• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

B&C 15DS115

Pieter1267

Active Member
Forum Donor
Joined
Nov 20, 2024
Messages
170
Likes
86
Hi

Im intrigued by the B&C DS115 pro sub driver. Im thinking about 2 in 2 ported boxes. I thought sealed config was better but several folks and even chatgpt said its fine in a ported box around 100liters.

I have dayton um22 and they lack a bit mid bass. Im not the first to say so. But i would like something that is punchy mid bass and good down low. I only use them for music and need some better techno output amongst other things. I reckon trade in some lower scale for that fantastic brute b&c ds115 motor force. I considered the sw115 but is probably not tight enought mid bass and more of the um22-15 i have now.

Anybody using the ds115? If so, what cabinet? For music: stay with 15 or is move to 18 a better balance? My room isnt big and feel 18 might be overkill. Aim is about 30-120hz usable range. 35-110 also ok.

Planning to feed it with inuke6000.

Oh, and im interested in letting chatgpt make a drawing and calculation for the cabinet.

Thoughts?
 
Thoughts?
You use widely accepted vocabulary as to describe subjective impressions. Question is, how it translates into real data, that one can use as a foundation for a decission, given the fact, that B&C uses just that data to advertise their product.

I see a deeper conflict when a target is decribed subjectively, and all means to get there are objective and for the time being unrelated to the former. Of course ChatGPT doesn't feel the conflict, the parrot it is. It is just talking along ;-)

I wonder if, using drivers like that, you're a pro, or if you tempt to listen to techno-edm in the wide open Atacama desert at a few yards distance?

In ernest, with every ported speaker box the port will exhibit tons of typical port noises, aka "chuffing" before the driver gets close to its limits. The port will limit the design, except the enclosure is big enough. Speaking of 300 liters or so.
 
You use widely accepted vocabulary as to describe subjective impressions. Question is, how it translates into real data, that one can use as a foundation for a decission, given the fact, that B&C uses just that data to advertise their product.

I see a deeper conflict when a target is decribed subjectively, and all means to get there are objective and for the time being unrelated to the former. Of course ChatGPT doesn't feel the conflict, the parrot it is. It is just talking along ;-)

I wonder if, using drivers like that, you're a pro, or if you tempt to listen to techno-edm in the wide open Atacama desert at a few yards distance?

In ernest, with every ported speaker box the port will exhibit tons of typical port noises, aka "chuffing" before the driver gets close to its limits. The port will limit the design, except the enclosure is big enough. Speaking of 300 liters or so.
Oh yes, chatgpt is like talking to some kind of overdoses on anti depressants person. It goes wichever way you want.

But at the moment its also just for experimentation. I want to learn more about subs, how it translates to home use and the current tech and drivers - various postings on this forum alone, claim at the moment ported subwoofers dont have the chuffing or uncontrolled or untight characteristics they had 10 years ago. Also, software evolved a lot. There are at least 4 or 5 programs like WinISD or Hornresp. I wonder how ChatGPT uses its calculations. Does it use the free programs on internet or is it able to use quality models?
 
Oh yes, chatgpt is like talking to some kind of overdoses on anti depressants person. It goes wichever way you want.

But at the moment its also just for experimentation. I want to learn more about subs, how it translates to home use and the current tech and drivers - various postings on this forum alone, claim at the moment ported subwoofers dont have the chuffing or uncontrolled or untight characteristics they had 10 years ago. Also, software evolved a lot. There are at least 4 or 5 programs like WinISD or Hornresp. I wonder how ChatGPT uses its calculations. Does it use the free programs on internet or is it able to use quality models?
If people believe so, then ChatGPT would simply replicate. It doesn't have any calculatory competences; results are taken from texts, which answer is most likely, measured against the internet mostly. It has been shown, see youtube, Sabine Hossenfelder.

ChatGPT is a language model, just language, with more or less(!) 'intelligence' than a parrot.

Anyway, WinIsd and such show clearly, in combination with results presented in the program "FlareIt", that the port is the limiting factor in most of contemporary designs. Even with gracefully rounded ends it will get into chuffing at about 10m/s. Be it or not, you'll find out with experimentation. Please report, it is always appreciated (fodder for our new parrot ;-)
 
this is from chatgpt:



From B&C's datasheet, typical parameters are:


  • Fs = 35 Hz
  • Qts = 0.29
  • Vas = 65 liters
  • Sd = 855 cm²
  • Xmax = 14 mm
  • Re = 5.6 Ω
  • Pe = 1700 W
  • Driver displacement ≈ 5 liters (estimated)


Step 2: Choose a Box Volume (Vb)​


Since you're tuning low (38 Hz) and this is a high-performance pro audio driver, a good box size is around:


Recommended box volume:​


Vb ≈ 100 – 120 liters
Let’s go with 110 liters net internal volume (before port & driver displacement).


Step 3: Calculate Port Dimensions​


We’ll use the cylindrical port formula to find port length for 38 Hz tuning:


Formula:​

L=(23562.5×D2)Fb2×Vb−(0.732×D)L = \frac{{(23562.5 \times D^2)}}{{Fb^2 \times Vb}} - (0.732 \times D)L=Fb2×Vb(23562.5×D2)−(0.732×D)
Where:


  • LLL = port length (in inches)
  • DDD = port diameter (in inches)
  • FbFbFb = tuning frequency (Hz)
  • VbVbVb = internal volume (in cubic inches)

Let's pick a port diameter:​


For an 855 cm² cone area, a 4–6 inch diameter port is minimal. To reduce chuffing, dual 4-inch ports or a single slot port is common.


Let’s use one 6" (15.24 cm) round port.




Convert volume to cubic inches:​

110 liters=6710 in3110\ \text{liters} = 6710\ \text{in}^3110 liters=6710 in3D=6"⇒D2=36D = 6" \Rightarrow D^2 = 36D=6"⇒D2=36
Now calculate:

L=23562.5×36382×6710−(0.732×6)L = \frac{{23562.5 \times 36}}{{38^2 \times 6710}} - (0.732 \times 6)L=382×671023562.5×36−(0.732×6)L=84825097156−4.392L = \frac{{848250}}{{97156}} - 4.392L=97156848250−4.392L=8.73−4.392=4.34 inchesL = 8.73 - 4.392 = 4.34\ \text{inches}L=8.73−4.392=4.34 inches


✅ Final Specs (Rounded):​


ComponentValue
Tuning Frequency38 Hz
Box Volume (Net)110 liters (3.88 ft³)
Port TypeRound
Port Diameter6 inches
Port Length~4.3 inches
Driver Displacement~5 liters
Gross Volume~115 liters (accounting for port + driver)


Optional Improvements:​


  • Use a slot port to reduce length and improve airflow.
  • Add flaring to the port to reduce noise (especially if high SPL).
  • Use damping on internal walls (but not blocking port)."



This does look impressive to me. Is it wrong?
 
From B&C's datasheet, typical parameters are:


  • Fs = 35 Hz
  • Qts = 0.29
  • Vas = 65 liters
  • Sd = 855 cm²
  • Xmax = 14 mm
  • Re = 5.6 Ω
  • Pe = 1700 W
  • Driver displacement ≈ 5 liters (estimated)
What I found in the data sheet of the B&C 15DS115, however, is quite different:

1747151088260.png


I played around a bit with the parameters from the data sheet in WinISD. But I can't get a reasonable BR tuning with it, the parameters lead to very small housings with low tuning, which always means extremely long ports if the diameter is to be appropriate.

I would choose a different driver.
 
I would first do a room measurement to find out if it’s not some room mode that is sucking the mid-bass. Next try to EQ the UM sub, either by DSP, by moving it around, or preferably, both.
 
I would first do a room measurement to find out if it’s not some room mode that is sucking the mid-bass. Next try to EQ the UM sub, either by DSP, by moving it around, or preferably, both.
I mean, isn't it dubious to state the demand in less well defined terms, and then try to fulfill the requeirements by just using technical data, that bear no relation to before mentioned terms?
Not the least, a beast like a 15" bass to be tuned to 38Hz for home use with, as was said, techno output?
I support the idea to get to terms with buying (cheap today) measurement equipment as to encircle the subjective problem first.

(And let ChatGPT dream on alone.)
 
I have the 21" version in a reasonable sized ported box, does the job, definitely agree mid bass slam etc at anything other than the most spirited home volume level is more about setup than the driver itself
 
What I found in the data sheet of the B&C 15DS115, however, is quite different:

View attachment 450637

I played around a bit with the parameters from the data sheet in WinISD. But I can't get a reasonable BR tuning with it, the parameters lead to very small housings with low tuning, which always means extremely long ports if the diameter is to be appropriate.

I would choose a different driver.
Thanks for correcting this! Good to see the chatgpt limitations. Its good but like MSO, you need to be very carefull what you put in.

Evenso, I talked to B&C itself and they say this;

"Almost our entire range of LF drivers is designed for ported enclosures, as sealed cabinets have very limited use in typical PA applications – which is our primary market. Therefore, we don't design our transducers with that type of enclosure in mind."

Thats odd, no?
 
What I found in the data sheet of the B&C 15DS115, however, is quite different:

View attachment 450637

I played around a bit with the parameters from the data sheet in WinISD. But I can't get a reasonable BR tuning with it, the parameters lead to very small housings with low tuning, which always means extremely long ports if the diameter is to be appropriate.

I would choose a different driver.
EDIT:

the 8ohm has Fs at 33. The 4ohm is 34. Does that make a difference? Edit 2: there are more differences.

This is the 4 ohm parameter

Parameters​

Fs

34 Hz​

Qes

0.17​

Qts

0.16​

Sd

855 cm² (132.53 in²)​

Xmax

16.5 mm​

Mms

273 g​

Le

3.2 mH​

Re

3.2 Ω​

Qms

4.7​

Vas

83 dm³ (2.93 ft³)​

η0

1.9 %​

Xvar

14 mm​

Bl

33.6 Tm​

EBP

200 Hz​



this is the 8ohm version

Parameters​

Fs

33 Hz​

Qes

0.18​

Qts

0.17​

Sd

855 cm² (132.53 in²)​

Xmax

16.5 mm​

Mms

254 g​

Le

4.5 mH​

Re

4.9 Ω​

Qms

5.2​

Vas

94 dm³ (3.32 ft³)​

η0

1.9 %​

Xvar

14 mm​

Bl

38.7 Tm​

EBP

183 Hz​

 
the 8ohm has Fs at 33. The 4ohm is 34. Does that make a difference? Edit 2: there are more differences.
Parameters will vary like that, once you model it then typically they will come out basically the same (The only woofer I can remember seeing come out v different was the BMS 18n862)
 
Thats odd, no?
not really, PA market is not generally going to mind some big boxes with large ports. Say it's a 50cm cube with a slot port across the width which yields a 500cm2 CSA so now you need something like an 80cm long port to hit their stated 40Hz tune which means the port runs to the rear and most of the way up the back (probably you make the box slightly taller to accommodate this). Now you have you simple to build box with a large port capable of taking ~1200W, keeping particle velocity largely in check and banging out ~125dB in its passband.

it's just an example, you can skin this however you like using freely available tools (e.g. hornresp)
 
What matters when dimensioning an enclosure are essentially the values for fs, Qts and Vas. And the 4 Ohm version doesn't differ enough from the 8 Ohm version to change my Assessment.

I wonder what kind of application the 15DS115 is intended for....

*edit':
Ein DSP gibt natürlich gewisse Freiheiten.
 
Very low Qts, high BL: probably would do well in a horn. Back loaded for instance.
Fun fact: in case the speakers are used remotely close to their capabilities at home in a small room the listener wil be deafend within seconds for a while, considering there are adequate mids and treble present with techno output, as has all been said. Please consider to add some 12dB reserve in order to overcome the temporal (?) hearing loss.

ChatGPT didn't know, but now you do.

A 'horn' shows so much resonances, that one might be tempted to design it so that its resonances counteract those from the room/positioning (Predicted In Room Response, PIR of the spinorama).

The LLM (Large Language Model) replicates common wisdom from the internet and such. Creativity isn't its stronghold, neither a critical reflection on what is said, if that isn't the same to begin with. Stick with it? Your choice.
 
What matters when dimensioning an enclosure are essentially the values for fs, Qts and Vas. And the 4 Ohm version doesn't differ enough from the 8 Ohm version to change my Assessment.

I wonder what kind of application the 15DS115 is intended for....

*edit':
Ein DSP gibt natürlich gewisse Freiheiten.
Danke!

Some guys at diysudio said the ds115 looked good in sealed box. How is that with you?

Codys Harbottle suggested to look at the Ciara 15.00. What software did you use to model it? I want to try myself.
Fun fact: in case the speakers are used remotely close to their capabilities at home in a small room the listener wil be deafend within seconds for a while, considering there are adequate mids and treble present with techno output, as has all been said. Please consider to add some 12dB reserve in order to overcome the temporal (?) hearing loss.

ChatGPT didn't know, but now you do.

A 'horn' shows so much resonances, that one might be tempted to design it so that its resonances counteract those from the room/positioning (Predicted In Room Response, PIR of the spinorama).

The LLM (Large Language Model) replicates common wisdom from the internet and such. Creativity isn't its stronghold, neither a critical reflection on what is said, if that isn't the same to begin with. Stick with it? Your choice.

And yet still i was blown away how much data it could produce. Shouldnt you just look at it as mso, winisd, hornresp: they all seem to have in common they only work as well as the correct data you put in.

I had fun trying to correct it though. It took 5 more tries before it started with correct parameters. Im glad guys like you correct me on the jubilations.
 
Fodder for the expert base: tune your Dayton woofers to 25Hz with a 4" port, stuff the enclosure with wadding, densly, leaving the area around the inner end of the port free. You've got a damped / lossy vented box with peculiar, but beneficial properties.

That's something to work with, getting into room equalizing. The speaker would not be the problem, never ever. Sure, there's lots of talking.
 
Last edited:
Shouldnt you just look at it as mso, winisd, hornresp: they all seem to have in common they only work as well as the correct data you put in.
No, it's apples and oranges. WinISD or Hornresp will give the same answer every time, provided you give them the same inputs. You can trust that if your input is correct, the model is used properly and the output it also correct.

With ChatGPT the output is not necessarily deterministic. You can ask these LLMs to do the same thing twice (same input) and get different responses each time (different output). So, you can't trust that your output is definitely correct if your input is.

I think for technical calculations like this, ChatGPT is a poor fit, but it would probably be good at helping you walk through using one of the non-AI tools to calculate stuff.
 
With ChatGPT the output is not necessarily deterministic.
As already said, Dr. S. Hossenfelder provides valuable input on that. You may find a plethora of AI related clips on youtube, searching for "Hossenfelder", "AI". Rarely I say that, but trust me. Your exsisting drivers are ready for the job, they only need your support (equalizer). CU
 
Back
Top Bottom