Reviving Lost technology of the Old Republic

[Evil Twin]

Do not be misled by the speculative prattling of Jonmarsh... the technology of the Old Republic is well known to the core cognoscenti, and steps taken to revive it are much farther along than the pitiful efforts of nostalgic experimenters...


Just recently I found a cache of artifacts from the old days, and completed initial testing on a bandpass design created in the prime of the Old Republic...


SPL response



Impedance



For one skilled in the Dark arts, there are a multitude of potential solutions waiting to be unlocked, depending on how you prefer to fine tune the end solution- all of the next possibilities use the same rear chamber volume, but are tuned by changes in the front system design alignment. This allows a degree of voicing control not feasible with conventional sealed and ported cabinets... but note the inevitable interaction with sensitivity- either a blessing or a curse, depending on your choices.

The most important factor is the availability of improved technologies in both drivers and cabinet materials- it is completely possible that a new level of performance is now achievable.

Fb = 50Hz - This is tuned for the lowest bass extension with a slight peaking of the very low frequency response. But this compromises sensitivity. Note the F3 point.






Fb = 60Hz - This is a "nominal" configuration close to the theoretical nominal optimum for this driver, assuming a roughly flat response and good extension- note the F3 point.





Fb = 65Hz - this is a configuration that might be said to optimize the mid bass level and articulation, as well as overall sensitivity. Note the F3- still very good by music standards.




As you can see from the graphs, these tuning efforts also interact with the basic sensitivity- this is an inevitable characteristic of bandpass enclosures. Here, the span of change in sensitivity is 4 dB- all for the same driver and rear enclosure volume. The Bandpass enclosure is relatively unique in how much the enclosure takes control of the driver behavior....






\

[5th element]

Do you really only end up with 80-85dB Re 2.83V with a 97dB Re 2.83V driver? Or is the stimulus voltage in peak to peak rather than rms?

[Evil Twin]

This is one of the necessary tradeoffs to achieve this control and bandwidth in a compact enclosure- and in this case the driver is selected to achieve a compact enclosure- by virtue of the starting Qts necessary to achieve a target Qtc.

It is not an intuitive concept - I will share a different design example showing how a more conventional driver (ATC 15 E 102 25 10) - a similar type of no ripple tuning can be derived, but the rear enclosure volume is for a similarly optimized minimum ripple design is 190L, with a front enclosure volume of 60L, and nominal front "port/impedance converter" Fb of 39Hz. This produces a system sensitivity of 86 dB, from a driver with a nominal sensitivity of 94.5dB/2.83VRMS.





If placed in a smaller cabinet like the one proposed for the 15NW100, it will have 3 to 5 dB peaking and correspondingly degraded transient response.







The key is not the sensitivity, but the Qts. One could say that the sensitivity is a side effect of the BL product needed to achieve a low Qts- in fact, Qts in a range often not very useful in conventional cabinets.


Returning to the 15NW100, one can see that an even more compact enclosure for music applications with an F3 of 30Hz is entirely feasible:





Here, the rear chamber volume has been reduced from 75L to 55L.

It could be noted that compared with Jonmarsh's Wavecor Ardent, the the rear enclosure volume is only slightly greater, the sensitivity is higher, and the LF extension deeper. Undoubtedly this could form the basis of a three or 4 way music system. Testing such concepts is envisioned for the near future.


In a closed box of the same size, the peak response frequency is about 240Hz, the Fb 63Hz, and the F3 113Hz. At 63Hz the response is down -8dB.

It may be fitting to liken the power of the bandpass enclosure to shape and mold to that of the Force- great power, which if wielded unwisely by the untrained or unknowledgeable may have disastrous results.

That may be why they are so popular for competition car audio systems... and rarely seen elsewhere.

[Dave Bullet]

Dark Lord, forget those who do not see the potential of such a device. I can see this being used as a phase-shift disruptor to augment the pulse alignment of your Tie Fighter's blaster cannons.

[Evil Twin]

Dark Lord, forget those who do not see the potential of such a device. I can see this being used as a phase-shift disruptor to augment the pulse alignment of your Tie Fighter's blaster cannons.

Your perception of cross application potential is both remarkable and worrisome... this could constitute a serious military technology breach if word of this technique gets to the wrong quarters...

[5th element]

Haha sonic cannons. Tuning ports at the ready.

I had indeed noticed that the cabinet volume was unusually small considering the size of driver. I assumed it was a low Are beast, which as you say is usually difficult to use and are usually more suited to horns.

I guess this allows it to work in such a cabinet as above. It's almost like instead of using a low sensitivity subwoofer, small cabinet and lots of power, for deep bass, you're using the cabinet to modify a high sensitivity driver into a low sensitivity one with good extension. An acoustic impedance device as you said above.

Is this more akin to the low Qts driver giving you a severe extended bass shelf alignment when tuned low. Then the front chamber and port act like an acoustic low pass, rolling off the rising response (and extending the bass too) and thus giving you a low sensitivity system with reasonably flat response?

[Juhazi]

Has anyone else thought about how high room spl would be achieved without braking the poor driver? And what about vent turbulence making noise?

[Zvu]

I checked that. It handles everything pretty well. Port resonance is kinda low but Kef elastic port could be used to mitigate it - something old, something new, something borrowed, something blue

Only thing, i don't get the spl ET did - that is if WinISD is precise enough. When i get home i'll check what Vituix says.

https://i.postimg.cc/PxRxPBxh/TS.png

https://i.postimg.cc/F1tHx33B/spl.png

https://i.postimg.cc/766hPxqm/signal.png

https://i.postimg.cc/Znvq3VBN/port-noise-at-100-W.png

https://i.postimg.cc/7Lqbc3dv/cone-excursion.png

[Evil Twin]

Has anyone else thought about how high room spl would be achieved without braking the poor driver? And what about vent turbulence making noise?

Impressive, most impressive!

This one is strong in the Force of intuition-

A bandpass enclosure, properly designed, can fully harness the capabilities of a driver and do so with a wide range of options to shape its performance. However, it does place taxing demands on the driver parameters and driver integrity, because the forces involved on the voice coil and cone structures can be much higher than a conventional enclosure.

With the driver being literally buried within the enclosure, these stresses may not be audibly evident until one is past the point of no return.

As regards the port, you will have similar issues about port velocity and compression as for a conventional bass reflex, and the same design measures must apply. But consider well the interesting differences between the behavior of a bass reflex port and the front chamber and port for a bandpass enclosure....

Consider a typical driver for a ported alignment- the Satori WO24P. It performs well in 50L... about the same cabinet volume as the rear volume test cabinet that will be used for the next test build for the bandpass.






The double peak impedance curve approximately shows the boundaries of useful levels of output from the port- you can that based on level this is from just below 20Hz to a bit above 40Hz, approximately a 1 octave span. Most of the LF output below 40Hz comes from the port. The resulting system has a 4th order roll off, 24 dB/octave.


I will briefly revisit the proposed 55L rear chamber design which will be evaluated. Now the "port" is being driven at low frequencies by the front of the driver. (there are compound alignments which use ports additionally on the rear, but these have very steep roll off rates and wretched transient response).




Now the spread on the impedance peaks is quite different, though we still have in essence a double tuned system, with a common rear Fb and front Fb. Interestingly, the LF roll off rate is only 12 dB per octave, with the expected implications for group delay and transient response.


Why would you make this effort, compared with the WO24P-8 in a similar size enclosure? Both are driving a port of sorts at low frequencies.

Which driver has more Sdp? (Must I even pose this question? )

Which driver has more power handling? 90W versus 2kW?

Which driver will unload below the lower resonant peak, due to what has become a cabinet with a hole in it? Which driver is constrained from excess motion by being in a sealed box?


The Satori is more sensitive as a ported box reflex system, but the other system approach can reach SPL's of 105 dB with less than 100W.

While great store is placed on the ability to think outside of the box, perhaps it is necessary to remind ourselves of the value of creative and less orthodox thinking inside the box...

Of course, the final measure must be the results, and those are not determined yet for this specific test case. Especially, it will be interesting to look at the distortion results.

But as this new development proceeds, so far I must quote my former master:



Events are unfolding exactly as I have foreseen...


Then too, there is the more pragmatic and modern approach- nothing ventured, nothing gained.

Should it be of interest to members of the forum I can post a complete general purpose design process for this type (first order) of bandpass design.

[Dave Bullet]

Should it be of interest to members of the forum I can post a complete general purpose design process for this type (first order) of bandpass design.

Yes ET, I'd be interested in such information. In particular, whether or not particular T/S parameter values indicate better or worse driver candidates for a bandpass design (similar to EBP for vented, or Qts for TL or open baffle etc...).

I'd also be interested in benefits to group delay and excursion plots over the operating bandwidth compared to "same sized" vented and sealed alignments (all things being equal - which they never are, but hey). Whilst I'm at it, I wonder if bandpassing reduces breakup distortion at the top of the operating range?

[JonMarsh]

I checked that. It handles everything pretty well. Port resonance is kinda low but Kef elastic port could be used to mitigate it - something old, something new, something borrowed, something blue

Only thing, i don't get the spl ET did - that is if WinISD is precise enough. When i get home i'll check what Vituix says.

https://i.postimg.cc/PxRxPBxh/TS.png

https://i.postimg.cc/F1tHx33B/spl.png

https://i.postimg.cc/766hPxqm/signal.png

https://i.postimg.cc/Znvq3VBN/port-noise-at-100-W.png

https://i.postimg.cc/7Lqbc3dv/cone-excursion.png

Hmmm, I'm reminded of a quote a former colleague of mine and I have used a lot in the past (he has since moved on to his dream job...)

In God We Trust, all others we verify...

[Juhazi]

I understand the benefit of BP for eg. subbing desktop speakers, when spl requirement is low. But for hifi in a living room or HT use, we need high capacity without port noise. Alas a BP might be challenging to integrate in a multiway one-box speaker.
The Satori WO24 used in the sim has 8" cone, Sd 255cm2 paper cone and soft low-damping surround. How about using SEAS L26ROY or some more basic 10" woofer because high Xmax is not needed?

[Scottg]

I like to use band-pass over a much more narrow range, a "hump", to effect extension AND I like to do it with the port resonance at about the same freq. for the mid-bass's bass-reflex (port) tuning freq. (..which is usually a much lower freq. than standard "alignment"). ex. 45 Hz for both tuning freq.s, where the midbass is often -6db from the average (with the port's contribution), and of course where the "top of the hump" for the band-pass is near (usually slightly less than) that same spl (-6db from the average).

[Evil Twin]

That is precisely my earlier point- the bandpass concept can be purposed to many different end goals, with corresponding changes in technique- however, the mathematics and the driver select process to reach a specific goal may be somewhat non-intuitive and unforgiving.


I will later post a basic design synthesis example based around the concept development I am engaged in now- in the near term, though, my focus in the coming days is on hardware testing and a new cabinet design that is coming together...

[Scottg]

-speaking of that, rather like the band-pass here, Synergy/Unity Horns: use a small resonant acoustic low-pass inside a horn. ..and like here: it's not so much about the resonant behavior of the vent increasing output, rather about a compact design with an acoustic low-pass providing a nice usable pass-band to combine with other drivers.