Understanding Speaker Frequency Response – The Alike

Note: Paul Dicomo was an incredible good friend to eCoustics and many people over the previous 2 a long time. Sadly, Paul misplaced a brave battle with most cancers in 2021 and the trade misplaced an excellent man who taught a technology of audio/video journalists the right way to perceive the merchandise that we cowl.

Paul contributed a variety of glorious articles over time and we hope {that a} new technology of readers finds nice worth in his writing.

By Paul Dicomo

The Secret Behind The Industry’s Most-Cited Spec.

Here’s a fast quiz: which of those two audio system sounds higher: Speaker A with a frequency response vary of 45Hz to 18kHz or, Speaker B with a spread of 20Hz to 25kHz? The reality is there’s merely not sufficient knowledge in these numbers to know something of worth. Taken out of context and with out different knowledge, a easy set of numbers don’t let you know a lot about actual world sound high quality. But individuals make audio shopping for choices primarily based on printed specs, such because the frequency response spec, on a regular basis. I’d prefer to demystify the method for you; allow you to in on a little bit trade secret about “The Frequency Response Spec.”

My Frequency Response

The Frequency Response specification makes an attempt to explain the vary of frequencies or musical tones a speaker can reproduce, measured in Hertz (recognized to old-timers as “Cycles per Second”). The vary of human listening to is mostly considered being from 20Hz, very low bass tones, by 20kHz (20,000Hz), the very highest treble. Presumably a speaker that might reproduce that vary would sound lifelike. Alas, it’s no assure. The most vital determinant of a speaker’s frequency efficiency will not be its width or vary, however whether or not it’s able to reproducing all of the audible frequencies on the identical quantity at which they have been recorded.

You don’t need the speaker to alter the “mix” of tones; that will spoil the timbre of voices and devices, making them sound unnatural. Ideally, you need the sounds which might be on the recording to be reproduced as they have been recorded, with out the speaker altering the sound. To say it one other approach: should you made a recording of all of the audible tones on the identical quantity and performed that recording by a speaker, you’d need all of the audible tones to come back out on the identical quantity. In reality, that’s a method of measuring audio system. A sign that’s comprised of all frequencies at equal quantity is fed right into a speaker that sits in a room with no reflective surfaces. A calibrated microphone is positioned in entrance of the speaker and feeds the speaker’s output right into a machine that plots the frequency vs. amplitude as proven in Figure A.

frequency vs. amplitude
Figure A

Now check out the graph in Figure B. That’s the frequency response of the Erehwon Model 10, with drivers and tweeters manufactured from pure Unobtainium (“Half the carbs, all the sound!”). The flat line on the graph signifies that the speaker is “flat”; it reproduces all of the musically related tones on the identical quantity. That doesn’t imply {that a} “flat” speaker will play all recorded sounds on the identical quantity — bear with me right here — it means that it’s going to deal with all sounds equally; it received’t impose its will on the music however will help you hear the music because it was recorded. Flat is sweet. Flat response signifies that the speaker reproduces sound precisely.

Erehwon Model 10
Figure B

Too dangerous that the Erehwon Model 10 doesn’t actually exist, and neither does Unobtainium. Today’s applied sciences permit speaker designers to get nearer to the “flat” supreme than ever earlier than, however they nonetheless fall far in need of “perfection.” So if a frequency vary spec will not be enough, what’s?

Frequency Response In Context

An enormous enchancment could be a frequency response quantity that additionally contains the amplitude tolerance, expressed as “XHz-YkHz +/- 3dB.” This tells you that the amplitude of the speaker’s response relative to frequency doesn’t deviate greater than 3 Decibels from the middle line. The “plus or minus 3dB” spec is considered an ordinary of kinds. The idea is that 3dB variations are “just perceptible,” so a speaker whose response curve lies inside that tolerance window is a fairly correct speaker. Let’s see if that concept holds water.

speaker frequency response
Figure C

Take a have a look at Figure C. This speaker has response that may be specified as 20Hz-20kHz +/- 3dB. Take a have a look at Figure D; it, too, can have the very same specification as Speaker C! Do you assume they are going to sound related? NOT! They received’t sound even remotely like each other. Speaker C could have “one note” bass and can make voices and different devices sound unnatural, however Speaker D will sound easy and extra pure.

speaker response curves
Figure D

If I had to decide on strictly by the response curves, I’d select speaker D as a result of its amplitude variations are smoother and gentler. In distinction, speaker C’s amplitude variations are extra excessive and “spikey.” Experience has proven speaker designers that these fast modifications in response produce a sound that’s extra fatiguing, much less pleasing and subjectively much less correct.

Advertisement. Scroll to proceed studying.

smooth frequency response
Figure E

Now have a look at the response of the speaker in Figure E. This speaker reveals a easy response curve with low amplitude variations so that you’d anticipate a reasonably pure sound; nonetheless, the bandwidth of those errors may be very broad, and expertise has proven us that even low quantity variations are audible in the event that they cowl a broad vary of frequencies. In this case, Speaker E would have wealthy bass, outstanding treble and be considerably recessed or “laid back” within the midrange. Audiophiles name this “The Smile Curve.” It’s not the fascinating trait it appears like nevertheless it’s a really “sellable” trait to naive patrons.

My Response To Frequency

Now that the significance (and limitations) of amplitude variations in frequency response graphs, you would possibly ask: “does the frequency range tell us anything at all?” Yes, it does. As lengthy as the amplitude tolerance (+/- 3dB), the frequency response vary or width tells you the way excessive or low the speaker goes. A speaker rated as 20Hz – 25kHz +/- 3dB will play decrease bass and better treble sounds than a speaker that measures 40Hz – 20kHz +/- 3dB. I wouldn’t wager cash that it might be the higher, extra pleasing speaker, however at the very least I’d know one thing of worth.

And now that you know the way to interpret these numbers, you’re able to run proper out and purchase a speaker simply by trying on the response curve, proper? I wouldn’t advocate it. Despite many advances in know-how over the previous 20 years, frequency response measurement is an imperfect science. The identical speaker measured by two completely different labs might yield completely different response graphs. And some firms simply plain cheat once they publish response curves. If it seems to be hand drawn, it in all probability was. (Yes, the graphs have been hand drawn for illustration functions.)

The Third Dimension
So far we’ve talked about frequency (the X axis of the graph) and amplitude (Y axis) however we unnoticed an vital third dimension: time. When a speaker responds to an impulse, for instance a rim shot — “THWACK!” — it ought to begin immediately and cease the moment the instrument stops making sound. If the speaker retains vibrating or resonating and making sound after the supply sound stops it’s altering, or “coloring,” the sound of the unique recording. And that’s dangerous.

Figure F Figure G
Upload Upload

Figure F reveals a bandwidth restricted impulse sign. You can see that it begins and stops abruptly. Figure G reveals that very same impulse popping out of a speaker. You can see that the sound persists after the impulse enter has stopped — it resonates or “rings.” The speaker is altering the timbre or character of the unique recording. In order to see to what extent and at which frequencies the “ringing” is going on, we use a complicated pc algorithm referred to as MLLSA (affectionately referred to as “Melissa” by engineers who don’t date a lot) to measure the response of a speaker in frequency, amplitude and time. Figure H is a MLLSA spectral decay graph of a prototype speaker. The third axis of this graph is time, so graph strains closest to you’re measurements taken later than those within the again. Think of it as a collection of slices with every slice being a frequency response graph taken at a distinct cut-off date.

If we have been to measure the proper speaker the MLLSA graph would seem like a straight line in again with no strains in entrance. Real audio system fall far in need of this supreme and proceed to resonate after an impulse has stopped, resembling in Figure H. Figure I is a Polk LSi9, and we are able to see that the speaker stops responding sooner within the midrange than the speaker pictured in Figure H, indicating that the LSi9 is a greater sounding speaker.

Figure H Figure I
Upload Upload

While no measurement method can totally describe the subjective sound of a loudspeaker, MLLSA and different frequency response measurements are of nice assist to Polk engineers in creating higher sounding audio system. Only a idiot would design a speaker primarily based on measurements alone and solely a complete idiot would design a speaker primarily based solely on subjective listening. A speaker which may sound good on a selected recording might in truth be flawed – it could have what is often referred to as a “euphonic coloration.” It could also be pleasing to the ear underneath sure situations, nevertheless it positive ain’t proper.

We use each measurements and subjective listening to design and consider audio system. The measurements save us time and are an incredible assist in pointing us in the appropriate design route, avoiding errors which will come again to chunk us later. The measurements give us a way of choosing which experimental designs are price listening to.

But now we have to be glad with the whole subjective expertise earlier than a brand new design turns into a Polk Audio speaker. We spend numerous hours listening to music and films. Several skilled listeners should hearken to a proposed design and log out on the sound earlier than a mannequin may even go into manufacturing. The Project Manager, Systems Engineer, VP of Engineering, Product Line Manager, and particularly Matthew Polk, all should agree that the prototype delivers the form of rewarding listening expertise that you just anticipate from Polk Audio.

What’s Your Frequency?

You now know the key: a frequency response specification is a really weak predictor of the particular efficiency of a loudspeaker. A frequency response chart could be extra useful, nevertheless it’s lacking the vital time measurement. You now know to search for total curve smoothness and to keep away from fast swings in amplitude. Some magazines and evaluate websites publish MLSSA graphs of reviewed audio system, and now you’ll perceive the right way to interpret them. More energy to you!

No matter how adept you could be at deciphering frequency response knowledge, it ought to solely be one knowledge level amongst many in selecting a speaker. There is a lot extra to a speaker’s efficiency than simply its response – like its dispersion and imaging, dynamic vary and element decision in addition to dimension, cosmetics and worth. Looking at good frequency response knowledge will help you eradicate audio system with apparent and obnoxious errors. Once you’ve eradicated the increase & tizz pseudo-fi audio system, you possibly can settle right down to cautious listening and making a extra knowledgeable alternative.

How Polk Specifies Frequency Response

Polk Audio publishes two frequency response specs: “Overall” and “-3dB.” “Overall” describes the frequency vary limits of the speaker inside an amplitude drop off of 9dB. Any frequency reproduced greater than 9dB down from the remainder of the frequencies will contribute little to the sound. The “-3dB” spec describes the frequency vary limits of the speaker inside an amplitude drop off of 3dB.

Advertisement. Scroll to proceed studying.

I simply wrote this massive article making the case that these sorts of numbers usually are not terribly helpful in making shopping for choices. So why does Polk use them? For higher or for worse, these numbers are the norm within the audio trade.

To not publish them would go away an impression that our merchandise weren’t aggressive. A greater query could be: why don’t we publish frequency response and MLSSA graphs along with the straightforward numbers? We really feel that these graphs wouldn’t be significant to the overwhelming majority of shoppers.

It takes years of working with measurements and loudspeakers earlier than you get a great sense of how the graphs correlate to subjective sound high quality. Incorrect interpretation of graphs can simply result in misinformation and dangerous selections. Finally, the variation in measurement strategies could make evaluating graphs from two completely different labs or producers unreliable and deceptive.

Paul Dicomo

By Paul Dicomo, Polk Audio – Marketing Manager

Originally Published: September 28, 2005

More Articles by Paul Dicomo:

All You Need to Know About Bass Management

Home Theater in a Box vs. Components. Which is Best For You?

Source link

Comments are closed.