the 7" woofer, 2K crossover, it's easy, it works. Finished this SR71 project last night. Lined them with lead, 25lbs each (they won't ring).
This is about a $400 project; this Zaph SR71 kit has good reviews as far as I know.
I'm a fan of the 7" woofer, 2K crossover, it's easy, it works. Finished this SR71 project last night. Lined them with lead, 25lbs each (they won't ring).
This is about a $400 project; this Zaph SR71 kit has good reviews as far as I know.
Attachments
Yeah, haven't seen many in club gigs, nor at the stadiums.🙂
All know that they are not the right speakers for 70 m2 room with 3,5 m height and 4,5 m listening distance.
Last edited:
My buddy has a 7" Vifa woofer, morel tweeter w/ Madsiound Leap crossover in a $10,000 system, it the best two channel sytem I every heard.
I built these for a co-worker who is just getting into HyFi.
I built these for a co-worker who is just getting into HyFi.
Last edited:
I can back up my claims they are rubbish at high levels. Simply simulate the frequancy modulation caused by 2mmpp cone movment at 80Hz while playing a 1KHz tone. For even greater effect simulate the effect of such a cone movment on a music track. Sounds like the 'loud' sound you get from overstressed small speakers. I have some example fies/code on my laptop which I can get after the weekend.
QSC K8s, K10s
I didn't realize it was a large room, that said my friend's syetem would work, he has 10" subs stacked to the ceiling and 1200 watts (4 - 300 watt mono amps).
I'd go for a powered speaker. The QSC K8s sound more like HiFy speakers han most. Lots of great reviews on the web.
QSC - K Series K8 Active Loudspeaker
I didn't realize it was a large room, that said my friend's syetem would work, he has 10" subs stacked to the ceiling and 1200 watts (4 - 300 watt mono amps).
I'd go for a powered speaker. The QSC K8s sound more like HiFy speakers han most. Lots of great reviews on the web.
QSC - K Series K8 Active Loudspeaker
Yes they are near the back wall and the woofer cone is only 500mm from the back wall.
The 3-way crossed over around 250Hz to the mid-woofer but in the end I preferred the 18W8530 to do from upper bass onward. Removing the lower bass and some mid-bass duties and placing it in an undersized enclosure meant the excursion was reduced by over 60% and add in half excursion due to parallel with the woofer means there's very little excursion at normal listening levels.
It's a 1st order on the 0.5 woofer but with an oversized zobel to make the roll off steeper past the crossover point. The FR graph shows raw drivers only, no baffle step included and no effects of the enclosures. Even so you can see summing right up to 1kHz but if you figure in baffle step losses and woofer off axis performance above 200Hz then the shape would become more realistic.
Attachments
Looks nice, I made the mind experiment you proposed and I see it as ~-6db@30 hz and about 5 db room for dealing with baffle step, pretty nice. Debending on the width of the baffle, you could have a 2 to 3 db peak somewhere between 400 and 900 hz, which should not be considered if it embraces less than 1/8th of an octave. 1/8th of an octave will be about 70-100 hz width of the affected area. If you track down that peak you could size the Zobel in a such manner that it addresses it. The zobel will still straighten the impedance from that frequency up but will serve as a notch too. It could become a necessity to enlrge the inductor a little bit in order to prevent too low impedance in that area.
As far as I am informed, the baffle step is alway ~6 db, but it always begins with slight peak which height depends on the phisical position of the driver on the baffle. This peak sometimes is only a db or two, but it can reach up to 4 db height above the mean baffle step of 6 db.
Best regards!
As far as I am informed, the baffle step is alway ~6 db, but it always begins with slight peak which height depends on the phisical position of the driver on the baffle. This peak sometimes is only a db or two, but it can reach up to 4 db height above the mean baffle step of 6 db.
Best regards!
I don't get why you come here. Is this place like therapy for you? Your posts are usually abrasive and don't help any one but your own ego. The fact you don't see why people use certain drivers is particularly short sighted.
Why not take up a different hobby? You'll probably be happier.
rabbitz: I love your speaker, it looks awesome! Any more info on the XO design and tweeter/woofer used?
The baffle width is 320mm. I measured the FR at the listening position when it was an OB mid in 2006. There was a dip in the response between 450Hz and 700Hz which I put down to OB losses (Fp=476Hz) and not room suckout which would be much lower at below 250Hz.
I haven't measured it since the top section was made into a sealed section in 2008.
Thanks.
The tweeter is Peerless 810921, mid-woofer is SS 18W8531G00 and the woofer is Vifa M22WR-09-08.
The crossover is an AR series without BSC for the top section and a parallel crossover for the 0.5 woofer. I've shown the details for reference only and the top section would not work for a 2 way as there's no BSC. It would need a larger L2 and other changes.
C+R tames the rising response of the tweeter.
Czz+Rzz is used for roll off shaping for the mid-woofer and phase at the crossover point.
R1 sets the tweeter level.
L1+C1 sets the tweeter crossover.
L2+C1 sets the midwoofer crossover.
L3 sets the woofer crossover and Cz+Rz helps to steepen the roll off.
Impedance of top section is 6R5 to 9R3 but minimum drops to 3R5 (can be an amp killer) when the 0.5 woofer is added in parallel (connecting in series does not work as becomes part of the series network).
I've added the modelled FR for the top section and the reverse null.
BTW.... Tony Gee has designed a lot of speakers that look like 3-ways but are in fact 2.5 ways with series crossovers for the upper section. They can work very well indeed.
Attachments
Last edited:
Back to the topic. I chose this path as I have only been able to design a good 3-way but I can design a great 2.5 way (or 2.2 speaker). So, I know my limitations and stick within it.
My last 3-way (see pic) sounded good but not good enough so was scrapped. So get too ambitious and it can end in tears. Different story if you can find a great established design or are a brilliant designer.
There's no winner or loser in 2-way vs 3-way as in the end, it's what's best for the intended application.
My last 3-way (see pic) sounded good but not good enough so was scrapped. So get too ambitious and it can end in tears. Different story if you can find a great established design or are a brilliant designer.
There's no winner or loser in 2-way vs 3-way as in the end, it's what's best for the intended application.
Attachments
Sorry I did not read your post earlier.
To crossover and keep the driver within wave length distance. 34cm-10cm
Last edited:
But that is a decision between two design drawbacks, whether to sacrifice some vertical dispersion in the higher frequencies or to crossover in a more sensitive region of the human ear. Also in the lower crossover are used bigger parts, it must be steeper to avoid issues at higher loads and it's prone to sibilance intermodulation and such.
On the other hand, higher freuency crossover would free you up from the need to design RLC addressing the tweeter impedance at resonance. But that's not an argument, because it is a consideration which is easier and not which is superior.
On the other hand Linkwitz and Riley address the directivity and dispersion issue with only time aligning the acoustic centers thus avoiding the limitations of vertical center to center distance.
But my opinion is inclined because of the prejudice that a tweeter is an inept piece of hardware that should operate only in the last octave 10 to 20 khz.
Then you dealed with it, or it even didn't existed. A well positioned driver on the baffle could return as low as ~1db peak in the beginning of the 6 db plato of the baffle step.
I am very curious, whether it was coincidence or you designed it like that.
It's very nice to see common speaker issues approached and addressed sucessfully 🙂
I do not understand what you try to say.
I know what I want to explain here, with the close placement I want to prevent to big interference distortion. That could cause holes and peaks in the response. I was not thinking about directivity.
- Status
- Not open for further replies.