[Wayne A. Pflughaupt]

brucek wrote:

I find that after some thirty two years working in engineering that I would rather play with my grandson than study the fundamentals of gain structure.

Can’t say as I blame you there! Sure wish mine lived closer...

Anyway – you remember those private discussions we had about this a few months ago. Well, eventually it dawned on me (sometimes I’m a little slow), “Waitaminnit, what about the sub itself?” Still later, it dawned even further, ”Hey, what about the receiver?”

Here’s what I’m finding. Behringer gives the 1124 ‘s noise spec as “>94 dB, unweighted, 20 Hz – 2O kHz,” an excellent spec that translates to about 120 dB dynamic range (the stuff of dreams for most of our home gear). A valid S/N spec has to have a qualifier like that, showing how the figure was achieved; otherwise it’s meaningless. In other words, when you see a S/N spec of simply “100 dB” it’s not worth a lot.

The BFD’s “unweighted” (a.k.a. Z-weighted) spec is best, because it considers flat, full range response. More commonly you’ll see something like “100 dB A-weighted.” As you know, A-weighting filters roll off the lowest and highest frequencies. As Rane says in some of their technical papers, "An A-weighted noise spec [can] ‘hide’ nasty low-frequency hum that make for bad noise specs. Sometimes A-weighting can ‘improve’ a noise spec by 10 dB. Words to the wise: always wonder what a manufacturer is hiding when they use A-weighting.”

Rummaging around the web sites of the popular receiver manufactures, a few have no qualifier for their S/N ratings. Most of them that do use a qualifier, it’s a strange “xx dB, IHF A-Network.” I’ve never seen this anywhere but with home receivers. I’ve posted questions on two Forums and haven’t had any luck finding any info on what it is (see this thread).

So – until I learn otherwise, I’m going with the assumption it’s A-weighting. This means that with the typical receiver’s noise spec, which ranges from 95-105 dB, depending on the manufacturer (that last figure is probably wishful thinking), this means there’s a good chance our receivers are doing at best 85-95 dB S/N unweighted.

With the powered subs, it gets even worse. It’s hard to find a manufacture that even publishes a noise spec, and if you can find one, you only get an unqualified figured. The few I have found range between 84-94 dB. Is that A-weighted, C-weighted, Z-weighted? Who knows? Like Rane says, they’re meaningless numbers.

On a different topic, a couple of things in that quote from the BFD manual jumped at me:

When using the Feedback Destroyer as a parametric equalizer (Parametric EQ mode) and applying extreme attenuation values in the low end range of the frequency spectrum (below 50Hz), you should perhaps set several filters to process the same frequency. Depending on the amount of attenuation applied, the use of only one filter may lead to slight distortion, which produces interference in the signal path. This is a natural physical phenomenon which should be avoided. Use for example two filters adjusted to a signal attenuation of about -12dB (with the same frequency and bandwidth). This produces the same effect achieved with one filter set to -24dB

You would think that educated audio engineers would know that “slight distortion” below 50 Hz is inaudible. Also, note their idea of an “extreme attenuation value” – quite a bit more than ours.

Regards,
Wayne