Bag End Infrasub, the latest subwoofer scam?

[gnarly]

Can anyone explain how, in practical terms, an integrator differs from a first-order low-pass filter in terms of frequency response?

IMNSHO, Integrator is pure marketing jabber...
A filter is a filter, be it analog or digital.
If they flatten sealed natural rolloff, with a 12/oct dB analog boost, and then use frequency dependent dynamitic limiting to prevent over excursion the boost will certainly cause....
.....the resultant excursion limited max SPL vs frequency, will have the same magnitude roll off and phase response, as simply applying a Linkwitz-transform or shelving or PEQ or whatever boost, with the appropriately constructed hpf to limit the boost's over excursion,
No free lunch, no new tech.

There is one cool thing about "flatten with dynamic limiting" though...
for small signal (and I do mean small), the very low end can be heard without phase rotation. ( ie before dynamic limiting kicks in)

 

[Flaesh]

over excursion

Just install as many subwoofers as you need. Quite realistic for home hi-fi.

 

[gnarly]

Just install as many subwoofers as you need. Quite realistic for home hi-fi.

Yep, exactly. Just comes down to displacement, huh?

 

[levimax]

Just install as many subwoofers as you need. Quite realistic for home hi-fi.

Especially for commercial subs with warranties you need excursion protection in case of an "accident".

 

[Titurel]

Just curious. Has anyone (other than me) who was involved in or following this thread actually owned a bag end infra-sub or even laid hands on one?

 

[Flaesh]

owned a bag end infra-sub or even laid hands on

Why do I need it?
Can you show the measurements of the subject in your MLP, preferably at different levels up to the maximum?

 

[Titurel]

Why do I need it?
Can you show the measurements of the subject in your MLP, preferably at different levels up to the maximum?

You don't seem to understand the concept of actually responding to the question posed.

I asked if anyone following or involved in this thread had any hands-on (first person) experience with the product.

I was not soliciting any other information or opinion.

In response to your questions. 1) I could not guess why you'd need it. 2) No longer owning one, as I mentioned earlier in this thread, I am unable to comply with your request.

 

[gnarly]

Just curious. Has anyone (other than me) who was involved in or following this thread actually owned a bag end infra-sub or even laid hands on one?

Not me. But been familiar with them for many years.
And more important (to me), I've made DIY copies of how how they work. Freqency dependent limiting is the only part of the processing probably out of the reach of most folks trying to emulate. them.

 

[Vladimir Filevski]

will have the same magnitude roll off and phase response,

No. Integrator is equivalent to analog low-pass filter with 0 Hz corner frequency. That kind of low-pass filter is impossible to make.

 

[Chrispy]

Just curious. Has anyone (other than me) who was involved in or following this thread actually owned a bag end infra-sub or even laid hands on one?

Along those lines has been years since I've seen them mentioned. Know nothing about the drivers they use either. I thought they ceased production but not sure.

 

[Titurel]

Along those lines has been years since I've seen them mentioned. Know nothing about the drivers they use either. I thought they ceased production but not sure.

Bag End still lists them on their website and at least one online retailer (Full Compass) offers them for sale. Their home office seems to be in Elgin, Illinois.

 

[gnarly]

It is not a subwoofer scam! You got it all wrong.
It is not "only +6 dB shelv"! A single integrator has 6 dB per octave slope, so for equalizing a second-order sealed enclosure, two integrators are needed, producing 12 dB per octave slope. You can equalize it to 1 Hz if you want, but for practical reasons Bag End subwoofers are limited to 8 Hz. This type of subwoofer is patented by Edward Long and Ronald Wickersham and is licensed to Bag End company.
I built this type of subwoofer some 20 years ago, with 18" RCF pro driver - it works as advertised by Bag End company. I limited it to 15 Hz.

I totally agree is is not a scam.
But I would call the term "integrator" to be nothing more than a first order (6 dB/oct) shelving filter).

here's the patent https://patentimages.storage.googleapis.com/28/fa/73/02d3569c740a6f/US4481662.pdf
They got a patent simply because they were among the first to use a sub below resonant frequency.

No. Integrator is equivalent to analog low-pass filter with 0 Hz corner frequency. That kind of low-pass filter is impossible to make.

Don't see where you get that. The integrator seem nothing more than a 6dB shelf to me, as said above.
And did you mean a high-pass filter at 0 Hz?


Maybe there is more going on than I've seen / understand? Please add links if so...
As I've solely been relying on the patent and past reviews such as this .... https://web.archive.org/web/20110726154245/http://www.bagend.com/elfm2eqreview.htm

Briefly, the ELF approach was the work of Ed Long and Ron Wickersham, two creative individuals with an impressive background in audio engineering. They realized that if a subwoofer could operate below its resonant frequency, not only would it produce less colored (read: distorted) sound, but also exhibit more linear frequency response. By matching the roll-off of a low-frequency loudspeaker driver mounted in a sealed box (infinite baffle) with a circuit that boosted those frequencies at the same rate, flat frequency response to very low frequencies (less than 10Hz) could be achieved. There was the added bonus of working well with even very small drivers and cabinets. The limiting factor in choosing the size of the driver is the maximum output level and not the low-frequency cutoff.

"I auditioned the ELF-M2 with Bag End's 10-inch ELF subwoofers and I was very pleased with the impact of these tiny subs. To protect the subwoofer drivers from damage caused by overexcursion at these extremely low-frequencies, the ELF design also includes a protection circuit called Concealment. This circuit progressively reduces the amount of gain, starting at the lowest frequencies, once the (adjustable) threshold has been exceeded. For a more detailed explanation of the ELF (Extended low frequency) principals, dig out your back issues for my review of the ELF-1 in the October 1993 issue of EQ magazine or call Bag End and ask for its booklet on ELF."

Putting the patent and that review together, all i see is low shelves and an implementation of dynamic limiting, called Concealement.
Concealement...hmmm...interesting name

 

[KL....]

No. Integrator is equivalent to analog low-pass filter with 0 Hz corner frequency. That kind of low-pass filter is impossible to make.

Thank you @Vladimir Filevski, Yes, a Double Helmholtz Effect will (perhaps) achieve LF (architecture/implementation is critical, isn't it?) but not with absolute phase nor absolute Linearity nor a 0 Hz corner frequency (f1 and f2), reasonable, although the Driver will/should achieve LF with (much) less Cone excursion, won't it?

Thank you @gnarly, Yes, not a scam, as @Vladimir Filevski has mentioned, and the following suggests much, doesn't it?
Re Concealment.... why would do you expect everything to be shown in the Patent, other than it is possable?

By matching the roll-off of a low-frequency loudspeaker driver mounted in a sealed box (infinite baffle) with a circuit that boosted those frequencies at the same rate, flat frequency response to very low frequencies (less than 10Hz) could be achieved. There was the added bonus of working well with even very small drivers and cabinets. The limiting factor in choosing the size of the driver is the maximum output level and not the low-frequency cutoff.

 

[gnarly]

although the Driver will/should achieve LF with (much) less Cone excursion, won't it?

Nope, no way.

Re Concealment.... why would do you expect everything to be shown in the Patent, other than it is possable?

I meant concealment as in....."the prior mentioned integrator magic" only works as described before the concealment circuity kicks in....
and that is what is being concealed imo.

Sounds like the 'concealment' filter is just a side-chain level-triggered EQ or hpf. Which means, at any SPL above the trigger level, the sub acts like any other sub used below resonance, with boost and a high-pass filter.

 

[Vladimir Filevski]

But I would call the term "integrator" to be nothing more than a first order (6 dB/oct) shelving filter).

Well, no.
Here is frequency response (black) and phase (red) of 8 Hz first order shelving filter:

And here is frequency response (black) and phase (red) of integrator:

here's the patent https://patentimages.storage.googleapis.com/28/fa/73/02d3569c740a6f/US4481662.pdf
They got a patent simply because they were among the first to use a sub below resonant frequency.

Yes, that is the whole point of patenting - to be the first.
Like the famous example of Alexander Graham Bell and Elisha Gray.

Don't see where you get that. The integrator ...
And did you mean a high-pass filter at 0 Hz?

Here is frequency response (black) and phase (red) of 8 Hz low-pass filter:

Compare it to the above integrator.

Maybe there is more going on than I've seen / understand? Please add links if so...

Here is a link for explanation about integrator:
https://en.wikipedia.org/wiki/Op_amp_integrator

Because sealed box has 12dB/octave slope, two integrators are required. And some high-pass filtering at 8 Hz, or so.

 

[gnarly]

Well, no.
Here is frequency response (black) and phase (red) of 8 Hz first order shelving filter:
View attachment 553366

And here is frequency response (black) and phase (red) of integrator:
View attachment 553368

Thanks for posting this followup.
First graph above is surely a first-order low shelf.
2nd graph, supposedly showing a straight line downward sloping magnitude response of the filter, and a constant flat phase of -90 degrees,
is an impossible analog filter for all I've ever understood.
Doesn't matter how implemented in the analog/minimum-phase world, non-level or bandwidth limited response cannot have flat minimum phase.

Any more material other the the wiki link? It's over my head math/terminology wise.

The sum of the integrator (two very shelf like filters if not plain shelves), along with a sealed subs natural 2nd order roll-off, should indeed give flat mag and phase (if properly executed.)
No disagreement there. But that necessitates the integrator NOT be flat mag and phase; it necessitates the integrator be a minimum-phase filter that has the inverse magnitude roll-oof or the sealed sub.

And I'll contend again...all that mag and phase flattening is fine until excursion is needed beyond the driver's capability, and then dynamic limiting kicks in.

Yes, that is the whole point of patenting - to be the first.
Like the famous example of Alexander Graham Bell and Elisha Gray.


Here is frequency response (black) and phase (red) of 8 Hz low-pass filter:

View attachment 553374

Compare it to the above integrator.

What order is that low pass? And to what/how is applied? In series with the boost filters I assume (be they integrators of shelves)

Here is a link for explanation about integrator:
https://en.wikipedia.org/wiki/Op_amp_integrator

Because sealed box has 12dB/octave slope, two integrators are required. And some high-pass filtering at 8 Hz, or so.

High-pass now? Like I've been saying? Hope so, cause I can't see any rational for a 8Hz low pass..

 

[gnarly]

Ok, I've searched integrator more thoroughly and see that the gain never stops, and that the circuit provides real freq dependent time delay (1/4WL)
I stand corrected and have learned something . Thank you Vladimir and guys.

 

[Vladimir Filevski]

What order is that low pass? And to what/how is applied? In series with the boost filters I assume

I am sorry, I wasn't specific - it is 8 Hz first order low-pass filter. Of course there should be some gain applied - in the same OpAmp filter circuit, or as additional simple gain block before the filter.

High-pass now? Like I've been saying? Hope so, cause I can't see any rational for a 8Hz low pass..

High pass filter before integrator is required, to stop unnecessary infra low signals overloading electronics and woofer.
Actually, 8 Hz second order low-pass filter has (almost) identical performance to the double integrator plus protective 8 Hz high-pass filter. I used 8 Hz corner frequency simply to correlate with the Bag End design.
Actually, I used 20 Hz second order low-pass filter as an example in my paper "Simple Bass Equalizer for Closed-Box Loudspeaker Subwoofers":
https://ieeexplore.ieee.org/document/4588104
A year or two after my paper, British REL announced the new subwoofer line, which apparently used the same technique as in my paper. After I contacted them to clarify, they simply erased the explanation from their adds how their subwoofer electronics works.

 

[Ted G]

I thought the theory in the video made sense. I also agree it’s likely overkill for me. But, I still want one or three.

 

[norman bates]

Bag end definitely pushed the phase as a selling point.