• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Power amplifier tests with respect to FTC: 16 CFR Part 432 (July 5, 2024) requirements on output power claims

If 80% of amplifiers fail the FTC test, does that equate to 80% of amplifier purchasers are unhappy with their purchases (or their purchases fail to meet their requirements)? If not, what good is the FTC test?
While the test may not directly capture customer dissatisfaction, its primary purpose is to promote fairness in the market and curb the normalization of inflated claims. If the test criteria fail to reflect real-world consumer needs, reassessment or adjustment would naturally be necessary.

That said, adjusting the standards to align with current market practices would undermine the goal of transparency. Setting the bar too low risks revisiting the era of exaggerated PMPO claims from the 1990s.

Let’s avoid revisiting debates about whether the FTC rules themselves are flawed.
 
Why do you believe it's important to gather data on how many amplifiers pass this test?
Why are you asking me the obvious? If you want to regulate the market, you need to know what the impact is. FTC specifically has an obligation to gauge this impact on any regulation it establishes. From the Amplifier Rule documentation:

"The Commission believes that the proposed amendment would not have a significant economic impact upon small entities, although it may affect a substantial number of small businesses. Specifically, the proposed change in the disclosure requirements should not significantly increase the costs of small entities that manufacturer or import power amplification equipment for use in the home. Therefore, based on available information, the Commission certifies that amending the Rule asproposed will not have a significant economic impact on a substantial number of small businesses"

[...]

"C. Small Entities to Which the Proposed Amendments Will Apply The Rule covers manufacturers and importers of power amplification equipmen tfor home use. Under the Small Business Size Standards issued by the Small Business Administration, audio and video equipment manufacturers qualify as small businesses if they have 750 or fewer employees.45 The Commission’s staff estimates that a substantial number of the entities covered by the Rule likely qualify as small businesses."

It is trivial to show that many companies of this size are hurt economically if none or few can comply with this standard and still have specifications that are market competitive. The Commission has done no work whatsoever in this area. Such a project would start with a sample of such companies, and surveying them to see what percentage can pass such a test and still be in business.

I also ask because John and @pma continue to claim that these tests are easy for "real amplifiers." I want to know how many such amplifiers pass this test, not this repeated hand waiving and expecting claims to be taken as facts.
 
While the test may not directly capture customer dissatisfaction, its primary purpose is to promote fairness in the market and curb the normalization of inflated claims. If the test criteria fail to reflect real-world consumer needs, reassessment or adjustment would naturally be necessary.
If the regulation is not about customer dissatisfaction, then it has no business existing!

On inflated claims, we can trivially catch them with the type of tests I run. Government skipped way past those and invented something no one has run before. And has since the regulation was announced. In other words, the government regulation itself is inflated! Big time!
 
Let’s avoid revisiting debates about whether the FTC rules themselves are flawed.
Yeh, let's not talk about the main topic itself. :(

Look, if the EPA mandated tomorrow that every car had to get 200 miles per gallon, you think that is reasonable? I am sure people would like such a thing but that is insufficient to put the force of government behind it. The requirement needs to be realistic. Yes, it can be a stretch vs where the market is today but can't be fantastical. The way you know where you are in that range is to gather data. Not comments from the Internet. And that is what I asked for.
 
It is trivial to show that many companies of this size are hurt economically if none or few can comply with this standard and still have specifications that are market competitive.
Isn't that just part of market competition? Why should standards be lowered across the board simply because inflated claims have become the norm? Catering to the lowest common denominator only perpetuates the problem, leaving everyone none the wiser.
If the regulation is not about customer dissatisfaction, then it has no business existing!
That perspective seems overly simplistic. A customer using their Topping amp won’t necessarily feel less satisfied, regardless of how it’s rated. Regulation is more about ensuring transparency for consumers and fostering fair competition among manufacturers.
On inflated claims, we can trivially catch them with the type of tests I run.
Perhaps, but that approach would require testing every amplifier on the market, which is impractical. Additionally, polls indicate that speakers are considered more important.
In other words, relying on a few third-party testers -each using slightly different methods -is not a viable substitute for proper regulation.
Yeh, let's not talk about the main topic itself. :(
Maybe I'm not communicating my points effectively. What I meant to say is that it’s generally acknowledged that the FTC rules do not A) accurately represent real-world use cases and B) are open to interpretation. Pursuing that discussion again would only lead to circular arguments, which I’m not interested in revisiting. The horse is dead. What I find more intriguing is the pursuit of a set of tests that are fair, objective, and representative as a start for something better.
Look, if the EPA mandated tomorrow that every car had to get 200 miles per gallon, you think that is reasonable? I am sure people would like such a thing but that is insufficient to put the force of government behind it. The requirement needs to be realistic. Yes, it can be a stretch vs where the market is today but can't be fantastical. The way you know where you are in that range is to gather data. Not comments from the Internet. And that is what I asked for.
Certainly, it needs to be realistic -that’s not my point. I understand that the FTC isn’t. I’m merely cautioning against taking a snapshot of the amplifier market in 2024 and using that as the baseline for future expectations and ratings. Doing so would represent a significant regression af transparency..
 
I’m suggesting that “peak power” should be measured with 250ms of pink noise. Continuous power should also use pink noise but obviously would need to be tested over a longer period, at least 5 minutes.
I'm thinking something like this would be most informative. Multichannel amps get a bit more complicated, but sticking with a mono example for now this would be a likely view of the V3 mono from Fosi:

Picture 1.png


I've invented the intermediate values between 250ms and 600s, but having had it on my bench for a bit I'd bet I'm close. I'd also bet nearly all amps have the same shape curve, which is why only the 250ms and 600s values are needed in most cases. These were 1KHz sine waves at a 1% distortion limit, but switching to pink noise isn't hard and in this case would produce a very similar result.
 
I'm thinking something like this would be most informative. Multichannel amps get a bit more complicated, but sticking with a mono example for now this would be a likely view of the V3 mono from Fosi:

View attachment 414990

I've invented the intermediate values between 250ms and 600s, but having had it on my bench for a bit I'd bet I'm close. I'd also bet nearly all amps have the same shape curve, which is why only the 250ms and 600s values are needed in most cases. These were 1KHz sine waves at a 1% distortion limit, but switching to pink noise isn't hard and in this case would produce a very similar result.
Very nice and informative!
Edit: t may be a stretch to assume that the curves are identical across different amplifiers. I would speculate that this largely depends on the linearity of the power supply, among other factors, protection circuitry etc.
 
Last edited:
Perhaps, but that approach would require testing every amplifier on the market, which is impractical.
I said FTC could have picked a test like I run. Not that people look to me for all the tests. They are also common and familiar tests for the industry.
 
Isn't that just part of market competition? Why should standards be lowered across the board simply because inflated claims have become the norm?
Inflated claims have become the norm in Pro and Car audio industries. Not consumers. Sure, there are exceptions with XYZ cheap amp but it is not the norm in larger scale of things.

Now, if you want to claim that a 400 watt amplifier from Hypex should be rated at 12 watts because that is what it produces @20Khz for 5 minutes, then not only do you not have a valid argument, but are working to put these companies out of business -- the very thing FTC claims is not doing.

As I said and am repeating for the third time, very simple tests could have been mandated for power ratings to unify things more. I run such at test:

index.php


This was company power claims: "
  • Powerful Output: It is only 12 watts per channel, but compared to transistor amplifier, it actually can output more than 60 watts."
My test easily blows through it and provides apples vs apples comparison to other amplifiers. So it is quite effective in cutting through inflated power claims.

Net, net, standardization of power measurements is good. What is not good is standardized that are so broken that the wind up screwing up the market and mislead consumers as FTC currently does.
 
Interesting observation - ADC digital filter lies (ringing). The real difference between amplifier rated sine power amplitude and white noise clipping amplitude is in fact much lower than ADC measurements would suggest.

IMG_0807.jpeg IMG_0809.jpeg
 
Now keep the volume knob where it is and replace it with a square wave
The square wave will be 2x higher average power than the sine. That is what will destroy the speaker - raw power.


Plus - the myth normally stated is that a lower power amp is worse for speakers, because when they clip (presumably much lower power than the speaker can normally cope with) the harmonics destroy the tweeter. This is particularly what is not the case. If you have a higher power amp capable of driving the amp close to destruction - then it is just raw power causing the problem.
 
Last edited:
HiFi audio amplifiers are supposed to be linear devices.
That means they treat all input signals the same way.
If they have gain, they apply an equal amount of gain to all frequencies.
They do that, whilst not adding unrelated signals of their own.

A power amplifier's job is to produce power (by swinging voltage and delivering current).
Lots of continuous linear power.
Well designed power amplifiers don't care what frequency they are dealing with- all frequencies are treated the same.
As much power as you want, for as long as you want.

A measly 5 minutes at full rated power, at any frequency across the audible bandwidth is not unreasonable. In fact it's not much of a hurdle at all for a well designed amplifier.

Amplifiers that cannot deliver their rated power across the audible spectrum on a continuous basis are poorly designed and/or over specified. Testing to FTC conditions often uncovers this. How is that not useful to consumers? And which type do you want to buy- the poorly designed, the over specified, or the better alternative?
Yes but why should I as a consumer care if an amplifier can deliver max power at 20khz for five minutes when there really ain't any realistic scenario ever when I'd need it?
Of course I want an amplifier that is well designed but in what way would an unrealistic use be a proof of that? I mean yeah sure IF it can handle that it might be proof that it's well built, but it's not like it's proof it's badly built either if it can't handle that since there's no real life use for or anyways.

Didn't Hypex themselves say that if they have to comply with FTC they need to spec their was it maybe 200w amplifier as 12w, but then when a consumer then buys it and plays music and then notice it can push out 200w of power while playing music, and can do that for years, then they'd realised that that FTC rule is just dumb and missleading since it doesn't have anything to do with real life.
 
I've invented the intermediate values between 250ms and 600s, but having had it on my bench for a bit I'd bet I'm close. I'd also bet nearly all amps have the same shape curve, which is why only the 250ms and 600s values are needed in most cases.
A stress test using sine waves needs to be much shorter duration compared to music. If you are going to cook the amp with such tests, I see no reason to go beyond 30 seconds.
 
Now, if you want to claim that a 400 watt amplifier from Hypex should be rated at 12 watts because that is what it produces @20Khz for 5 minutes, then not only do you not have a valid argument, but are working to put these companies out of business -- the very thing FTC claims is not doing.
That is not at all what I’m arguing, Amir.
Perhaps I’m not expressing my point clearly, or maybe you’re taking a defensive stance. I’ve also mentioned several times that I agree on the need for more realistic testing.
Net, net, standardization of power measurements is good. What is not good is standardized that are so broken that the wind up screwing up the market and mislead consumers as FTC currently does.
And I agree with that.
 
A stress test using sine waves needs to be much shorter duration compared to music. If you are going to cook the amp with such tests, I see no reason to go beyond 30 seconds.
He missed an m, I believe...:)
It was supposed to be 600ms. I don't think he meant 10 min.


Edit: It was indeed 600 seconds.
 
Last edited:
Interesting observation - ADC digital filter lies (ringing). The real difference between amplifier rated sine power amplitude and white noise clipping amplitude is in fact much lower than ADC measurements would suggest.

View attachment 414995 View attachment 414996
Question: why do you have such a toy $200 oscilloscope? This is a "real" scope:

Keysight_DSOX2022A-1.jpg



Sure, it costs 10 times more than yours. And has bandwidth and sampling rate you don't need for audio. But you know, we are talking about toys vs real instruments!
 
I’m suggesting that “peak power” should be measured with 250ms of pink noise. Continuous power should also use pink noise but obviously would need to be tested over a longer period, at least 5 minutes.
You can't measure power limits with pink noise, because you can't measure the THD of noise.
 
You can't measure power limits with pink noise, because you can't measure the THD of noise.
We can with FSAF and it looks like this:

1734596032937.png

1734595724489.png


Any noise can be used (the above is white) ,even music,it's so good REW has a brand new window for it.

Problem is that it will have to be implemented in the known analyzers software and also be able to cross-check in real time.
I suspect it will be time-hungry.

Seems doable though.
 
HiFi audio amplifiers are supposed to be linear devices.
That means they treat all input signals the same way.
If they have gain, they apply an equal amount of gain to all frequencies.
They do that, whilst not adding unrelated signals of their own.

A power amplifier's job is to produce power (by swinging voltage and delivering current).
Lots of continuous linear power.
Well designed power amplifiers don't care what frequency they are dealing with- all frequencies are treated the same.
As much power as you want, for as long as you want.

A measly 5 minutes at full rated power, at any frequency across the audible bandwidth is not unreasonable. In fact it's not much of a hurdle at all for a well designed amplifier.

Amplifiers that cannot deliver their rated power across the audible spectrum on a continuous basis are poorly designed and/or over specified. Testing to FTC conditions often uncovers this. How is that not useful to consumers? And which type do you want to buy- the poorly designed, the over specified, or the better alternative?
There's something about chasing engineering excellence in this argument - just like DACs chasing 120+ SINAD. It's admirable, I want it but I'm not going to pay over the odds for it when there is no practical benefit.

Some people might want that, that's cool. I get the engineering excellence (or competence if you prefer) angle.

What I do want is a reliable way to compare amplifiers.

What I want is to know (as far as reasonably possible) that my next amplifier will go as loud as I think it should, won't keep clipping or dropping off into protection and will keep doing that for it's full lifetime without unexpectedly burning it's insides out. I don't want to think about it.

An amplifier that was FTC rated at 100W would do that (obviously I'd look at the features and other measurements, not just one metric). I couldn't afford it though, and it would be over specified for me - a waste of money and engineering and a poor choice.

What are those real world test that tell me what I need to know?

There are encouraging signs in this thread, as well as entrenched positions and some unbecoming language.

FTC regulations exist. Fine. Let's move on.
 
Back
Top Bottom