Hi Everyone,
I'm looking to build a 3-way based on the sb23nrxs45-8. I am specifically looking at a popular DIY design that I don't really want to mention, so as not to go off topic with love or hate of particular designer.
The design as is, uses the SB23nrxs45-8 in a 35 liter cabinet with a 35hz tuning.
Using a simulator, this looks to give an F3 of ~40hz with a 1.5db bump around 70hz. This is pretty much what the original design notes seem to have come up with.
Choosing a 50 liter box tuned to 30hz looks to give and F3 of around 35hz with fairly flat response.
70 liters tuned to 26hz gives and F3 of ~30hz. The modeled response here is very flat.
Here is my question: Are there any reasons not to choose the larger enclosure?
With the 35L box, excursion is held just below X-Max. The 50L box pushes the woofer just passed its limit, and the 70L box pushes it beyond by 2mm or so. But this is at the rated max wattage of 90w, correct?
Group delay from the 50L box is not so dissimilar from the 35L box. The 70L box is modeled at about 10 milliseconds delay for the 20-30hz frequencies.
Any thoughts?
Should I bother going down this road, or just build the speakers as designed?
(Ha ha! I know. That is something only I can decide.)
I'm looking to build a 3-way based on the sb23nrxs45-8. I am specifically looking at a popular DIY design that I don't really want to mention, so as not to go off topic with love or hate of particular designer.
The design as is, uses the SB23nrxs45-8 in a 35 liter cabinet with a 35hz tuning.
Using a simulator, this looks to give an F3 of ~40hz with a 1.5db bump around 70hz. This is pretty much what the original design notes seem to have come up with.
Choosing a 50 liter box tuned to 30hz looks to give and F3 of around 35hz with fairly flat response.
70 liters tuned to 26hz gives and F3 of ~30hz. The modeled response here is very flat.
Here is my question: Are there any reasons not to choose the larger enclosure?
With the 35L box, excursion is held just below X-Max. The 50L box pushes the woofer just passed its limit, and the 70L box pushes it beyond by 2mm or so. But this is at the rated max wattage of 90w, correct?
Group delay from the 50L box is not so dissimilar from the 35L box. The 70L box is modeled at about 10 milliseconds delay for the 20-30hz frequencies.
Any thoughts?
Should I bother going down this road, or just build the speakers as designed?
(Ha ha! I know. That is something only I can decide.)
If the 70 liter (~2.5cuft) cabinet fits in your home layout, it will produce the best bass tuned to Fs=27Hz as you simulated. To reduce panel resonances, remember to add a few horizontal multi-hole shelf(window) braces, and a cabinet top-to-bottom brace which locks everything together. You still want your tweeter at seated ear level ~36"-39", which with normal tweeter dispersion will also sound coherent to standing adults.
Hi LineSource,
Thanks for responding. 70L would make the box a 1 meter tall floor stander, which would be about ideal--tweeter height wise.
In WindISD, if I run the numbers the anonymous designer measured, the 70 liter models more like EBS alignment, with droop starting at 200hz, and shelving between 50 and 32hz (F3) and then drops off more steeply. Perhaps this is why he picked the 35L/35Fb.
Also in WinISD, the model shows no X-Max problems at all. At first I was just messing around with Speakerboxlite because they already had the SB woofer parameters in their database.
Can you think of any trade-offs other than size? The designer said when I emailed, that a larger enclosure/lower tuning would likely lead to sloppy bass. Hmm... under-damped?
Thanks for responding. 70L would make the box a 1 meter tall floor stander, which would be about ideal--tweeter height wise.
In WindISD, if I run the numbers the anonymous designer measured, the 70 liter models more like EBS alignment, with droop starting at 200hz, and shelving between 50 and 32hz (F3) and then drops off more steeply. Perhaps this is why he picked the 35L/35Fb.
Also in WinISD, the model shows no X-Max problems at all. At first I was just messing around with Speakerboxlite because they already had the SB woofer parameters in their database.
Can you think of any trade-offs other than size? The designer said when I emailed, that a larger enclosure/lower tuning would likely lead to sloppy bass. Hmm... under-damped?
Yes, larger vented box have sloppier, more uncontrolled low end.
Smaller vented box acts more like a closed box as you go smaller in size, but you get a small hump in the mid- higher bass as a trade-off, which is not as bad (even preferred in some situations) as the sloppy low end with a shelf, if you ask me.
Check the manufacturer recommendations, 42L vented marked as damped and 58L as less-damped, 35L would be fine in my opinion.
8'' SB23NRXS45-8 :: SB Acoustics
Last edited:
depending on your room, if you have very stiff and heavy walls i would stay away from any bump in the low end anechoic freq response. Woofers with strong magnets in small boxes tuned low usually gets the most balanced low end in such rooms. However, rooms with flimsy walls can work great with peaky bass speakers. And there are hundreds of other things to consider as well.
If you wonder why the manufacturer recommend smaller than "optimum" vented box for this driver, just check their EBP value (Fs/Qes). 65.8 in this case, which is tending towards for a closed box, but usable for vented too with good result.
Last edited:
Thanks Denibeni, Celef,
Part of the inconsistency comes from the two sets of measurements: the manufacturer's and the designer's own measurements of his samples. The 35L/35FB box with his measurements sims as a pretty flat FR plot in WinISD.
It would be nice to have an F10 in the low twenties, but I don't want to sacrifice overall SQ.
I'm currently using Event Oplas, which in my room have considerable measured output down to 20hz. It is nice to have that low bass for live recordings, orchestral, organ works, grand piano, field recordings, etc. Not really noticeable for most jazz or rock, depending on how the drum kit was mic'd.
What I really should do is put together two rough boxes, one 35L and one at whatever size looks to be the best compromise in WinISD and measure (and listen) in my room.
Not really sure I am that disciplined right now, though! Too many other things going on, unfortunately.
Part of the inconsistency comes from the two sets of measurements: the manufacturer's and the designer's own measurements of his samples. The 35L/35FB box with his measurements sims as a pretty flat FR plot in WinISD.
It would be nice to have an F10 in the low twenties, but I don't want to sacrifice overall SQ.
I'm currently using Event Oplas, which in my room have considerable measured output down to 20hz. It is nice to have that low bass for live recordings, orchestral, organ works, grand piano, field recordings, etc. Not really noticeable for most jazz or rock, depending on how the drum kit was mic'd.
What I really should do is put together two rough boxes, one 35L and one at whatever size looks to be the best compromise in WinISD and measure (and listen) in my room.
Not really sure I am that disciplined right now, though! Too many other things going on, unfortunately.
- Status
- Not open for further replies.