By Jack Sharkey, August 30, 2018
In Part 1 yesterday we discussed harmonics, distortion and clipping. In Part 2 we'll discuss Total Harmonic Distortion, Signal-to-Noise Ratio and sensitivity. You don't need a deep understanding of everything covered in this two part series, but a little background may be just enough to help you make an educated decision when purchasing your next piece of audio gear.
Total Harmonic Distortion
In the audio world, Total Harmonic Distortion is the percentage increase in harmonics from the source through final output. A complete system has a THD number, as does each individual component.
Without going into the actual physics of how harmonics are created by audio components, here's what happens: components in the chain develop sympathetic and unsympathetic frequencies of the signal they are processing. Theoretically and practically, the higher the quality of the component the fewer harmonics created. We can get a feel for the audio quality of a component by measuring what and how many harmonics are created at certain frequencies and volumes. This is basically the measuring stick we use to see how a component performs.
Typically, speaker THD will be higher than amplifier THD because of the nature of a speaker's job. Speakers are the critical link in the distortion chain and its intuitive that we should expect more THD from our speakers than from any other component in the chain.
In short THD in amplifiers of any level of decent quality stopped being a real issue in the late 1970s. Except for a very few environments and circumstances, THD specs <1% are negligible and not something you really need to worry about. Find the tightest spec you can but bear in mind the money you spend to decrease your THD from 0.07% THD to 0.02% or even 0.05% may not translate into much of a difference if your budget is comfortable at 0.07%. When it comes to THD less is not that big a deal if you can't afford less.
Like the song says, you got 99 problems, and THD ain't one of them.
Signal To Noise Ratio
If you listen to a lot of cassettes and 8-tracks, SNR might be an issue, otherwise you can pack your SNR concerns away with your leisure suits and platform shoes. In the digital world SNR is not a thing, and even on pre-digital recordings that have been remastered and reformatted into digital files, SNR is not a concern. Of course, SNR with vinyl is most definitely a thing, but generally people who listen to vinyl embrace the fact that noise is part of the experience.
The lower the number (as represented in dB) the less efficient a speaker is. The less efficient a speaker is the more power you will need to send it to get the same SPL as from a speaker with a higher sensitivity number. If Speaker A is more sensitive than Speaker B, it does not specifically mean that Speaker A is a better speaker. It simply means that Speaker A does a more efficient job of converting electrical energy into sound.
Sensitivity is derived by measuring 1 Watt of power at 1 meter with a constant frequency (often 1kHz) and finding out how much of that 1 Watt of power is turned into heat, not sound energy. The result is specified in dB.
Typically, the smaller the speaker the less efficient it will be. But, sensitivity is only a measure of the efficiency of the power transfer from electrical energy to music. A better performing speaker may simply not be as efficient as a lesser performing speaker because of the components and design that makes the speaker better in the first place. It's never a clear-cut as we'd like.
Putting The Numbers To Work For You
Let's put some of these numbers to use, using an LS50 for example.
The top end power for an LS50 is rated at 100W continuous power. That's a lot of power that will develop a lot of SPL through the relatively efficient LS50. But, let's say you're lucky and you get to blast your Motley Crue and Neil Diamond CDs for hours at a time. One hundred watts doesn't leave you a lot of headroom, so you have to be careful about clipping (impedance drops as voice coils warm up causing your amp to work harder). The harder an amp works the more likely it is that clipping will occur.
This is where it is sometimes advisable to have an amp that is spec'd higher than the speaker because you will lessen the chance of putting the amp into a clipping condition. But, when you blow your speakers by over-powering them, you did not get that advice from me. Off-the-record, you most certainly can power a speaker with an amp that is spec'd higher than the speaker's power requirements - you just have to ensure no one ever turns the volume past a point that the amp delivers more power than the speaker can handle. Otherwise you'll be calling your speaker company's service department.
The truth is there is no easy calculation. We can't use the pro-audio calculations we read about in Part 1 but we can cheat on that calculation a little. Here's where it helps to work with a dealer you trust and who knows more about the audio game than just a few talking points about video games and smartphones. But while you're on your way to a trustworthy audio dealer, here's a few things to keep in mind:
How loud do I listen to music?
How often do I "crank" the volume?
What type of music do I listen to?
How big is the room I will be trying to fill with sound?
Those Four Questions Will Help Answer the Question: Which Amplifier Is Best For My Speakers
But these two specs are probably the mose important ones you should consider:
Maximum Power Out: This is a peak measurement. Here is where I would use the pro-audio calculation and double the power rating of the speaker and multiply it by 1.25, so an amp that is capable of producing instantaneous peak power around 250W (PMPO) would be a good fit for a speaker rated at 100W.
Minimum Power Out: This is a continuous rating and is the number we should be most concerned about. If I wasn't in the position to buy an amp capable of continuous power at the upper end of my speaker's power rating I would only use an amp capable of producing 1.25 times twice the minimum acceptable wattage for my speaker. Translation: The LS50 requires 25 Watts of continuous power so I would oblige them by powering them with a minimum of 60 Watts ((25x2)*1.25) (50*1.25=62.5)
Note: The above specifications consider each channel (L and R) individually, so the numbers are per channel, not total.
You may fall in love with speakers that don't have a great efficiency number. That's okay! If you love the way they sound embrace the excess heat! Just keep in mind that you'll need to bump your amplifier power numbers up a little to get the desired SPL. On the other hand, cheaper, inefficient speakers don't typically deal well at higher power.
So those are the specs and numbers but in the real world of audio today the differences between brands and models (when it comes to published specs) is generally pretty minimal. At the end of the day, the most important spec is how a particular component and group of components sound. It's about the listening experience, and if you're buying gear from a manufacturer who puts the time and expense into properly engineering their products, you're really only going to benefit from that without a lot of legwork on your part. But then again, sometimes the joy of the hobby is knowing the details. But for most of us as long as we remember these three important things you're going to be okay:
Deal with people who know what they're talking about and whom you trust. The audio world has gone full tech and the numbers can be befuddling without the proper background and insight. If you're not going to learn the numbers make sure the person you're buying from understands them.
Listen to each component whenever possible but understand that how it sounds in the showroom is not how it is going to sound at home. If you're buying without listening, on-line for example, buy from a manufacturer whose reputation for not fudging the numbers can be trusted.
At the top of the audio manufacturer foodchain it really all comes down to how it sounds. Not necessarily how the sound is produced.
Note: This series is meant to present a broad overview of the subject as succinctly as possible in an attempt to give a basic understanding of the subject. By necessity, some detail and depth of explanation have been left out. The opinions expressed in this piece are the author's own and do not necessarily reflect those of KEF.