Thank you for your personal story.
Well I am glad that you are so "blessed" with such hearing.
It only doesn't have much to do with the things we were talking about before.
I always talk in a general way so it's understandable but mostly usable for other people.
I have my own way of hearing and doing things as well, I just don't see the need of sharing those because it won't be helpful to other people since they are not me.
Btw, measurements are just tools, not the end goal which you seem to suggest.
Relying on hearing is great, but not useful for either sharing information or getting back to it a certain time later.
Even less so for understanding the world around on an objective way.
Ok. I readed the Baffle step papers already.
I thougth it could be handled by the low pass filter damping, the LCR often used to tame the impedance peak after the Fb of a vented.
I saw the way to tame the baffle step with a specific serie coil added, like for instance what you see in The Edge, but founded it two serie R and inductance in the end as an overcomplicated bass and mid filter. I didn't see too much of those specialised circuitry with a standalone coil for the BSC only in the few commercial speakers I had.
I never listen to my loudspeakers near the wall, I found it kills the soundstage and also the bass lift is always too much boomy VS tigth. I usually put the front baffle at 1m to 1.2 m from the front wall ! Acceptable yet.
I assume at 11 feets listening distance one is not needing more than 3 to 4 dB of baffle compensation on the 8 dB that small baffle exhibit ?
I thougth it could be handled by the low pass filter damping, the LCR often used to tame the impedance peak after the Fb of a vented.
I saw the way to tame the baffle step with a specific serie coil added, like for instance what you see in The Edge, but founded it two serie R and inductance in the end as an overcomplicated bass and mid filter. I didn't see too much of those specialised circuitry with a standalone coil for the BSC only in the few commercial speakers I had.
I never listen to my loudspeakers near the wall, I found it kills the soundstage and also the bass lift is always too much boomy VS tigth. I usually put the front baffle at 1m to 1.2 m from the front wall ! Acceptable yet.
I assume at 11 feets listening distance one is not needing more than 3 to 4 dB of baffle compensation on the 8 dB that small baffle exhibit ?
Only in-room measurements (at the listening spot and averages over the room) will tell. Point remains that small baffles have a rather acute transition from 4π to 2π radiation, which leads to the 6dB rise. The frequency band involved is mostly between 700 and 1000Hz (as The Edge shows). So you have tot take care the room gain and the baffle step together don't create a gap at the band between about 200-300Hz and 700Hz. When floor bounce also cancels any small frequency band in that range, you're set for 'disaster'.
I'm a numpty and have decided to totally ignore these aspects under simulation. On the other hand some posts explain why. They will be in our lounge. The lounge will remain as it is with all sorts in it. There are also limited speaker placement possibilities.
In my case simulation shows a couple of dB of BS. If correct not much point in doing anything about it. It also shows expected results in a 2pi space. Again this doesn't predict what may or may not happen. All I can do under simulation is get an idea what will happen if parts need to be added to correct these areas.
I can set other spaces but not my space - that just about sums the entire area up.
Also no scope for more complex equalising other than what an AV receiver can do via it's mic.
X-over for your baffle (275x415mm): FR of the drivers have been manipulated with the diffraction tool to generate off-axis responses.
Woofer position: 168mm from bottom side
Tweter position: 338mm from bottom side/118mm from left side (20mm offset)
CtC is 170mm
Check that the drivers position is correct for your cabinet and remember that this is just a simulation.
Woofer position: 168mm from bottom side
Tweter position: 338mm from bottom side/118mm from left side (20mm offset)
CtC is 170mm
Check that the drivers position is correct for your cabinet and remember that this is just a simulation.
I just waded through the long Arkadia thread using a translator. http://www.hifi-forum.de/viewthread-205-485-7.html
The guys on the thread didn't think much of the Arkadia design and were trying to modify it e.g. with U22REX. They all got frustrated that their notch filters didn't eliminate the 4K peak. Conclusion was the CA22NRX doesn't do much above 1K.
This set me to thinking about the 22W8534 which Profiguy has been using. Looks like a smoother roll-off. It does F3=56Hz in a sealed box. Better than the 4 ohm version F3=62Hz which Troels uses in his Discovery 81 design, which in any case is huge. So once again no exact crossover but maybe an easier 8" to work with. I'd like to stay with an 8" mid-bass but we seem to be at the limits of the possible in a 2 way. Not impossible but challenging.
What do you guys think of the 22W8534?
The guys on the thread didn't think much of the Arkadia design and were trying to modify it e.g. with U22REX. They all got frustrated that their notch filters didn't eliminate the 4K peak. Conclusion was the CA22NRX doesn't do much above 1K.
This set me to thinking about the 22W8534 which Profiguy has been using. Looks like a smoother roll-off. It does F3=56Hz in a sealed box. Better than the 4 ohm version F3=62Hz which Troels uses in his Discovery 81 design, which in any case is huge. So once again no exact crossover but maybe an easier 8" to work with. I'd like to stay with an 8" mid-bass but we seem to be at the limits of the possible in a 2 way. Not impossible but challenging.
What do you guys think of the 22W8534?
I find fiber glass to sound very good and neutral and SS drivers to have better bass than many. For the rest read what I wrote about it above in the thread. And do not forgett you can not have it all. Hence perhaps the testimonia of ptofiguy about the 22RNX, he liked the mid despite a 8" for hifi is often made for bass upper bass (low mid) purpose.
Notice Seas made also excelent fiber glass drivers like the 6" Excel that was in the Proac D15 for instance.
But this Discovery line is very honnest, I have the SS 10F, a very good 3" if not one of the most neutral (especially in 4 ohms). take a 8" Augerpro WG for the SB26 ADC and choose steep filter (24 dB high pass), try 12 to 18 dB low pass on the woofer. Just a theoric approximation of course.
Notice Seas made also excelent fiber glass drivers like the 6" Excel that was in the Proac D15 for instance.
But this Discovery line is very honnest, I have the SS 10F, a very good 3" if not one of the most neutral (especially in 4 ohms). take a 8" Augerpro WG for the SB26 ADC and choose steep filter (24 dB high pass), try 12 to 18 dB low pass on the woofer. Just a theoric approximation of course.
Used in Troels Discovery 3WC MKII (LR2 on his stepped baffle), and the 4ohm version is used up to 1600hz LR4 in a closed-box Heissmann 2-way https://heissmann-acoustics.de/en/suzie/ with the high-pass cap (GHP; geschlossen mit Hochpass-Kondensator) trick popular in German DIY to squeeze a db or so more out of the bass. A lot can be gleamed from the example construction PDF on the page.
Seems a very capable driver that is easy to work with in some sims I did some time ago. The horn-tweeter used in that Heissmann design is not expensive and doesn't need a waveguide, although it needs a lot of padding with its very high sensitivity.
Much lower XO point in the MK I was published by Troels but removed after he made the MK2 version. It can still be viewed on the archive page
I'd think it'd be easy for Dave B or shadowplay to knock up a sim for you on your baffle using the 8ohm version; taking cues from the Heissmann design but maybe a more relaxed response curve instead of ruler flat
. That Scan tweeter is also pretty benign to play with in sims.
Last edited:
Andy maybe you should look more as you originally started with a Seas CA18RNX 6 1/2" driver. Paper cone, long throw. The reduction in dia reduces peaking and paper doesn't have the rather bizarre off axis behaviour alu can tend to have as F goes up. The smaller size also alters the F where beaming gets significant.
F3 at the bass end. It relates to cabinet volume as well as Fs. Take a driver that I have looked at the Dayton 8" reference model. It can do an F3 of ~30Hz in a 40L box. No chance with 20L. What tends to happen as with the Visaton driver in a small box I have worked on is that a vented cab rolls off more like sealed in the regions of interest. There should be no over bloated bass. There may be from room gain.
The trade off. At the bass end at some particular F the smaller cone has to move more than the larger one. So on the smaller speaker spl might be limited by xmax. That spl may be lower than what you need to use, The cone has to move too far and gets out of it's linear range. There may be some 20Hz around. At some point it could reach the mechanical limit, even worse. So in smaller cabinets the shift to 8" is more concerned with this aspect than the bass level reached.
Funny thing about that Dayton 8". Aluminium cone and the polar response doesn't start going awol until 3k. I have a phobia a about this sort of cone at higher frequencies.
The new idea about which driver to use - the spl curves don't thrill me to bits. There is always compromises involved and maybe the suggested cross over and driver are the best way forwards. The cross over must be doing something about the peak to give the results shown. I've copied it actually. Any inductor can be tuned to some frequency with a capacitor. What isn't clear is the sizes needed. So I will bung each in a design to see what they do. This will give me a clue. The parts also seem to be keeping the impedance range in check.
F3 at the bass end. It relates to cabinet volume as well as Fs. Take a driver that I have looked at the Dayton 8" reference model. It can do an F3 of ~30Hz in a 40L box. No chance with 20L. What tends to happen as with the Visaton driver in a small box I have worked on is that a vented cab rolls off more like sealed in the regions of interest. There should be no over bloated bass. There may be from room gain.
The trade off. At the bass end at some particular F the smaller cone has to move more than the larger one. So on the smaller speaker spl might be limited by xmax. That spl may be lower than what you need to use, The cone has to move too far and gets out of it's linear range. There may be some 20Hz around. At some point it could reach the mechanical limit, even worse. So in smaller cabinets the shift to 8" is more concerned with this aspect than the bass level reached.
Funny thing about that Dayton 8". Aluminium cone and the polar response doesn't start going awol until 3k. I have a phobia a about this sort of cone at higher frequencies.The new idea about which driver to use - the spl curves don't thrill me to bits. There is always compromises involved and maybe the suggested cross over and driver are the best way forwards. The cross over must be doing something about the peak to give the results shown.
I've copied it actually. Any inductor can be tuned to some frequency with a capacitor. What isn't clear is the sizes needed. So I will bung each in a design to see what they do. This will give me a clue. The parts also seem to be keeping the impedance range in check.
Thanks for that! The Heissmann claims -3db at 49Hz. How is this possible? I have to look up high pass cap. No doubt it has downsides or everyone would use it.
Generally, inserting a series cap about 560uf or greater into the woofer of a particularly volumed closed box interacts with the impedance such that the low-end response is boosted at the knee. It is not an exact science but can be modeled in Vituixcad and then tweaked in practice. Example here and in the thread linked before.
Not sure why it isn't used more often, maybe just cause closed boxes are not the fad.
Edit - I simmed it working in VCad with the 10" version.
Not sure why it isn't used more often, maybe just cause closed boxes are not the fad.
Edit - I simmed it working in VCad with the 10" version.
Last edited:
and a SB26ADC ANdy already has will be a better match in a 8" @augerpro WG for it ! The SB26ADC has so-so waterfall below 2500 hz but jas so low HD that a cut-off at 1500 hz is certainly feasible in such WG.
It will ask the magic of impedance flatening RLC from Shadowplay because the SET amp though.
Maybe the Seas 22 RNX is warmer in the mid or better sound (see profiguy input), but not ssure better bass.
compared to bass reflex it does not increase power handling (by excursion reduction) and it turns the system into a 4th order highpass (if i'm not mistaken!), which increases group delay compared to closed box, but usually not as much as bass reflex (also not as much low bass gain).
you can also model/simulate it in hornresp with the filter wizard.
3d order that is. But active correction of the highpass function is even more flexible I think. Just tore down a 30y old system of mine with B6 active tuning (with 6dB boost - straight to 25Hz) a la Don Keele. It has worked more than flawless for years and the woofers never ever came in the danger zone.