Hi,
Im using the parts express dayton ultimaxes diy kit (sealed) in my 2 channel (2.1) setup. Im varying the crossover but sofar the best sounding is around 100hz. Sadly the Daytons seem to not be as tight above 70hz and ive been reading more and more about how they are actually more a home theater sub driver and they are great below 70. Fantastic even.
So im looking at an upgrade driver that goes well 20hz-110hz thereabouts and maybe using in the same cabinet, or make a new one with winIS, or buy a package at GSG. What would your preference be for replacing the daytons? This is regardless my room condition etc. Just purely looking at the driver itself. Ideally Id have a subwoofer with a good quality of the ultimaxes below 50hz and to about 20hz, but also one that gets nice and tight at higher frequencies.
Im using a Behringer 6000dsp, and a minidsp. Been experimenting with MSO and REW, using UMIK1. MSO is nice though it is catering for home theater sub setups more than music usage imo. I have Revel F208 main speakers, driven by a Quad 606mki, and a crown xsl 1502. I aim to have a system capable between 103db spl for some nice techno moments, about 80 for average music and sometimes lower levels for classical or jazz. My budget is anywhere about twice what the dayton ultimaxes are. Im not considering manufacturer subwoofers like SVS (had them, liked them). I want to stay with diy. Thanks!
Im using the parts express dayton ultimaxes diy kit (sealed) in my 2 channel (2.1) setup. Im varying the crossover but sofar the best sounding is around 100hz. Sadly the Daytons seem to not be as tight above 70hz and ive been reading more and more about how they are actually more a home theater sub driver and they are great below 70. Fantastic even.
So im looking at an upgrade driver that goes well 20hz-110hz thereabouts and maybe using in the same cabinet, or make a new one with winIS, or buy a package at GSG. What would your preference be for replacing the daytons? This is regardless my room condition etc. Just purely looking at the driver itself. Ideally Id have a subwoofer with a good quality of the ultimaxes below 50hz and to about 20hz, but also one that gets nice and tight at higher frequencies.
Im using a Behringer 6000dsp, and a minidsp. Been experimenting with MSO and REW, using UMIK1. MSO is nice though it is catering for home theater sub setups more than music usage imo. I have Revel F208 main speakers, driven by a Quad 606mki, and a crown xsl 1502. I aim to have a system capable between 103db spl for some nice techno moments, about 80 for average music and sometimes lower levels for classical or jazz. My budget is anywhere about twice what the dayton ultimaxes are. Im not considering manufacturer subwoofers like SVS (had them, liked them). I want to stay with diy. Thanks!
Last edited:
If you need to cross over higher than 100Hz, then I would recommend that you do not use a 15" subwoofer. Try a 12" instead. I have a pair of the CSS SDX 12" in about 1.5 cufu each and they are great, with excellent SQ.
https://www.css-audio.com/online-store/CSS-SDX12-12-XBL^2-Subwoofer-p110031759
I powered these with a NX3000, so your 6000DSP should work just fine. Keep in mind that the power delivery from these amps falls off below 30Hz. For example, dyno testing shows the NX3000 can only produce about 475W at 20Hz (continuous) into 4 Ohms and only for 10 seconds. But this is plenty of power to drive the CSS sub. If you need more power, look for a better amp.
https://www.css-audio.com/online-store/CSS-SDX12-12-XBL^2-Subwoofer-p110031759
I powered these with a NX3000, so your 6000DSP should work just fine. Keep in mind that the power delivery from these amps falls off below 30Hz. For example, dyno testing shows the NX3000 can only produce about 475W at 20Hz (continuous) into 4 Ohms and only for 10 seconds. But this is plenty of power to drive the CSS sub. If you need more power, look for a better amp.
What crossover settings are you using? Are they second-order, third-order, or fourth-order? Is the filter topology Linkwitz–Riley, Butterworth, or something else? The miniDSP should offer some of these as options for the low-pass and high-pass filtering.
The published frequency response of the Ultimax II 15" DVC Subwoofer 2 Ohm Per Coil driver is shown below for reference. If you are trying to use a 100Hz crossover frequency, then it would seem prudent to be using at least a 4th-order 24dB/octave low-pass filter to remove as much of the driver's response above 200Hz as possible.
Hi witwald,
Bear with me i dont see the quote option in your post.
Im now using LR24dB. After some chat with Harbottle's designer im now testing Buttersworth agan, and a high pass filter (6db) at 25hz. I was contemplating that but didnt find any info on putting a high pass filter down low in a sub driver on any forum!
Either way, it sounds better but after all considerations I think im leaning towards a B&C 15DS115. Im also leaning towards switching towards a ported design instead of sealed. Ive seen many great B&C SW115 ported design but they go to 3k hz and I really only need 20-100hz. But the 15DS115 is said to be not good for ported (even if B&C themselves argue otherwise). What do you think? After that, I cant start experimenting with slopes again. I was suprised to see the highest order of subdriver manufacturers, Harbottle, advising on Buttersworth btw.
Bear with me i dont see the quote option in your post.
Im now using LR24dB. After some chat with Harbottle's designer im now testing Buttersworth agan, and a high pass filter (6db) at 25hz. I was contemplating that but didnt find any info on putting a high pass filter down low in a sub driver on any forum!
Either way, it sounds better but after all considerations I think im leaning towards a B&C 15DS115. Im also leaning towards switching towards a ported design instead of sealed. Ive seen many great B&C SW115 ported design but they go to 3k hz and I really only need 20-100hz. But the 15DS115 is said to be not good for ported (even if B&C themselves argue otherwise). What do you think? After that, I cant start experimenting with slopes again. I was suprised to see the highest order of subdriver manufacturers, Harbottle, advising on Buttersworth btw.
The 1st-order high-pass filter set to 25Hz would, of course, help reduce the power going to the subwoofer at very low frequencies. If you are using a vented-box enclosure that is tuned quite low, this filter helps reduce large cone excursions when infrasonic frequency content is present. The woofer would otherwise tend to flap around while producing no significant output. Under these circumstances, it is more useful to include a 2nd-order high-pass filter. Some amplifiers have one, and it is usually labelled as a "subsonic" filter. Note that the original design paper by Thiele back in the 1960s included vented-box designs with 2nd-order high-pass filtering that also had some peaking. Hence, high-pass filtering low-frequency loudspeakers is a well-known technique.
In its 8-ohm version, the B&C 15DS115 has Fs = 33Hz, a very low Qts = 0.17, and a quite small Vas = 94 litres for a 15-inch driver. Any standard vented-box alignment is going to result in a quite high value of Fb in a small enclosure, which is not a good combination for a subwoofer.
Below is a potential design for the 15DS115 in a 50-litre closed-box enclosure. With parametric EQ of +7dB at 25Hz with Q=1.50, we can get around 106dB SPL at Xmax for a peak power input of around 230W re 8 ohms. A 4th-order Linkwitz–Riley low-pass filter set to 80Hz is included, and produces a –6dB cut-off at around 100Hz. The second-order low-frequency roll-off will help to obtain a good transient response.
In its 8-ohm version, the B&C 15DS115 has Fs = 33Hz, a very low Qts = 0.17, and a quite small Vas = 94 litres for a 15-inch driver. Any standard vented-box alignment is going to result in a quite high value of Fb in a small enclosure, which is not a good combination for a subwoofer.
Below is a potential design for the 15DS115 in a 50-litre closed-box enclosure. With parametric EQ of +7dB at 25Hz with Q=1.50, we can get around 106dB SPL at Xmax for a peak power input of around 230W re 8 ohms. A 4th-order Linkwitz–Riley low-pass filter set to 80Hz is included, and produces a –6dB cut-off at around 100Hz. The second-order low-frequency roll-off will help to obtain a good transient response.
Last edited:
If a little bit more output level is desired, the parametric EQ in the previous closed-box design can be replaced with a 2nd-order high-pass peaking filter set to 26Hz and Q=2.0. This produces the following results. Here we have an F3 = 24.8Hz, which is a little higher than the 22.1Hz obtained in the previous design. Our maximum SPL is now 108dB, with a peak power required of 215W, for a nominal power input of 170W re 8 ohms. Of course, the low-frequency roll-off is now 4th-order, so the transient response won't be as good as that of the previous design.
Last edited:
50 liters. Amazing! How do i go about and build such a box? I remember from the Dayton kit it also needs internal stuff. But that kit I glued together in less than 20 minutes.
And do you think going for a sealed design is worth it? I read such good things about ported designs, even with thi DS115 but admittedly, most take the sw115. This DS has a much strnger motor though. People suggest either a horn based design or sealed. And sealed design with this DS give fantastic results. Ill think i give this a go.
A subsonic filter, that is a high-pass filter that is usually set to 25 Hz or lower. This filter is very useful to protect driver from over excursion. That in turn can lead to permanently damaged membrane or voice coil. This is a real issue with sealed, EQed subwoofers when playing loud.
I have set myself 2nd order high-pass filter for that purpose at around 20 Hz to my sealed 15" subs. When driven really loud, the membrane travels wildly, but has never shown any sign of stress.
I have set myself 2nd order high-pass filter for that purpose at around 20 Hz to my sealed 15" subs. When driven really loud, the membrane travels wildly, but has never shown any sign of stress.
Oh but I dont use it for protections, I just use it because lower the measurements go a bit wild. I never contemplated that it is actually even better to just kerb the output within the parameters and characteristics of the subject sub.
When I use chatgpt, it says 50 liters is a bit low. Others have commented on this as well. Wouldnt it be better to design a 100liter+ cabinet for this 15DS115?
When I use chatgpt, it says 50 liters is a bit low. Others have commented on this as well. Wouldnt it be better to design a 100liter+ cabinet for this 15DS115?
edit: chatgpt came up with this:
Subtract 2×19 mm (wall thickness) from each axis:
Internal volume:
522×470×394=96,517,080 mm3=96.5 L522 \times 470 \times 394 = 96,517,080\ \text{mm}^3 = 96.5\ L522×470×394=96,517,080 mm3=96.5 L
Add ~13.5 L from port and driver displacement →
Net volume ≈ 110 L
This is the internal depth of the slot port shelf (front-to-back), inside the box.""
I asked chatgpt to make a vented box, 38hz frequency tune. It came up with these measurements. Is it possible to run this somehow in winISD to verify?
| Feature | Value |
|---|---|
| Gross Volume | ~114 L |
| Net Volume | ~110 L (after port/driver) |
| Tuning Frequency | 38 Hz |
| Port Area | 330 cm² |
| Port Dimensions | 500 mm (W) × 66 mm (H) × 146 mm (L) |
| Port Location | Front-facing (bottom) |
| Wall Clearance | Only 25 mm (1 inch) needed behind the box |
| Panel | Size (mm) | Qty |
|---|---|---|
| Front/Back | 560 × 508 | 2 |
| Sides | 432 × 508 | 2 |
| Top/Bottom | 560 × 432 | 2 |
| Port Shelf | 500 × 394 | 1 |
| Port Side Walls | 66 × 146 | 2 |
🔹
Subtract 2×19 mm (wall thickness) from each axis:
- Internal Width: 522 mm
- Internal Height: 470 mm
- Internal Depth: 394 mm
Internal volume:
522×470×394=96,517,080 mm3=96.5 L522 \times 470 \times 394 = 96,517,080\ \text{mm}^3 = 96.5\ L522×470×394=96,517,080 mm3=96.5 L
Add ~13.5 L from port and driver displacement →
Net volume ≈ 110 L
🔊
- Port Width: 500 mm (nearly full box width)
- Port Height: 66 mm
- Port Area:
500×66=33,000 mm2=330 cm2500 \times 66 = 33,000\ \text{mm}^2 = 330\ \text{cm}^2500×66=33,000 mm2=330 cm2 - Target Tuning Frequency: 38 Hz
- Port Length: 146 mm
This is the internal depth of the slot port shelf (front-to-back), inside the box.""
I asked chatgpt to make a vented box, 38hz frequency tune. It came up with these measurements. Is it possible to run this somehow in winISD to verify?
The vented box is not going to be a good solution. The VituixCAD simulation is shown below using Vb=110 litres, Fb=38Hz, port area Ap=330cm^2, and port length Lp=13.3cm. The latter is reasonably close to your value.
This design has a dip of around 6dB at 70Hz. It could of course be equalized to be much flatter.
Below is the response with some parametric EQ applied (+5dB at 90Hz, . In my estimation, it's now reasonably workable as a solution for a subwoofer in a room. However, it's a less smooth response than I would normally like to see.
This design has a dip of around 6dB at 70Hz. It could of course be equalized to be much flatter.
Below is the response with some parametric EQ applied (+5dB at 90Hz, . In my estimation, it's now reasonably workable as a solution for a subwoofer in a room. However, it's a less smooth response than I would normally like to see.
What an amazing software you use. What is it?
And can you make me a ported/vented box with a tuning around 40 that does work for this driver?
this is what chatgpt commented on your measurements
Amazing isnt it?
He suggest 3 options:
1. raise freq tuning to 40 or 42
2. smaller box (95 liter), keep tuning at 38
3. reduce port area
And can you make me a ported/vented box with a tuning around 40 that does work for this driver?
this is what chatgpt commented on your measurements
📊
🔈
- Tuning frequency is confirmed at ~38 Hz (peak in port output).
- There is a notable ~6 dB dip around 70–75 Hz, as you mentioned.
- This is a classic sign of underdamping in the upper bass.
- Likely caused by a combination of:
- Large port area (331 cm²)
- Relatively low Q (box too “loose” acoustically)
- Large box volume (110 L)
📉
- Excellent cone control around tuning: excursion minimum near 38 Hz.
- Xmax (14 mm) is exceeded only below 30 Hz, so it’s well-behaved.
🌬️
- Port velocity peaks at ~21.8 m/s at tuning frequency — close to the 17 m/s comfort limit (red dashed line).
- This is acceptable but on the higher side.
- You may hear some chuffing at very high drive levels, but it’s not severe.""
Amazing isnt it?
He suggest 3 options:
1. raise freq tuning to 40 or 42
2. smaller box (95 liter), keep tuning at 38
3. reduce port area
Last edited:
Here’s a revised version of your box that should smooth the dip significantly:
The improved enclosure offers a much more even output in the critical 50–80 Hz range, which will sound punchier and better integrated in most real-world setups — especially near walls or boundaries.
Would you like:
| Parameter | New Value |
|---|---|
| Box Volume (net) | 100 L |
| Tuning Frequency | 40 Hz |
| Port Area | 275 cm² (e.g. 500 × 55 mm) |
| Port Length | Recalculate (≈ 180 mm) |
🔴original design
- Strong peak near tuning (~38 Hz)
- Clear dip around 70 Hz, ~6 dB down from the peak
- Smoother low-end extension, but less upper bass energy
🔵fixed config
- Flatter overall response with significantly reduced dip
- Better balance across 40–100 Hz
- Slightly less deep bass, but tighter and more musical response
✅
The improved enclosure offers a much more even output in the critical 50–80 Hz range, which will sound punchier and better integrated in most real-world setups — especially near walls or boundaries.
Would you like:
- The exact port dimensions for this improved design?
- Or a WinISD project file to experiment further yourself?
We need an edit: some guy at another forum discovered these calculations are based on wrong TS parameters. The DS115 has different specs.
edit 2: this is the spec sheet for the 8ohm version. Its vastly different than the 4 ohm version
Parameters
Fs33 Hz
Qes0.18
Qts0.17
Sd855 cm² (132.53 in²)
Xmax16.5 mm
Mms254 g
Le4.5 mH
Re4.9 Ω
Qms5.2
Vas94 dm³ (3.32 ft³)
η01.9 %
Xvar14 mm
Bl38.7 Tm
EBP183 Hz
edit 2: this is the spec sheet for the 8ohm version. Its vastly different than the 4 ohm version
Parameters
Fs34 Hz
Qes0.17
Qts0.16
Sd855 cm² (132.53 in²)
Xmax16.5 mm
Mms273 g
Le3.2 mH
Re3.2 Ω
Qms4.7
Vas83 dm³ (2.93 ft³)
η01.9 %
Xvar14 mm
Bl33.6 Tm
EBP200 Hz
Last edited:
Please, when you say based on the "wrong TS parameters", what does that "guy at another forum" actually mean? I simply looked up the 15DS115 8-ohm version's datasheet and plugged in the TS parameters that were supplied there. Which of the TS parameters did I input incorrectly?
Of course, I may have overlooked the minor fact that you potentially wanted to use the 4-ohm version of the 15DS115 driver. I say "minor", because its TS parameters don't really affect the tuning outcome to any great degree. I already used a calibrated eyeball to contemplate the differences between the two sets of TS parameters, before choosing the 8-ohn driver as being representative of either.
Below is a simulation of the 15DS115 4-ohm driver in a 100-litre enclosure and tuned to 40Hz. It's very similar to the previous results for the 15DS115 8-ohm driver.
Unfortunately, it appears that these two drivers are not well suited for low-frequency vented-box subwoofer applications. If you are happy to EQ the frequency dip a little, then they can be made to have a smoother response.
Last edited:
Im sorry, you didnt do anything wrong. ChatGPT did, it took the wrong parameters.
Ive got WInISD working and come to teh conclusion the 15DS115 isnt working as sub, either ported or sealed or perhaps only with heavy DSP.
I then went and tried the SW115 in 15" which went a lot better. Then I tried the 18" SW which went even better. ABout 160 liters ported box wold make teh SW work. Of course, its way way overkill but I see plenty of people using pro drivers trying as big a sdriver as they can, for instance in that devastator design, to have both good low and good mid bass.
Me, I would be happy with a 160liter or maybe a bit bigger and 18". Can you run this 18" B&C SW through your software? And maybe the 15"" too. What do you think about that driver?
Again, sorry for the misunderstanding. You have been very kind and helpfull guiding me through this forest. And im learning! Probably, in the end, maybe a pro driver isnt the best home solution but I should at least try and see if I like it or not. If not ill sell it and try something else. Its our hobby, in the end.
As an aside: what a wealth of knowledge in this forum. Almost any driver the DIYaudio subwoofer forum pops up in google. 🙂
Below is the simulation of the 18SW115 in a Vb = 160-litre enclosure tuned to Fb = 28Hz. Once the low-pass filter is added, the F3 should be around 28Hz or so.
If we keep the enclosure volume at Vb = 160 litres, we can also try a closed-box enclosure. If we apply parametric EQ of +10dB at 28Hz with Q=1.00, we get the response shown below. Peak power input is around 200W at Xmax. The F3 is around 28Hz, which isn't too bad, plus the roll-off rate below the cut-off frequency is only second order. The penalty we pay is in maximum output SPL, but there is still a maximum output capability of 113dB.
If we keep the enclosure volume at Vb = 160 litres, we can also try a closed-box enclosure. If we apply parametric EQ of +10dB at 28Hz with Q=1.00, we get the response shown below. Peak power input is around 200W at Xmax. The F3 is around 28Hz, which isn't too bad, plus the roll-off rate below the cut-off frequency is only second order. The penalty we pay is in maximum output SPL, but there is still a maximum output capability of 113dB.