Small 2-way speakers with linear on-axis and power response characteristics (Scan Speak and SB Acoustics drivers). H&V off-axis measurements included

[Inertiaman]

Still, the idea is to add one panel on the back not necessarily just 18mm, 22mm (or more) works too,
Then relocating the ports might work even if it means cutting a bit of the plastic flange.

If the objective is only to enable a longer port, you don't necessarily need to add material on the entire back panel. You can add material only at the port, to offset it backwards.

I imagine a ~ 100mm diameter x 24mm thick disc of MDF or baltic birch, with an inner circle cutout for the port O.D., maybe outside edge chamfered or radiused (only for cosmetic reasons). Glue it on the back at an appropriate place for the (eventually) chosen port.

The rear of the speaker may look unconventional with a "bulge" of sorts, but not necessarily ugly (for my tastes anyway).

 

[Chris*42]

I could not find a viable solution with the SB PRs (their fs are too low).
However I think I found ways that could work out for those who already built the speaker with minimal change.

@XMechanik, please correct me if I am wrong but the 8L specified seems to be the internal volume of the cabinet without removing the volume occupied by the Driver/Port/Xover. A rough estimation tells me that about 0.6L should be removed

#1 idea is to

1. use two Monacor ports extended to their maximum length (210mm)
2. double the thickness of the front panel (glue another 18mm baffle on the existing one), it can also help with the vibrations of the cabinet.
It is critical to significantly increase the diameter driver hole in the existing from baffle to avoid resonances
The volume added this way should be partially to compensating for the added port.
The internal box depth of the cabinet is 227mm and the front baffle is 18mm thick, with a port 210mm long this only provides (227-210+18 =) 35mm clearance from the wall.
This is not enough, but by doubling the front baffle thickness we get (35+18 = ) 53mm which is still a bit short but better.
3. optional, chamfer or round over the now 36mm thick baffle to maybe decrease the the diffraction of the cabinet (smoother response)

The ports are not fully meeting the targets but close enough and should provide ample headroom at anything but the maximum SPL capabilities of the driver...
View attachment 367293
#2 idea is to

1. use one MBR-50 Monacor port extended to 224mm (MBR-50, internal diameter 51mm, length 150 to 280 mm, Sv=20,4 cm2)
2. double the thickness of the front panel (glue another 18mm baffle on the existing one), it can also help with the vibrations of the cabinet.
It is critical to significantly increase the diameter driver hole in the existing from baffle to avoid resonances
The volume added this way should be partially to compensating for the added port.
The internal box depth of the cabinet is 227mm and the front baffle is 18mm thick, with a port 224mm long this only provides (227-224+18 =) 21mm clearance from the wall.
This is not enough, but by doubling the front baffle thickness we get (21+18 = ) 39 mm which is still a bit short but better.
3. optional, chamfer or round over the now 36mm thick baffle to maybe decrease the the diffraction of the cabinet (smoother response)

The port is not fully meeting the targets but close enough and should provide ample headroom at anything but the maximum SPL capabilities of the driver...
View attachment 367294

If I were to start from scratch I would redraw the box to have indeed 8L net tuned to 51-53Hz but with a larger port that meets the 5% of speed of sound criterion even at max excursion.
The cabinet would have to be larger, the depth would definitely need adjusting to both accommodate the longer port and achieve the 8L target.
Height (increased towards the "bottom") could also be adjusted if required be adjusted with minimal impact on the response.
The port surface area is the minimal viable value to meet the 5% of speed of sound criterion that would need to be adjusted to the nearest standard PVC pipe and the length recalculated.
View attachment 367292
Or alternatively use the same MBR-50 port with 8L at +/-52Hz
View attachment 367296
Side note, it simple to understand with the execution simulations why I always add a high pass filters to my EQ to prevent bottoming the driver at high SPL... Q.E.D.

I could not find a viable solution with the SB PRs (their fs are too low).
However I think I found ways that could work out for those who already built the speaker with minimal change.

@XMechanik, please correct me if I am wrong but the 8L specified seems to be the internal volume of the cabinet without removing the volume occupied by the Driver/Port/Xover. A rough estimation tells me that about 0.6L should be removed

#1 idea is to

1. use two Monacor ports extended to their maximum length (210mm)
2. double the thickness of the front panel (glue another 18mm baffle on the existing one), it can also help with the vibrations of the cabinet.
It is critical to significantly increase the diameter driver hole in the existing from baffle to avoid resonances
The volume added this way should be partially to compensating for the added port.
The internal box depth of the cabinet is 227mm and the front baffle is 18mm thick, with a port 210mm long this only provides (227-210+18 =) 35mm clearance from the wall.
This is not enough, but by doubling the front baffle thickness we get (35+18 = ) 53mm which is still a bit short but better.
3. optional, chamfer or round over the now 36mm thick baffle to maybe decrease the the diffraction of the cabinet (smoother response)

The ports are not fully meeting the targets but close enough and should provide ample headroom at anything but the maximum SPL capabilities of the driver...
View attachment 367293
#2 idea is to

1. use one MBR-50 Monacor port extended to 224mm (MBR-50, internal diameter 51mm, length 150 to 280 mm, Sv=20,4 cm2)
2. double the thickness of the front panel (glue another 18mm baffle on the existing one), it can also help with the vibrations of the cabinet.
It is critical to significantly increase the diameter driver hole in the existing from baffle to avoid resonances
The volume added this way should be partially to compensating for the added port.
The internal box depth of the cabinet is 227mm and the front baffle is 18mm thick, with a port 224mm long this only provides (227-224+18 =) 21mm clearance from the wall.
This is not enough, but by doubling the front baffle thickness we get (21+18 = ) 39 mm which is still a bit short but better.
3. optional, chamfer or round over the now 36mm thick baffle to maybe decrease the the diffraction of the cabinet (smoother response)

The port is not fully meeting the targets but close enough and should provide ample headroom at anything but the maximum SPL capabilities of the driver...
View attachment 367294

If I were to start from scratch I would redraw the box to have indeed 8L net tuned to 51-53Hz but with a larger port that meets the 5% of speed of sound criterion even at max excursion.
The cabinet would have to be larger, the depth would definitely need adjusting to both accommodate the longer port and achieve the 8L target.
Height (increased towards the "bottom") could also be adjusted if required be adjusted with minimal impact on the response.
The port surface area is the minimal viable value to meet the 5% of speed of sound criterion that would need to be adjusted to the nearest standard PVC pipe and the length recalculated.
View attachment 367292
Or alternatively use the same MBR-50 port with 8L at +/-52Hz
View attachment 367296
Side note, it simple to understand with the execution simulations why I always add a high pass filters to my EQ to prevent bottoming the driver at high SPL... Q.E.D.

Just as an example, there are many other PRs on the market:

The SB15SFCR (Sd = 178 cm2, Fs = 21 Hz, Mms = 62 g) can be tuned to the desired Helmholz frequency of 45 Hz in a cabinet of 8 l with an additional mass of ca. 16 g.
But this 5" x 8" PR has to be placed on the left or right side panel, this would result in a little bit funny design! But I´m sure there are solutions with PRs mounted on the rear panel.

 

[differenceclouds]

If I were to modify the shape of the cabinet, what should stay constant? The internal volume? The volume + depth? Thinking about doing something a bit trapezoidal like certain 80s boutique speakers.

 

[Inertiaman]

If I were to modify the shape of the cabinet, what should stay constant? The internal volume? The volume + depth? Thinking about doing something a bit trapezoidal like certain 80s boutique speakers.

If you want the response to behave identically to the tested speaker, you need to maintain the front baffle dimensions. You also want to maintain the same volume. Depth can be changed +/- to adjust volume for other variables (material thickness, port dimensions, etc).

 

[Maiky76]

Just as an example, there are many other PRs on the market:

The SB15SFCR (Sd = 178 cm2, Fs = 21 Hz, Mms = 62 g) can be tuned to the desired Helmholz frequency of 45 Hz in a cabinet of 8 l with an additional mass of ca. 16 g.
But this 5" x 8" PR has to be placed on the left or right side panel, this would result in a little bit funny design! But I´m sure there are solutions with PRs mounted on the rear panel.

Sealed/closed that always works but the LF response takes a big hit (-6 to 8dB below 100Hz) and you'll still need a high-pass filter for power handling if a sub is added.

As mentioned previously ~52Hz would be my personal target but that's just... me.
By my calculation this PR with no added mass and 8L yields a 52Hz tuning frequency but is overdamped (lower output) compared to the dual MBR35.

If the volume remains constant, i.e. circa 7.4L, then the PR with no added mass would result in a 54Hz tuning frequency and with 16g yields about 48Hz, one would need ~27g for 45Hz.
However, both tunings are overdamped compared to a single MBR35 but with more linearity that is for sure.
The caveat is that the net volume of the speaker would be further decreased with the addition of the PR and the internal damping would have to be beefed up.

The bottom line is DIY! There are variations that offer different tunings, sealed, single and dual MBR35, single MBR50 PR etc.
Those interested in building the speaker should find for themselves what works best.

BTW Winisd is a simple tool to perform the sims but it's windows only.
I use my tools because I have full control over the approximations/variables/hypothesis used for the sim.

 

[torgeirs]

Gongrats with simulation program and result!
The DI is superb
I wondered about the 15 ohm resistor in series with the tweeter. See it is only 5 watts, but it will take much of the hf energy.
Is it getting warm at all when playing loud?

Have a feeling this constuction could be ported to a mixed solution with dedicated hf amp. The amp could be low watts if 15 ohm resistor was ommitted. The first cap could be put in front of the amp instead. With much smaller value of cource

 

[Rick Sykora]

He's the designer of XMachina, a great automatic passive crossover designer. Does the AI even know why it does what it does?

About XMachina

XMachina is a passive crossover designer. XMachina works unassisted. Just tell XMachina your goals, load your drivers characteristics and ...

xmachina-ai.blogspot.com

Seems a lot like VituixCAD’s optimizer app. Wonder how they compare?

 

[dcibel]

Not the same at all. VituixCAD optimizer simply adjusts component values in an iterative process to try and match a target curve. Circuit topology is completely up to the user. Xmachina physically creates the crossover schematic through it's process.

 

[Rick Sykora]

Not the same at all. VituixCAD optimizer simply adjusts component values in an iterative process to try and match a target curve. Circuit topology is completely up to the user. Xmachina physically creates the crossover schematic through it's process.

sounds pretty cool, but one is a CAD tool and as the name suggests, Optimizer is an optimization tool for an existing design. There is overlapping functionality. Sounds like Xmachina is the better tool to create the design and (as the OP did), move it to VituixCAD later.

 

[dcibel]

Of course, as shown in post 1 of this thread

 

[Rick Sykora]

Of course, as shown in post 1 of this thread

Yep, that is how, now I understand why. Thanks!

 

[XMechanik]

Following is a system spl characteristic comparison from Klippel NFS (red) and gated/NF with VituixCad crossover simulation (green).


I expected 330Hz resonance to show up on NFS (as gated measurements has some smoothing effect) but it didn't.

The discrepancy below 300Hz is probably a result of not using baffle diffraction on the NF measurement. I neglected it due to the small enclosure dimensions, expecting a "single dB" error (post #4). Apparently the effect is more prominent, at least 2.5 dB as I read it from the chart.

Below 60Hz both characteristics get closer together which is a bit strange. It seems that lack of diffraction is not the only issue with my data processing. The effect occurs where port response becomes strong (near the BR tuning frequency) so one explanation that comes to mind is that the scaling factor of the BR port response did not exactly match the NF measurements performed.

 

[D!sco]

Did you have to match the port response "by eye"? That's always the most guesswork in VCAD for me.

 

[XMechanik]

When it comes to the enclosure and its tuning, the HxWxD dimensions have been chosen so that the inner resonant frequencies do not overlap.


The BR tuning procedure wasn't very advanced, the tuning frequency was selected to obtain the highest level near 40Hz in the simulations, even at the expense of a slight sloping of the characteristics starting above 100Hz. Following is the comparison of simulation and Klippel NFS measurement (the simulation looks even a bit better than the result of my NF measurements ).


The port diameter of 3.5cm seems a bit too small indeed (post #11) and I was considering replacement. Apparently the volume levels I was using were low enough that I eventually forgot about it.

 

[XMechanik]

Did you have to match the port response "by eye"? That's always the most guesswork in VCAD for me.

I think it's calculated automatically after entering the port and cone diameters.

 

[S=klogW]

When it comes to the enclosure and its tuning, the HxWxD dimensions have been chosen so that the inner resonant frequencies do not overlap.
View attachment 368191

The BR tuning procedure wasn't very advanced, the tuning frequency was selected to obtain the highest level near 40Hz in the simulations, even at the expense of a slight sloping of the characteristics starting above 100Hz. Following is the comparison of simulation and Klippel NFS measurement (the simulation looks even a bit better than the result of my NF measurements ).
View attachment 368192

The port diameter of 3.5cm seems a bit too small indeed (post #11) and I was considering replacement. Apparently the volume levels I was using were low enough that I eventually forgot about it.

Would you be interested in running your sims with either two MBR-35 per enclosure or one MBR-50, please? Also, I would be very interested in a variant of the Mechano23 which had the higher 52Hz port tuning. I believe this will increase the speaker sensitivity from ~100Hz upwards and then the bass frequencies could be handed off to a sub-woofer. I do hope this endeavour is of interest to you and thank you for your generous work democratising top-perfoming loudspeakers.

 

[Rick Sykora]

Here is a hybrid crossover with a lower parts count and eeks out a bit more sensitivity too...

Yes, this one does not tamp down as much on the woofer resonance but at 24 dB down, am quite sure that it is not audible.

Thanks @XMechanik for sharing your design! Was surprised that the tweeter's response raw response is uglier than I expected but you tamed it well.

 

[XMechanik]

Gongrats with simulation program and result!
The DI is superb
I wondered about the 15 ohm resistor in series with the tweeter. See it is only 5 watts, but it will take much of the hf energy.
Is it getting warm at all when playing loud?

Have a feeling this constuction could be ported to a mixed solution with dedicated hf amp. The amp could be low watts if 15 ohm resistor was ommitted. The first cap could be put in front of the amp instead. With much smaller value of cource

There should be no problem as long as you don't drive it with 14V 3kHz constant amplitude sine signal (10V RMS).


Power dispatching simulations with is "Pink noise" and "M-noise" signals (which are more realistic pictures of normal use) show there is a significant power headroom.

 

[differenceclouds]

Could the 8ohm model woofer (SB13PFCR25-8) be used with the proper resistor in front of the same tweeter, or would this change other characteristics too much?

 

[torgeirs]

There should be no problem as long as you don't drive it with 14V 3kHz constant amplitude sine signal (10V RMS).
View attachment 368212

Power dispatching simulations with is "Pink noise" and "M-noise" signals (which are more realistic pictures of normal use) show there is a significant power headroom.
View attachment 368214 View attachment 368213

Should not be any worries then.
What do you think about a dual amp solution where the C-R-L part is made into a sallen key RC filter with between -8 to -12 dB gain?
The amp could be about 1/10 of the watts compared to the LF amp.
The 2 uF could be there at speaker protection/ 6dB oct filter + easier to make 2. order active filter than 3. order active filter