Does A More Powerful Amplifier Make Speakers Sound Better?

[Destination: Moon]

Couldn't this be readily measured?

 

[hardisj]

Can the speaker take the power?

I'm sorry to plug my own video, but there is no sense in me typing all this up when I have already provided and explained it in video form.

For those who can't be bothered with video:
Compression/enhancement: points at which the speaker behaves non-linearly in frequency response relative to lower input power. Where is this point? How do you determine it?

Below is an example of a passive speaker I reviewed (Elac UBR62). The DUT is provided with a quick stimulus sweep at 76dB @ 1m. This is the reference. The other "lines" are the relative difference in output over frequency. Red = 86dB... you can read the legend for the rest. The point of this is simple: demonstrate how a speaker's frequency response changes as a function of output levels. Any deviation from 0dB is a departure from the expected gain (i.e., 76dB to 86dB should result in 10dB gain). The departure from this expectation is provided simply as a relative delta from that difference caused by either compression (-dB) or enhancement (+dB; typically resonance or distortion driven). (Note: I don't go higher than 102dB @ 1m for reasons I have addressed elsewhere. I don't want to go OT here; you can do your own tests if you wish). The ideal would be for the output to never change as volume is increased infinitely. We know, however, that is just not practical. There are limitations a speaker reaches.

Spoiler: More Info

Dynamic Range (Instantaneous Compression Test)

Spoiler: More Info

The below graphic indicates just how much SPL is lost (compression) or gained (enhancement; usually due to distortion) when the speaker is played at higher output volumes instantly via a 2.7 second logarithmic sine sweep referenced to 76dB at 1 meter. The signals are played consecutively without any additional stimulus applied. Then normalized against the 76dB result.

The tests are conducted in this fashion:

1. 76dB at 1 meter (baseline; black)
2. 86dB at 1 meter (red)
3. 96dB at 1 meter (blue)
4. 102dB at 1 meter (purple)

The purpose of this test is to illustrate how much (if at all) the output changes as a speaker’s components temperature increases (i.e., voice coils, crossover components) instantaneously.

In the Elac UBR62 I tested you can see that as you increase volume (or have a transient) from 76dB to 86dB, the FR is roughly the same. As you go from 76dB to 96dB you lose as much as 1dB at some frequencies. But moreso, you can see how non-linear the response is becoming. At 102dB at 1 meter, things deteriorate quickly. So, it's reasonable to assume that the max SPL for this speaker is somewhere in the 96dB region because above that it becomes quite non-linear with deviations of as much as -2dB in the midbass region.


Another example below is of a powered speaker (Kef LS50 Wireless II) with built-in electronics to purposely limit the output of the speaker in order to protect it from damage. Note the scale is different from above. And it has to be because at 102dB @ 1m, the speaker is actually only achieving 100dB >200Hz and as low as 93dB at 40Hz. This is significant DSP-based limiting.

The bottom line is quite simple: there is a point where there is too much power. Not just in terms of distortion (which is certainly arguably audible) but mainly in terms of the actual frequency response. We all agree we can read a frequency response graphic. This is wasted power that doesn't even get translated as additional SPL; it's just lost in the ether. Quite literally, a waste of power. So keep this in mind when you are looking at such things; at some point the response changes depending on how much input voltage it receives.

 

[tomtoo]

Although the interesting thing is still dynamic power, right? The ability to cope with transistence peaks. crest factor.

https://en.wikipedia.org/wiki/Crest_factor

Take for example this little vintage NAD receiver:

NAD 7240PE
Power output: 40 watts per channel into 8Ω (stereo)

https://www.hifiengine.com/manual_library/nad/7240pe.shtml

How well does it handle peaks? It is designed to handle it in a sensible way, but check what it can do in dynamic power in 2 Ohm ( to use 2 ohm speakers for a long time I would not do that with a NAD 7240PE).

So how many watts are needed and when?

Then for the sound quality as usual other factors to consider Signal to noise ratio, distortion, damping factor (low such can be reflected in distortion and ,even uneven frequency response .... I think)

As others have mentioned. What kind of music you play. Modern compressed, a thick even carpet of tones then not many watts are needed. It does not place any major demands on the amplifier. That in itself is good, for the amplifiers. Not so fun for our ears.

With this you open up a can of worms.
Many people like me, like thing easy. RMS power is easy, and gives you a good orientation. Dynamic power is very complex. What time? 1ms,20ms,100ms? A long 30Hz sin bass? A short kick drum bass? How long is long enough for highly compressed music or for uncompressed. And this are just some questions you could ask.
With RMS power, you know this amp can deliver this power at least for some minutes.
Its on you to open up this can of worms. I stay easy give me RMS power and i know what i have. Give me as example pmpo power and i start laughing.
So tell me why i should care for such shiit?

https://www.techtree.com/content/ne...bo-series-wireless-karaoke-party-speaker.html

You see? Thats the can of worms you like to open. A toy rated at 10kW. Did you ever looked at a 10kW PA?

Thats a small 3kW PA.

https://www.thomann.de/intl/the_box_pro_achat_rock_n_roll_bundle.htm

How you like to compare?

 

[hardisj]

FWIW, I provided my short-term power testing above. But in my reviews I also provide long-term testing as well so we can see how/if the response changes as stimulus is applied for about 2 or 4 minutes. But that's less "dynamic" and more "extended" output sustainability. So I won't get into that here. But all of this stuff is found in my reviews for the past 4 or 5 months, if you want to look at more examples (site link in my signature).

 

[NTK]

...
So how many watts are needed and when?
...

On the answer to that question, IMHO @Charles Sprinkle has it correct. See his posts (emphases mine):

...
That having been said, if you play a pure sine wave at exactly the frequency of the minimum impedance for an extended period of time, you should expect that the amplifier will thermally protect itself from this abuse. Music has a crest factor of between 12dB and 15dB or even more. Even when pushed all the way into limit, the amplifier is going to see more than 9dB of crest factor with music or other actual program material. Even with only 6dBCF, the amplifier would be within thermal power specs given by the amplifier manufacturer. (And this at minimum impedance.)

So which way is the right way to measure power? I have something to say about that. The traditional way of rating "power" as V^2/R with sine waves is in my opinion complete nonsense. Power amplifiers don't amplify power. They amplify voltage. People don't hear watts. They hear SPL. We provide output "power" ratings as a reference, and yes, they are honestly measured and specified as described above. But I hope we all understand that actual power produced into a reactive load with complex impedance using real music or program material is going to be substantially less than headline "power" ratings.

...
Now if I was selecting an amplifier to power a passive loudspeaker, here's how I would do it. I would test the loudspeaker to see at what voltage it reached 10% THD. I would select an amplifier to provide this much voltage by reverse calculation of "power" specification, and verify that the amplifier could provide the peak current requirements. For peak current capabilities of an amplifier (when not directly specified), reverse calculate the RMS voltage with the lowest rated impedance, then multiply that number by 1.4 to get peak voltage. Peak voltage across the lowest rated impedance yields peak current capability. Of course there are other factors to consider when selecting an amplifier, but this will ensure that the amplifier doesn't clip or shut down before full output is achieved.

 

[abdo123]

@hardisj The ELAC UBR62 would need 75W to reach 102 dBSPL @ 1m. If you include ~3dB of headroom so the amplifier doesn't clip then you're in the ballpark of 150W.

So yeah, even the speakers that 'can't handle it' would need a 75W-150W amplifier.

Also do you think the compression tests are indicative of the woofer's maximum linear excursion? That would truly be 'the limit'.

 

[hardisj]

Also do you think the compression tests are indicative of the woofer's maximum linear excursion? That would truly be 'the limit'.

not necessarily.

When you say “linear excursion” keep in mind there is an IEC standard for this which defines it as a % change in Bl(x), Cms(x), Le(x) and/or Li(x). However, some manufacturers base it on physical geometry (i.e., coil height + gap). And then some use peak-to-peak and others use one-way. I use the Klippel method which is the IEC standard. But, bottom line is you're almost guaranteed to not be comparing apples to apples. Especially if "linear" means "mechanical" to you (which it does; "linear" just sounds cooler but they actually mean mechanical limits).

 

[pma]

Speakers are able to withstand (and radiate) incredibly high electric power in a burst test, which simulates real conditions well. See my thread on burst tests and measurement

https://www.audiosciencereview.com/...istortion-of-speakers-with-sine-bursts.11794/

54Vpeak, 38Vrms, 243W with a not too big Seas W18NX001. But the amp has to give it and not to clip!

 

[escksu]

The bottom line is quite simple: there is a point where there is too much power. Not just in terms of distortion (which is certainly arguably audible) but mainly in terms of the actual frequency response. We all agree we can read a frequency response graphic. This is wasted power that doesn't even get translated as additional SPL; it's just lost in the ether. Quite literally, a waste of power. So keep this in mind when you are looking at such things; at some point the response changes depending on how much input voltage it receives.

Yes, fully agreed on this. Just like amps have a limit on how much power it can give, speakers too have a limit on how much power it can take. Cheers!!

 

[DVDdoug]

Does A More Powerful Amplifier Make Speakers Sound Better?

A 20 or 30W amplifier putting-out 20W is no different than a 200W amplifier putting-out 20W (assuming no clipping and all else being equal).

is clipping audible or is just that it could damage speakers?

That's actually a myth that clipping damages speakers. It's the higher volume/power that damages speakers. When you get into clipping the total power does go up and low-mid frequencies get clipped while the high frequencies (which normally have less energy) keep going-up with the volume control. It's mostly just the louder high frequencies rather than the clipping that kills tweeters. ...It actually gets a little more complicated than that and often the distortion-related harmonics are lower than the original high frequencies and with severe distortion you can actually loose highs.

And, with a higher power amplifier you are more-likely to turn-up the volume and over-power the speaker because might not hear distortion until the speaker itself distorts.

Couldn't this be readily measured?

Unfortunately no... You can check for clipping with an oscilloscope or some amps have clipping indicators and/or power meters.

 

[DanielT]

With this you open up a can of worms.
Many people like me, like thing easy. RMS power is easy, and gives you a good orientation. Dynamic power is very complex. What time? 1ms,20ms,100ms? A long 30Hz sin bass? A short kick drum bass? How long is long enough for highly compressed music or for uncompressed. And this are just some questions you could ask.
With RMS power, you know this amp can deliver this power at least for some minutes.
Its on you to open up this can of worms. I stay easy give me RMS power and i know what i have. Give me as example pmpo power and i start laughing.
So tell me why i should care for such shiit?

https://www.techtree.com/content/ne...bo-series-wireless-karaoke-party-speaker.html

You see? Thats the can of worms you like to open. A toy rated at 10kW. Did you ever looked at a 10kW PA?

Thats a small 3kW PA.

https://www.thomann.de/intl/the_box_pro_achat_rock_n_roll_bundle.htm

How you like to compare?

That bit. Well, no. It gets a little tricky to keep track of how much and under what conditions dynamic power is enough. I did not think of that. Best to close can of worms and aim for ssensible RMS power.

 

[DSJR]

For some reason in the 80's and 90's, party animals (or those with teenage kids who had parties) would buy Mission speakers which back then were popular review fare in the UK. I've lost count of the number of service returns we had on a Monday after a weekend partying. Small amps heavily driven into clipping and speakers firing into people's drunken bodies so losing 'volume.' The crossover components were inserted into niches at the back of the plastic terminal plate and the coils would heat up like electric fires, softening and distorting heavily said plastic niches. tweeter caps would often be severely damaged and of course, ALL the tweeters in these returns were blown as a matter of course! people looked at me in surprise when I suggested they set the speakers high so they played over people's heads or better still, hired some PA speakers for the party. Bose do or did a pro version of the 901 and turned round with the angled baffles facing you, they 'sounded' really quite good for continuous high level listening and my boss used such speakers with cheap PA amp regularly for family get-togethers in local halls or outside...

 

[Koeitje]

200W?? You sure?? Because 200W is actually alot of power and is very costly as well. Most of the mainstream amps don't even hit 100W.

100W can easily clip during peaks in dynamic recordings with mainstream loudspeakers.

 

[Sgt. Ear Ache]

Does Capitalizing Each Word Make Them More Important? Sorry, couldn't resist.

My answer: maybe. Depends.

a Facebook "friend" of mine - he's actually a guy I went to school with but haven't had any actual real world interactions with since then - posts a couple of daily "updates" which are almost always profoundly boring but for some reason he capitalizes Every Single Word In The Entire Post. I find it incredibly weird and distracting but I'm so oddly fascinated by the relentless lack of anything interesting in his updates that I can't put him on ignore, lol...

now...what were we discussing??

 

[atsmusic]

100W can easily clip during peaks in dynamic recordings with mainstream loudspeakers.

years ago I had some kplisch speakers that were very efficient and I had a 200 watt into 8 ohm adcom amp. It had the clipping lights on it and would light up all the time when I was listening loud. So I think you are right.

 

[oldsysop]

Today's speakers are garbage.
There are hundreds of vintage speakers with sensitivity of 94 or higher.
That and an 80 / 100w AB amp is all it takes.

 

[David Harper]

It's also your post as well as the title.....did you leave something out like maybe the specifics for consideration/comment?

No. He didn't leave anything out. His question is plain and simple. And very good. What was it that you didn't understand about the O.P.?

 

[David Harper]

Does A More Powerful Amplifier Make Speakers Sound Better?

Good question and one I have always wondered about myself. IMO it is possible that a powerful amp can make a speaker sound better. But there are a lot of variables involved in that scenario. Not the least of which is placebo. I switched from a yammy AVR with 95 wpc to an amp with 200 wpc and I
was certain I heard a difference on my maggie speakers. But maggies are notoriously difficult so that may be why. Maybe that small amount of headroom provided by a big amp really does sound different?

 

[Doodski]

Get a minimum in 2021 of 200wpc (genuine measured power @8R /~350W@4R) and be happy for a long time.

Why settle for non-linearity in power output? The genuine measured power could be 400W@4R then 800@2. I mean where does it end? I still think a 75W-100W w/linear output is enough for most home audio nearfield listening. For a bigger room and speakers then more is better. Then we get into multiple pairs of speakers with multiple power amps and lotsa bass and a system to be heard. So not sure about the whole 350W to 500W@8R thing comes from when one can have bigger and more speakers.

 

[Berwhale]

Ooops, talk about anticlimax. Sorry about that.

Attached image NAD's own data on dynamic effect. Not too bad, though. If they are right now, that is.

Although the interesting thing for this thread is not the NAD receiver but this with dynamic effect in general.
I just used that NAD reviver as an example.

The PE on the end of 7240PE stands for Power Envelope. My understanding is that NADs approach was to rate their amps conservatively for average levels (e.g. 40W) and then use the 'PE technology' to handle peaks (e.g. 160W). I don't claim to undestand the PE technology, I remember reading about it in my youth when my brother was considering adding a NAD PE power amp to his 3020a.