Only one CNC operation? - perhaps, but unless you're aware of new nano-bot router that can carve the inside of that spiral, (and how would you cast, then remove the form if one solid piece) it would need to be a mated pair of moulds - so one "operation" per side. I think the cost of material would pale to that of run time on the appropriate 5-axis router. Carving the outside of relatively small 3D shapes is not all that complicated, but the inside is a different story.
Regardless of one's opinion of their sonic merits / bank for buck ( Euro, etc), I think it's fair to say the production design engineers involved in the B&W Nautilus series have the support of sufficient financial resources to select from any current manufacturing technologies an materials, and there must be a reason they fabricate them as they do. In fact I give them as much credit for making the idea of this series a commercially viable reality -albeit at a premium price- as the acoustic designers for the "what if " concept.
Regardless of one's opinion of their sonic merits / bank for buck ( Euro, etc), I think it's fair to say the production design engineers involved in the B&W Nautilus series have the support of sufficient financial resources to select from any current manufacturing technologies an materials, and there must be a reason they fabricate them as they do. In fact I give them as much credit for making the idea of this series a commercially viable reality -albeit at a premium price- as the acoustic designers for the "what if " concept.
Last edited:
One Step forward, two steps backwards.
Step forward - my CAD design needed no alterations on the CNC router. Worked like a charm.
Two steps back - Given some sound advice and realised that there is no point make something that looks great but sounds awful. Like everyone, I don't have money to burn.
XRK197, could I ask you to help me with the akabak sims using my current design. And advice what modifications are required to my design.
Step forward - my CAD design needed no alterations on the CNC router. Worked like a charm.
Two steps back - Given some sound advice and realised that there is no point make something that looks great but sounds awful. Like everyone, I don't have money to burn.
XRK197, could I ask you to help me with the akabak sims using my current design. And advice what modifications are required to my design.
Attachments
Not a bad end result!
In plywood it could look awesome!
Where did you find this?
I need the dimensions of the snail (dia and length when unrolled and is it linear expansion?). I need driver model or TS params. I need to know what your goal with the nautilus TL is. Is it to absorb the back wave or to enhance bass? You won't get bass enhancement with a closed tail.
I can assume a driver if I know what is size class you are planning to use. Looks like a 4 inch class driver. If you want bass enhancement with a spiral, the end needs to be open. Or flip it around and have it expand to a horn mouth with driver in a chamber with throat leading to small end of spiral in middle. You basically have a single channel Cornu.
The Snail dimensions when unrolled are –
1721.808581mm Length (in its current state, though this would increase by up to 250mm(Est.) if a 10” driver installed.
Mouth Diameter – 6” currently
Volume 43.91L
To start off, I am going to keep to the closed design.
I didn’t realise that the transmission line was only applicable to open ported speakers?
Though it does pose the question, if I do open the tail, how large does a port have to be? Is there a ratio?
I’m looking for an 8” Full range driver, though I am finding it difficult to find one for a sealed unit.
Any recommendations are welcome.
I have had a closer look at my catalogue of music I listen to and what I do to the EQ’s. On reflection, I generally like to increase the frequencies from 500Hz to 16KHz incrementally. And I also do the same on the other end of the specrum, 64Hz and 32Hz.
So if you imagine, I create a U shape on the EQ.
This I guess is an oxymoron. As a sealed/full range speaker will not provide the base, though it will give me everything else. Thus a subwoofer may become an addition.
Also I have just realised that this is a prototype. If I wanted to change it to Sealed, open ported/transmission line, horn(Cornu) etc. then I can. (But this is something to think about later!)
PS, I have been speaking to Daniel who is the guy who made the Loud snail. Really nice chap.
1721.808581mm Length (in its current state, though this would increase by up to 250mm(Est.) if a 10” driver installed.
Mouth Diameter – 6” currently
Volume 43.91L
To start off, I am going to keep to the closed design.
I didn’t realise that the transmission line was only applicable to open ported speakers?
Though it does pose the question, if I do open the tail, how large does a port have to be? Is there a ratio?
I’m looking for an 8” Full range driver, though I am finding it difficult to find one for a sealed unit.
Any recommendations are welcome.
I have had a closer look at my catalogue of music I listen to and what I do to the EQ’s. On reflection, I generally like to increase the frequencies from 500Hz to 16KHz incrementally. And I also do the same on the other end of the specrum, 64Hz and 32Hz.
So if you imagine, I create a U shape on the EQ.
This I guess is an oxymoron. As a sealed/full range speaker will not provide the base, though it will give me everything else. Thus a subwoofer may become an addition.
Also I have just realised that this is a prototype. If I wanted to change it to Sealed, open ported/transmission line, horn(Cornu) etc. then I can. (But this is something to think about later!)
PS, I have been speaking to Daniel who is the guy who made the Loud snail. Really nice chap.
The problem with this kind of design in general is that it has no large baffle. This means that in general there will be little bass. You have to correct this by a filter network or an active equalizer. If you do not have much experiences with crossover design I would suggest using an equalizer in the beginning. With active equalizing even bass from a closed enclosure is not a problem, if the driver itself has the needed parameters (large diameter, large Xmax).
Sim of Snail with FR13
Here is an AkAbak sim of the 173 cm long Snail TL. I assumed a rectangular cross section starting at 6 in (H) x 6 in (W) and tapering to a 1.5 in (H) x 6 in (W). I used moderate stuffing inside the whole line and speaker is positioned 36 in from back wall, 40 in high above floor, and depth of cabinet is assumed to be 16 in deep. The Visaton FR13 is a 6.5 in class high Qts driver (Qts=1.25).
Here is the SPL vs Freq for the sealed tail end:
Here is the SPL vs Freq for the vented tail end (vent exiting to the side), black is overall output, red is driver direct radiation, green is vent output:
As you can see, the sealed version has no bass, and the vented version has bass but it is very peaky and rough, and extension is not very deep. Switching to a low Qts driver doesn't help much. Given these results, I am not sure if I would proceed with building this speaker.
Here is an AkAbak sim of the 173 cm long Snail TL. I assumed a rectangular cross section starting at 6 in (H) x 6 in (W) and tapering to a 1.5 in (H) x 6 in (W). I used moderate stuffing inside the whole line and speaker is positioned 36 in from back wall, 40 in high above floor, and depth of cabinet is assumed to be 16 in deep. The Visaton FR13 is a 6.5 in class high Qts driver (Qts=1.25).
Here is the SPL vs Freq for the sealed tail end:
Here is the SPL vs Freq for the vented tail end (vent exiting to the side), black is overall output, red is driver direct radiation, green is vent output:
As you can see, the sealed version has no bass, and the vented version has bass but it is very peaky and rough, and extension is not very deep. Switching to a low Qts driver doesn't help much. Given these results, I am not sure if I would proceed with building this speaker.
Attachments
Really doubt that in a sealed system any particular Qms, Qes, or Qts has an effect on the sound at all. In a sealed system it all boils down to Qtc, in which .707 is usually the goal without simulations.
If there is a single t/s parameter that has an effect on sound quality other than the obvious Fs, Sd, and Sensitivity, in my humble opinion it would be Le. The lower the inductance, the less energy is stored.
Anyway, from what I understand the goal of this design is essentially make a sealed enclosure with a transmission-line like interior space to avoid acoustic energy reflection into the back of the cone.
My question is, has any consideration of using polyfill been given? The most obvious way to reduce reflections onto the back of the cone is to absorb energy. By the general rule of thumb, that audio enthusiasts have come up with, 1lb per 1 cu. ft. of polyfill is optimal.
I think it's worth noting that I've used Acousta-Stuf vs regular polyfill, and highly suggest regular polyfill. Acousta-stuff was irregular in its density and had a distinct chemical smell. Whereas regular polyfill is extremely even in density, has no smell, and is actually sold as hypoallergenic pillow stuffing.
Hope this helps!
If there is a single t/s parameter that has an effect on sound quality other than the obvious Fs, Sd, and Sensitivity, in my humble opinion it would be Le. The lower the inductance, the less energy is stored.
Anyway, from what I understand the goal of this design is essentially make a sealed enclosure with a transmission-line like interior space to avoid acoustic energy reflection into the back of the cone.
My question is, has any consideration of using polyfill been given? The most obvious way to reduce reflections onto the back of the cone is to absorb energy. By the general rule of thumb, that audio enthusiasts have come up with, 1lb per 1 cu. ft. of polyfill is optimal.
I think it's worth noting that I've used Acousta-Stuf vs regular polyfill, and highly suggest regular polyfill. Acousta-stuff was irregular in its density and had a distinct chemical smell. Whereas regular polyfill is extremely even in density, has no smell, and is actually sold as hypoallergenic pillow stuffing.
Hope this helps!
Qt, Fs, & Vas are the critical parameters for simulating a sealed box.
Without damping xrk's sims are VERY hard to interpret. The whole point of the 1/2 w line is to absorb all the backwave, to completely remove the ripple/ A simple sealed sim based on the box volume will tell you more.
dave
There is stuffing in the sims - hard to relate to lbs/cu ft as the implementation in AkAbak is as an acoustic resistance in units of Pa/m^3. I use a value that based on measurements compared to the sims is probably more like 0.25 to 0.5 lb/cu ft. I will try a lower Le driver and see how that changes things. Certainly the efficiency of the driver shows up. What surprised me was how ineffective the spiral taper was at absorbing the reflected energy. Maybe higher density of stuffing is needed. But then you might as well use straight rectangle and stuff it like an aperiodic line.
Last edited:
Exactly. Maybe i wasn't entirely clear
What i meant is that if speaker driver A and speaker driver B were exactly the same except for variations in their Qts, Qes, and Qms, they would sound exactly the same aslong as their final (sealed) enclosure gave them the same Qtc, and the baffle of the final enclosure was the same.
I don't know of any reference that refutes this point. I was just trying to clarify that given the same Qtc-Qts, Qes, and Qms aren't the best t/s parameters to look at when choosing a speaker driver on the internet. You're much better off looking for a speaker driver off other aspects, such as voice coil inductance.
Last edited:
With sufficient dampint there would be no ripple. Line length needs to be at least half-wavelength of the resonant peak.
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.