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

[amirm]

Deliberately choosing not to follow regulations is deceptive.

So you are deceptive and dishonest when you exceed the speed limit with your car?

 

[restorer-john]

@Old_School_Brad There will be honest manufacturers (like McIntosh and Accuphase) and there will be a sea of manufacturers who decide to keep doing what they are doing, that is being blatantly, deliberately, deceptive. They'll keep trying to get away with deception for as long as they can because compliance costs money. It increases costs and produces a better product for consumers. Products that will last longer. They don't want that.

Vote with your wallet- don't buy cheap, overrated garbage sold with ludicrous claims. Use Amir's testing as a helpful guide and do your own extensive testing if you have doubts. Call out the BS when you find it.

 

[amirm]

That interpretation is incorrect.

And you know that how? You consulted the FTC and they said mere simulation is good enough?

 

[Old_School_Brad]

So you are deceptive and dishonest when you exceed the speed limit with your car?

That's not quite the same thing. My choice of words might not be semantically accurate.

Dishonesty involves deliberate misrepresentation, while deception refers to the intent to mislead.

Therefore, opting not to comply with the FTC regulations that they are legally required to follow would be considered deceptive, rather than dishonest.

And you know that how?

It is evident from the way it is written.

 

[amirm]

Therefore, opting not to comply with the FTC regulations that they are legally required to follow would be considered deceptive, rather than dishonest.

A regulation must be enforced and strongly for it to have any impact on the market. No company wants to volunteer to either derate their power, or increase cost/retail price when the rest of their competitors are not. Their business will sink while their competitors smile all the way to the bank. A wise business decision would call for ignoring the regulation until someone comes knocking on your door.

 

[Old_School_Brad]

A regulation must be enforced and strongly for it to have any impact on the market. No company wants to volunteer to either derate their power, or increase cost/retail price when the rest of their competitors are not. Their business will sink while their competitors smile all the way to the bank. A wise business decision would call for ignoring the regulation until someone comes knocking on your door.

I understand that they may not want to, but they actually must comply if they intend to follow the law.

What leads you to believe that some other competitors aren't complying? It feels like you're sitting on some insider knowledge that I’m not aware of, and it doesn’t sit well with me, as it suggests, going back to your speeding reference, that if others are speeding, it’s acceptable for me to do the same. That reasoning and absence of regulation will only benefit dishonest manufacturers. In the meantime, average consumers will genuinely believe they have a compact half cigar box enclosure capable of delivering 1400 W.

Also, choosing to deliberately ignore legally required regulations seems like an unwise business decision and poor advice, in my opinion.

 

[DonH56]

The rule is below (as was provided earlier by @EJ3 and probably others before; this is from the copy of the Federal Register that one of the FTC authors sent me last year -- I don't have the full thing downloaded, just the update in the Federal Register).

"Shall be obtainable" means products must be able to attain their rated power for five minutes and achieve no more than 1% THD+N measured at the end of those five minutes for all frequencies from 20 Hz to 20 kHz. There is nothing I see about what testing is required to verify compliance. The title of the section is "Standard test conditions" which implies this is something they must test. A legal interpretation of "shall be obtainable" is that manufacturers must test "everything" to meet the spec, but the scope of testing is open for interpretation. That is the problem: from (e) one manufacturer could interpret this to mean they must test every frequency and power level from 250 mW to rated power for five minutes and measure THD+N, which is impractical, while another could perform no testing at all and certify their product can meet the rule based on whatever criteria they establish, again an unreasonable approach. In practice some intermediate compromise would likely be reached in conjunction with the FTC.

If I was designing an amp to meet this spec, I would first contact the FTC for clarification about their intent and what is required to verify compliance to the spec. That would likely lead to an explanation of design characterization, product line testing, and back and forth conversations about exactly what is possible to test and hopefully some clarity added to the spec. I would also clarify some additional points such as number of harmonics included in THD, noise bandwidth, and if they really mean THD+N or SINAD (or how they define "noise") and what procedure they want used to measure THD+N (there are several, again depending upon how noise is defined), confirm they mean a purely resistive 8-ohm load impedance, and so forth.

As an aside, I would have include verbiage requiring the same compliance test for any other rated impedances such as 4-ohm loads.

Federal Register /Vol. 89, No. 114 /Wednesday, June 12, 2024 /Rules and Regulations 49801

§ 432.3 Standard test conditions.
* * * * *
(e) Any power level from 250 mW to
the rated power shall be obtainable at all
frequencies within the rated power band
of 20 Hz to 20 kHz without exceeding
1.0% of total harmonic distortion plus
noise (THD+N) at an impedance of 8
ohms after input signals at said
frequencies have been continuously
applied at full rated power for not less
than five (5) minutes at the amplifier’s
auxiliary input, or if not provided, at the
phono input. Provided, however, that
for amplifiers utilized as a component
in a self-powered subwoofer in a selfpowered
subwoofer-satellite speaker
system that employs two or more
amplifiers dedicated to different
portions of the audio frequency
spectrum, any power level from 250
mW to the rated power shall be
obtainable at all frequencies within the
subwoofer amplifier’s intended
operating bandwidth without exceeding
1.0% of total harmonic distortion plus
noise (THD+N) at an impedance of 8
ohms after input signals at said
frequencies have been continuously
applied at full rated power for not less
than five (5) minutes at the amplifier’s
auxiliary input, or if not provided, at the
phono input.
* * * * *

 

[Old_School_Brad]

The rule is below (as was provided earlier by @EJ3 and probably others before; this is from the copy of the Federal Register that one of the FTC authors sent me last year -- I don't have the full thing downloaded, just the update in the Federal Register).

"Shall be obtainable" means products must be able to attain their rated power for five minutes and achieve no more than 1% THD+N measured at the end of those five minutes for all frequencies from 20 Hz to 20 kHz. There is nothing I see about what testing is required to verify compliance. The title of the section is "Standard test conditions" which implies this is something they must test.

I don't see how the section title suggests any implications; the text that follows is what truly matters.

A legal interpretation of "shall be obtainable" is that manufacturers must test "everything" to meet the spec,

Is it? I don't think that's accurate.

The phrase suggests that something should be accessible or available, but it does not define the methods by which that accessibility is attained.

 

[kyuu]

The phrase suggests that something should be accessible or available, but it does not define the methods by which that accessibility is attained.

This is a really weird hill you're dying on here. How exactly can a manufacturer claim they meet the specified parameters without some sort of testing regime? The notion that they can verify compliance with some "simulations" or whatever is absolutely silly and you know it. Manufacturers complying in that way would make the regulation worse than meaningless. Even with testing, there has to be a standard testing regime for every manufacturer to follow. If there's really multiple valid ways to interpret and test for compliance, then the resulting numbers would not be comparable with one another and the regulation is meaningless.

 

[Old_School_Brad]

This is a really weird hill you're dying on here. How exactly can a manufacturer claim they meet the specified parameters without some sort of testing regime? The notion that they can verify compliance with some "simulations" or whatever is absolutely silly and you know it. Manufacturers complying in that way would make the regulation worse than meaningless. Even with testing, there has to be a standard testing regime for every manufacturer to follow. If there's really multiple valid ways to interpret and test for compliance, then the resulting numbers would not be comparable with one another and the regulation is meaningless.

I addressed that specific point in post no. 1660 in response to Amir.

 

[Doodski]

I addressed that specific point in post no. 1660 in response to Amir.

Are referring to for example let's say the likes of Sony which to my understanding via computer can design, layout and then start the assembly machines without prototyping. Then using the design parameters to state whether compliance is in effect or not?

 

[Old_School_Brad]

Are referring to for example let's say the likes of Sony which to my understanding via computer can design, layout and then start the assembly machines without prototyping. Then using the design parameters to state whether compliance is in effect or not?

Yes, that’s an example.
However, my point on this is exaggerated, meant to illustrate that the regulation does not actually mandate manufacturer testing. In practice, manufacturers would likely conduct some physical testing to verify performance.

 

[rdenney]

...

In my work/world, "continuous" means "for the life of the product". Since I have worked on circuits for things like space applications and mission-critical components for the medical and military fields, if something was rated for continuous operation, it had to pass stringent design reviews and testing to prove it could handle that. Low-level circuits rarely had issues passing lifetime tests. For high-power RF amplifiers, not something I designed, it was common to have a maximum continuous ("unlimited" time of operation) power rating, and a peak rating with defined peak (pulse) power and duration specified -- including a "cool-down" period between max power peaks. Things like commercial/military radar systems and medical scanners typically had such ratings. Early ultrasound scanners had some serious heat problems, for example, when the handheld scanner got hot on a lady's tummy.

...

From a defense department or space program perspective, the rule probably seems reasonable, because those transmitters and amplifiers will be sending indefinitely continuous data streams that are fairly constant in signal level, and because the consequences of failure are extreme.

But let's consider something more realistic in high-power RF amplifiers: Amateur radio. Amateur radio amplifiers are routinely rated for intermittent output. The terms is ICAS--Intermittent Commercial and Amateur Service. For transmitter amplifiers, ICAS is something like a 50% duty cycle, and even that is used for single-sideband modulation. SSB amps are rated for peak envelope power, which answers the question of whether the peaks of my speaking into a microphone will clip the amp. FCC rules for maximum allowable power are also measured as peak envelope power. That power is calculated from the current delivered at the peak voltage in the modulated signal. Of course, in RF sideband operation, the audio signal modulates an RF carrier only on one side of the ground reference. The receivers construct the negative signal. For AM operation, where both positive and negative signals are modulated, the power is derated to 25% of the SSB PEP rating. For RF amplifiers intended for continuous service, such as broadcasting a continuous RTTY data stream or for FM broadcast where the carrier is continuous, the rating structure would be CCS, or Continuous Commercial Service. That's the same rating that would be used for indefinite continuous key-down for CW (i.e., morse code "continuous wave"), but, of course, CW is intermittent in actual use. We use key-down to tune the loading of vacuum tubes for RF transmitters, but we all know not to exceed 6 or 8 seconds if we don't want to see holes magically appearing in the plates on the tubes. Learning how to retune the tube loading after a change in frequency quickly is one of the skills. Amps rated for indefinite continuous key-down are grossly overdesigned for ICAS, and usually have a PEP for sideband signals as much as twice the key-down rating.

Examples of high-end, high-power Acom amplifiers:

* Power Output: 1,500W PEP, 1,200W continuous carrier; for continuous carrier transmissions longer than 15 minutes the external auxiliary fan must be installed

This one costs about 75% more:

* Power Output: 1,500W PEP, 1,500W continuous carrier; for continuous carrier transmissions longer than 15 minutes the external auxiliary fan must be installed

Note that they never report PEP greater than 1500 watts, because that's the maximum legal limit for amateur radio. But the second of these will easily produce 2000-2500 watts PEP for sideband or normal CW in ICAS.

The amp I use is a lower-powered (but still high-quality) Ten-Tec Centaur, rated at 600W PEP ICAS. It's published rating is 600W PEP SSB/CW, 350W RTTY.

Hams that buy amps for sideband operation can get a 1500-watt PEP ICAS amp that will work perfectly for them at the maximum legal power output. If they are CW operators, that amp will also still work fine, because the spaces between the dots and dashes, and the time spend listening to the other party, reduce the duty cycle. If they want an amp for continuous RTTY broadcast or for AM operation at peak envelop power, they'll need to spend more but the ratings that support those use cases are provided.

Point being, for intermittent applications or those that use non-continuous signals, there are standards for rating power output that are appropriate to the application. The issue isn't whether an amp's FTC rating doesn't represent what it can do in the worst scenario (at least for five minutes), the issue is whether that value actually tells uneducated consumers something important about playing music loud. Clipping occurs when an amp can't fill a peak waveform, not whether it can't fill a continuous waveform indefinitely at some much reduced amplitude.

Personally, I think a single number is foolish. Even fuel economy for cars has two ratings plus a "combined" rating that everyone knows is just a model that may or may not apply to the way they drive. But if I had to have a single number, I'd want it to represent peak output when playing worst-case actual program material, and as long as the signal is actual program material (or some reasonable model thereof), it should operate indefinitely without exceeding the rated temperature of the hot bits.

Rick "appreciates Don's perspective but the FTC isn't rating things for guys like him" Denney

 

[RickS]

A regulation must be enforced and strongly for it to have any impact on the market. No company wants to volunteer to either derate their power, or increase cost/retail price when the rest of their competitors are not. Their business will sink while their competitors smile all the way to the bank. A wise business decision would call for ignoring the regulation until someone comes knocking on your door.

Agreed, there is risk of doing business that I experienced in corporate decision making all the time. At one point, our decision making got so risk averse that our major customers complained. Once the legal department reviewed, we adjusted our approach.

For a major manufacturer, there are so many issues that if you try to address all of them, you likely will do nothing else and risk going out of business. Risk assessment is just part of the job. Even if a regulation had more enforcement than this one does, if the investment to conform is more expensive than the potential penalty, management might take the risk. Not endorsing bad behavior, but risk benefit analysis happens all the time.

 

[Old_School_Brad]

From a defense department or space program perspective, the rule probably seems reasonable, because those transmitters and amplifiers will be sending indefinitely continuous data streams that are fairly constant in signal level, and because the consequences of failure are extreme.

But let's consider something more realistic in high-power RF amplifiers: Amateur radio. Amateur radio amplifiers are routinely rated for intermittent output. The terms is ICAS--Intermittent Commercial and Amateur Service. For transmitter amplifiers, ICAS is something like a 50% duty cycle, and even that is used for single-sideband modulation. SSB amps are rated for peak envelope power, which answers the question of whether the peaks of my speaking into a microphone will clip the amp. FCC rules for maximum allowable power are also measured as peak envelope power. That power is calculated from the current delivered at the peak voltage in the modulated signal. Of course, in RF sideband operation, the audio signal modulates an RF carrier only on one side of the ground reference. The receivers construct the negative signal. For AM operation, where both positive and negative signals are modulated, the power is derated to 25% of the SSB PEP rating. For RF amplifiers intended for continuous service, such as broadcasting a continuous RTTY data stream or for FM broadcast where the carrier is continuous, the rating structure would be CCS, or Continuous Commercial Service. That's the same rating that would be used for indefinite continuous key-down for CW (i.e., morse code "continuous wave"), but, of course, CW is intermittent in actual use. We use key-down to tune the loading of vacuum tubes for RF transmitters, but we all know not to exceed 6 or 8 seconds if we don't want to see holes magically appearing in the plates on the tubes. Learning how to retune the tube loading after a change in frequency quickly is one of the skills. Amps rated for indefinite continuous key-down are grossly overdesigned for ICAS, and usually have a PEP for sideband signals as much as twice the key-down rating.

Examples of high-end, high-power Acom amplifiers:

* Power Output: 1,500W PEP, 1,200W continuous carrier; for continuous carrier transmissions longer than 15 minutes the external auxiliary fan must be installed

This one costs about 75% more:

* Power Output: 1,500W PEP, 1,500W continuous carrier; for continuous carrier transmissions longer than 15 minutes the external auxiliary fan must be installed

Note that they never report PEP greater than 1500 watts, because that's the maximum legal limit for amateur radio. But the second of these will easily produce 2000-2500 watts PEP for sideband or normal CW in ICAS.

The amp I use is a lower-powered (but still high-quality) Ten-Tec Centaur, rated at 600W PEP ICAS. It's published rating is 600W PEP SSB/CW, 350W RTTY.

Hams that buy amps for sideband operation can get a 1500-watt PEP ICAS amp that will work perfectly for them at the maximum legal power output. If they are CW operators, that amp will also still work fine, because the spaces between the dots and dashes, and the time spend listening to the other party, reduce the duty cycle. If they want an amp for continuous RTTY broadcast or for AM operation at peak envelop power, they'll need to spend more but the ratings that support those use cases are provided.

Point being, for intermittent applications or those that use non-continuous signals, there are standards for rating power output that are appropriate to the application. The issue isn't whether an amp's FTC rating doesn't represent what it can do in the worst scenario (at least for five minutes), the issue is whether that value actually tells uneducated consumers something important about playing music loud. Clipping occurs when an amp can't fill a peak waveform, not whether it can't fill a continuous waveform indefinitely at some much reduced amplitude.

Personally, I think a single number is foolish. Even fuel economy for cars has two ratings plus a "combined" rating that everyone knows is just a model that may or may not apply to the way they drive. But if I had to have a single number, I'd want it to represent peak output when playing worst-case actual program material, and as long as the signal is actual program material (or some reasonable model thereof), it should operate indefinitely without exceeding the rated temperature of the hot bits.

Rick "appreciates Don's perspective but the FTC isn't rating things for guys like him" Denney

I don’t see anyone suggesting that a single FTC value should replace all the others.
Continuous power ratings help keep manufacturers honest by preventing exaggerated claims. While the FTC rating might not perfectly match real-world use, switching or keeping a peak-based rating wouldn’t help, since many already misunderstand the vastly different peak power numbers that stem from all the different test durations, from 20 ms to 2-5 sec.

Just like fuel economy ratings consider different driving conditions, amplifier ratings should reflect both steady and peak performance instead of relying on just one number.
Since the FTC rule already sets a basic standard, the solution isn’t to eliminate continuous testing but to enhance it by including dynamic ratings in the evaluation -these ratings are already offered in various forms, while continuous testing is not.

 

[MaxwellsEq]

Clipping happens during short bursts, but an amp that struggle under continuous load may also struggle with certain peaks, depending on the duration

I've not seen that happen. Are there examples?

 

[Old_School_Brad]

I've not seen that happen. Are there examples?

That was a mistake on my part. You caught me before I had a chance to edit it out.

 

[MaxwellsEq]

That was a mistake on my part. You caught me before I had a chance to edit it out.

Well you may be right. If an amplifier has a fully regulated PSU, it may have no peak capacity over the sustained power. I've not seen it, though.

 

[DonH56]

From a defense department or space program perspective, the rule probably seems reasonable, because those transmitters and amplifiers will be sending indefinitely continuous data streams that are fairly constant in signal level, and because the consequences of failure are extreme.

But let's consider something more realistic in high-power RF amplifiers: Amateur radio. Amateur radio amplifiers are routinely rated for intermittent output. The terms is ICAS--Intermittent Commercial and Amateur Service. For transmitter amplifiers, ICAS is something like a 50% duty cycle, and even that is used for single-sideband modulation. SSB amps are rated for peak envelope power, which answers the question of whether the peaks of my speaking into a microphone will clip the amp. FCC rules for maximum allowable power are also measured as peak envelope power. That power is calculated from the current delivered at the peak voltage in the modulated signal. Of course, in RF sideband operation, the audio signal modulates an RF carrier only on one side of the ground reference. The receivers construct the negative signal. For AM operation, where both positive and negative signals are modulated, the power is derated to 25% of the SSB PEP rating. For RF amplifiers intended for continuous service, such as broadcasting a continuous RTTY data stream or for FM broadcast where the carrier is continuous, the rating structure would be CCS, or Continuous Commercial Service. That's the same rating that would be used for indefinite continuous key-down for CW (i.e., morse code "continuous wave"), but, of course, CW is intermittent in actual use. We use key-down to tune the loading of vacuum tubes for RF transmitters, but we all know not to exceed 6 or 8 seconds if we don't want to see holes magically appearing in the plates on the tubes. Learning how to retune the tube loading after a change in frequency quickly is one of the skills. Amps rated for indefinite continuous key-down are grossly overdesigned for ICAS, and usually have a PEP for sideband signals as much as twice the key-down rating.

Examples of high-end, high-power Acom amplifiers:

* Power Output: 1,500W PEP, 1,200W continuous carrier; for continuous carrier transmissions longer than 15 minutes the external auxiliary fan must be installed

This one costs about 75% more:

* Power Output: 1,500W PEP, 1,500W continuous carrier; for continuous carrier transmissions longer than 15 minutes the external auxiliary fan must be installed

Note that they never report PEP greater than 1500 watts, because that's the maximum legal limit for amateur radio. But the second of these will easily produce 2000-2500 watts PEP for sideband or normal CW in ICAS.

The amp I use is a lower-powered (but still high-quality) Ten-Tec Centaur, rated at 600W PEP ICAS. It's published rating is 600W PEP SSB/CW, 350W RTTY.

Hams that buy amps for sideband operation can get a 1500-watt PEP ICAS amp that will work perfectly for them at the maximum legal power output. If they are CW operators, that amp will also still work fine, because the spaces between the dots and dashes, and the time spend listening to the other party, reduce the duty cycle. If they want an amp for continuous RTTY broadcast or for AM operation at peak envelop power, they'll need to spend more but the ratings that support those use cases are provided.

Point being, for intermittent applications or those that use non-continuous signals, there are standards for rating power output that are appropriate to the application. The issue isn't whether an amp's FTC rating doesn't represent what it can do in the worst scenario (at least for five minutes), the issue is whether that value actually tells uneducated consumers something important about playing music loud. Clipping occurs when an amp can't fill a peak waveform, not whether it can't fill a continuous waveform indefinitely at some much reduced amplitude.

Personally, I think a single number is foolish. Even fuel economy for cars has two ratings plus a "combined" rating that everyone knows is just a model that may or may not apply to the way they drive. But if I had to have a single number, I'd want it to represent peak output when playing worst-case actual program material, and as long as the signal is actual program material (or some reasonable model thereof), it should operate indefinitely without exceeding the rated temperature of the hot bits.

Rick "appreciates Don's perspective but the FTC isn't rating things for guys like him" Denney

I have said that high-power RF amps often include intermittent power ratings with on and off (cool-down) duty cycle periods specified. Very common, and would work with the intent (I think) of the FTC rule: continuous power rating, then add a peak power rating that is adequately defined (e.g. the IHF 20 ms burst test) for headroom. You are reading too much into my post without reviewing prior context. Which is the same problem I have and why I should really just stay out of this mess; I am too lazy to post a book with all relevant experience, and then end up being lectured because my responses are either not clear or insufficiently detailed, about a topic I appreciate but realistically is just a bunch of keyboard warriors (myself included) whining about.

I have not renewed my Ham license in many years, another expensive hobby I have not gone back to, but have plenty of experience with Ham radios, commercial radio/TV systems, medical transmitters (e.g. 60 MHz ultrasound scanners), and government (military and otherwise) systems. I like seeing a continuous power rating as it has implications for other things like thermal management, then a headroom spec based on (defined) short-term output, would cover amplifiers for music and other (e.g. radar, communication, telemetry) applications. A radar transmitter is a good example of a design having very high peak power with much lower continuous power ratings. That is why the audio comparison. And you have already provided example where continuous power matters.

My argument with the FTC spec is lack of a clear definition both in scope and compliance testing. I think I have been consistent but this has gone on for so long I am sure someone could find something somewhere. If the FTC opens up a new comment period, I will contribute as I did last time. In this case, I think part of the problem is that people with knowledge of test standards assumed a base experience level not reflected in the final rule.

 

[Old_School_Brad]

Well you may be right. If an amplifier has a fully regulated PSU, it may have no peak capacity over the sustained power. I've not seen it, though.

No, no. I was just typing faster than my brain was working. That’s what happened there..

It's not unusual for some amplifiers to have little to no peak power capacity compared to their continuous power. In my view, this may be more related to the continuous test duration not being truly continuous or the fact that the "continuous" test duration is too similar to the peak power test duration.