Relationship Between Driver Frequency Range and Claimed Range of Speaker Cabinet

[Obrienp]

I suspect this topic may have been covered in one way or another but I was unable to find the topic in a search. Please point me to the previous thread if the subject has been covered.

 

I have recently been looking at the claimed usable frequency ranges of bass drivers from various manufacturers: Eminence, Celestion, Lavoce, Faital, etc. Very few claim to have a useable range below 45 hz but these same drivers are used in many cabinets that claim frequency ranges of 40 hz or below. BF claims 30 hz from their Emminence based cabs but I know their drivers are a proprietary design.

 

What is going on?  Are some cabinet manufacturers telling porkies about the cabilities of their cabinets, or is there something in the internal design of the cabinets that increases the useable frequency range beyond that of the drivers? Alternatively, are the manufacturers making drivers for the OEM market that have a better frequency range than the units they sell to consumers?

 

P.S. I was looking at the market because I heard of an empty cabinet that was going cheap. Having looked at what was available and the price of drivers, I decided it was not the bargain it appeared. Why would you want a driver that couldn’t reproduce a low E, let alone an open B?

Edited Saturday at 18:36 by Obrienp

[nekomatic]

Well the short answer is that the bass response of a speaker is completely dependent on the cabinet (volume and tuning of the reflex port if it has one), so the low end frequency response quoted in the speaker specs doesn’t really tell you anything helpful by itself.
 

There is software that will work all this out for you as long as the necessary data for the speaker is available, and friendly people on this forum will often run some calculations for you if you ask. Do you know the volume of the empty cab, and what diameter driver did it take?

[Obrienp]

@nekomatic Thank you for the response. That is reassuring, although confusing for the uninitiated like me.

 

The cabinet in question was a 2 x 12 but I dithered long enough for somebody else to snap it up anyway.
 

I was more concerned about the possible ramifications of the driver specifications that I turned up during my research, which is why I asked the question. I still don’t understand how a driver that the manufacturer has rated at eg 45 hz - 14,000, can deliver 35 hz - 20,000 hz, if put in the right enclosure. I’ll just have to take it on trust.

 

I’ve decided that I am going to stick to ready made cabs, rather than going down the DIY route.

Edited Saturday at 23:17 by Obrienp

[SimonK]

 

[itu]

1) An ordinary bass cab does not have to produce but the first harmonic. Therefore the usable f range should start from around 60 Hz. HiFi is another world.

 

2) A bass with a coil-magnet pickup can produce frequencies up to around 4 kHz.

 

3) Back to HiFi, and the linearity of a f response. If the range is expressed with the Hz only, it is useless. If the (±3 or ±6) dB limits are mentioned (or they can be seen in the curve), the audible range can be estimated. A ±20 dB curve is worthless, but suitable for marketing...

 

4) An element may have some specs. When it is built to a cabinet, the end result depends on the build quality, dimensions, a possible X-over, are the seams air tight, stability of the materials, stiffness and so on.

 

 

A shorter version:

a) The bass needs a cab that is able to produce reasonable sounds between 60 - 4 000 Hz (and between maybe ±9 dB). Bass is not HiFi stuff.

 

b) A good, solid cab with relatively big dimensions is most probably functional. Has to be listened with the amp and the bass you are using.

 

c) Everyone can make a cabinet, few can build a good one. Plain numbers of one component do not reveal the performance of the whole system.

[Obrienp]

I’m almost there.

 

Isn’t the range of a 4 string bass something like 41 - 4,000 hz in standard tuning? Don’t you need to be able to reproduce the 41 hz, or are you saying we don’t really hear anything below 60 hz? What about 5 strings with a low B?

[itu]

30 Hz x 2 = 60 Hz.

 

As I wrote earlier, this 60 Hz is the first harmonic of B. Ear needs this to create the lowest f = 30 Hz. It does not have come out from the speaker.

 

Human hearing is around 20 Hz - 20 kHz. Yes, it is possible to use subwoofers to go lower. Problems:

 

A) How to control the lowest (mush) range?

 

B) Wavelength is quite long at the low end. Is the space big enough that the frequencies are usable?

 

C) Lowest frequencies need a big box and consume LOTS of power. Is the system feasible to carry around?

 

D) The highest f that comes straight out from the neck (24 frets, 4 string in EADG tuning) is around 500 Hz. Put there three harmonics (1 kHz, 2 kHz, 4 kHz) and there you have the upper limit of the need. This is not very easy to push out from a big cone. X-over and a horn?

Edited Sunday at 14:06 by itu

[Obrienp]

@itu Thank you. I think I’ve got there now.

Edited Sunday at 14:07 by Obrienp

[itu]

Basic theory (that was cut short last time because of a cat):

Sound travels 340 m/s in the air (20 °C).

 

Wavelength of a 20 Hz wave is 340 m/s / 20 Hz = 17 m. 40 Hz is half of the previous, 8.5 m, and so on.

 

Standing wave means that some distance from a wall to another parallel wall happens to be the same as some wavelength. Therefore to avoid standing wave the space should consist of non-parallel or well damped walls.

 

Standing waves can be best tamed with (parametric, or 31-band) equalizers. Basically if you know the height, width, and length of a room, you can easily calculate standing waves and their harmonics, and kill the ringing with an eq.

 

As cabinets are usually the shape of a box, there are also three main frequencies (and their harmonics) affecting the sound, if they are not tamed with suitable damping material in the walls.

 

Horn or high range driver has usually its own case (or does not need one at all).

 

A box volume limits either from the two: lowest reproducible f, or sensitivity. If the box is small, you can go low, but you need extra power. If the box is big, and you don't have to go so low, you may get a louder unit.

 

A bass cab can be considered as a PA speaker, that doesn't have so neat f response, but can push lots of sound out. A good quality HiFi speaker may not be very efficient (like 0.1 %) whereas a PA speaker may have 2 % efficiency. In power numbers a 100 W amp creates up to 2 watts of sound in PA system, while at home the level is ridiculously low, like 1/10 W. But the response can be pretty straight.

 

Why is a non-linear response OK? Because it is one part of the sound. Besides the response is usually some sort of a band pass filter (BPF), like with bass it could be 60 Hz - 4 kHz that removes over three octaves of the sounds we are able to hear.

 

Ear can sense 10 octaves (some papers claim 16 Hz as the start):

20 - 40 (Hz)

40 - 80

80 - 160

160 - 320

320 - 640

640 - 1250 (yes, you saw the rounding, and understand the reason in the next lines)

1250 - 2500

2500 - 5000

5000 - 10000

10000 - 20000

 

In a band context there are several other instruments and sounds (drs, keys, g, voc...) that compete with the high end parts of the bass' response. Therefore we could cut quite a lot of the high end without issues. Remember a g-word response starts from 80 Hz, which is only one octave higher compared to bass. It is not even feasible to mix many similar frequencies to create any more mush. In studios the (frequency) bands are limited more or less between instruments to get better overall sound.

 

If this raises any ideas/questions, I may continue. Now that cat tries to reach my keyboard.

Edited Monday at 19:34 by itu

[Bolo]

On 26/01/2025 at 14:59, itu said:

Human hearing is around 20 Hz - 20 kHz

For 17-24 year olds that have stayed away from noise, drugs, alcohol  and have no history of ear infections etc.

It is the optimal human hearing, available to about 5% of mankind.

[itu]

20 Hz - 20 kHz is the average. I heard about a person who has reached 37 kHz in tests. Like
@Bolo said, an old fart like me may have issues reaching 10 kHz. Small venues and loud drummers have done lots of harm to my hearing.

 

But: far more serious is to lose the 1 - 5 kHz area, because that is the critical area of speech recognition.

[JPJ]

Another thing is that driver manufacturers will often talk of the frequency range in terms of +/- 3db. As driver frequency is a curve, the 45Hz might be at -3db, but the speaker will continue to produce much lower frequencies but at progressively lower spl (or ‘volume’ or ‘loudness’). At least thats how i understand it. 

[bremen]

On 28/01/2025 at 18:34, itu said:

20 Hz - 20 kHz is the average for a teenager who doesnt own earbuds or go to clubs. I heard about a dog who has reached 37 kHz in tests.
 

Ftfy 😉

 

The answers to OPs original question are 100% good points well made.

[itu]

On 29/01/2025 at 20:05, JPJ said:

Another thing is that driver manufacturers will often talk of the frequency range in terms of +/- 3db. As driver frequency is a curve, the 45Hz might be at -3db, but the speaker will continue to produce much lower frequencies but at progressively lower spl (or ‘volume’ or ‘loudness’). At least thats how i understand it. 

1) True in a closed box. The problem is that if you want to produce reasonable loudness, you have to push the eq to reach lower frequencies. That equals lots of power and driver excursion, both of which are always limited to some extent.

 

2) Not true in a reflex box. The cabinet cuts lower frequencies very effectively after the tuning f. The driver simply cannot produce any feasible loudness under that. The element acts like it had no cabinet around itself.

 

A HPF (high pass filter that cuts lower f) is your friend. You have better power reserve behind you, and handling noises are limited. Win-win.

 

±3 dB is very strict, and suitable for HiFi stuff. I might use ±6, or even ±9 dB limits when we talk about PA systems. HiFi is about flat response, PA is about power.

(That's why I hate the term FRFR, which is somewhere between flat response and power, but I think closer to power.)