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

[amirm]

1. Can the amplifier deliver its FTC rated power and rated distortion on a continuous* basis at 20Hz?
2. Can the amplifier deliver its FTC rated power and rated distortion on a continuous* basis at 20,000Hz?

*5 minutes minimum.

This is not compliant with regulation. It says that "...at said frequencies have been continuously applied at full rated power for not less than five (5) minutes..." Frequencies is plural. It clearly implies the amplifier needs to be subjected to wideband signal with bandwidth from 20 to 20 kHz.

In the old spec, you could say how you tested things. In the new one, you are not allowed.

As I mentioned, a proper specification would give precise instructions on how the test needs to be run with example. The regulation as written is ambiguous.

 

[amirm]

Reckon that there should still be minimum quality standards. Consumers should not be hoodwinked with false information, and products should be fit for purpose.
It is really just a case of deciding what the suitable standards are.

In the case of TI/Infenion chip-based desktop amps, and Hypex/Purifi modules for larger amps, you have plenty of that due to companies performing substantial design verification. This is far more than Joe designer performs on some class AB design of his own. This is why the satisfaction rate is so high for these amps, despite the companies putting them in finished goods don't have such resources.

I test these amps and brutally push them and always marvel at how they put up with all that. At times I make a mistake and put in 10 volts instead of 1 volt and amps just shut down or clip severely. No smoke comes out.

Importantly, for those of you who based your decision on my reviews, the tests are standardized and data can fully be compared between products.

Doubtless it would involve comprise, but really should not be so difficult surely.

Alas, it is. It is outside of the design envelope of many amplifiers to produce power near/at clipping for 5 minutes straight when no consumer ever uses their amp that way. The government is pushing for amplifies to be overdesigned whether you need or want it.

 

[amirm]

Yes, that is the huge difference between consumer audio electronics and industrial electronics. Says me who has been working since 1979 directly or in relation with the field of industrial electronics, both in military and civilian area.

Hence the reason your comments are frequently not applicable. Specialized gear is not cost sensitive but mission critical. What goes into them is vastly different than high volume mass consumer products where costs/size matter and reliability does not need to be assured to that degree. I have managed engineering at both and they approach could not be more different. We would sell digital effect/processing systems to major TV networks that would get used live and at major events like the Olympics. Typical box would cost $30,000 so we had plenty of margin. But then thing had to work or we would get thrown out of a major network operation. No one in that space puts a fosi device in the loop. Different business. Different requirement.

Long time back we were selling computers to Nasa for flight simulation. They were used on the Shuttle program. All the gear they were using from us were long discontinued. They had tested them, and verified them so they used them. New gear no matter how much better, was unproven and could not be deployed without years of testing. Again, different world than consumer products where the typical product cycle is as fast as 6 months.

 

[Old_School_Brad]

Alas, it is. It is outside of the design envelope of many amplifiers to produce power near/at clipping for 5 minutes straight when no consumer ever uses their amp that way. The government is pushing for amplifies to be overdesigned whether you need or want it.

There's no need for them to "overdesign" the amplifier. The manufacturer simply needs to disclose the amplifier's FTC-rated power. No redesign of the amplifier is required -just the addition of a new specification. For example, if an amplifier marketed as 100 W fails to meet FTC rules at that level but complies at 50 W, the manufacturer can list the FTC-rated power as 50 W. This won't affect the consumer who uses this amplifier, as its performance for them remains unchanged. The only difference is that future amplifiers could be compared using the FTC standard. The manufacturer could still include details like 1 ms burst power further down in the brochure if desired.

 

[Sokel]

(as a side note,this FSAF measurement seems to open its own can of worms,can the experienced ones give it a try please?Preferably with a known DAC and pink noise or music,cause what I see is strange... )

 

[Mnyb]

Alas, it is. It is outside of the design envelope of many amplifiers to produce power near/at clipping for 5 minutes straight when no consumer ever uses their amp that way. The government is pushing for amplifies to be overdesigned whether you need or want it.

So amp should not function with a valid input signal indefinitely ? would not this be the proper design margin for longevity ? Producing exactly for the consumer need is not really possible ( where and what music how loud ? ) also one must factor in component variations etc so some design margin is needed ? Just like an elevator does not fall to the bottom if 11people instead of 10 people steps in ?

 

[pma]

Importantly, he is testing at just 1 kHz??? What happened to 20 to 20 kHz for 5 minutes?

Maybe he does it better in the last amp, the NC252MP which he put in a giant case:

No, not at just 1kHz. The reasoning has turned to something like "presidential campaign".

PMA NC252MP power amplifier

4. Measurements and specifications (with respect to IEC 60268-3)

Article 4.3.4. Maximum effective output power as a function of power and frequency

https://pmacura.cz/NC252MP_IEC_powerfreq_4R_BW45kHz.jpg

Article 4.3.5. Maximum distortion limited (0.1%) output power - 240W/4ohm, one channel driven

https://pmacura.cz/NC252MP_IEC_powerdistlimit_4R_BW45kHz.jpg

Article 4.4. Characteristics of d.c. offset protection circuits
One can see that NC252MP transfers DC average level of 10.69V to the output, without any action of the module DC protection. This is dangerous and may lead to speaker voice coil damage in case of DCerror signal at the amplifier input. After 1s, when the input signal DC jumps of 20dB, the protection circuit finally takes action. It shuts down the amplifier output in 7.8ms after heavy DC clipping.
Further test is NC252MP response to gradually increased input DC voltage. The protection circuit took action after the output DC was as high as 26V.

https://pmacura.cz/NC252MP_outputDCoffset_test2_sm.jpg

This is tested with respect IEC standard and covers much more tests than a mere 1kHz power test.

 

[amirm]

So amp should not function with a valid input signal indefinitely ?

Valid signal is music. Not 20 to 20 kHz constant sweep for 5 minutes. To be sure, you can design an amplifier that can do this for 100 hours or whatever You just have to be willing to pay for it the extra measure. And put up with larger enclosure, fan noise, etc. So it doesn't come for free. Currently the market doesn't demand this as folks are quite happy with what they are getting.

 

[pma]

So amp should not function with a valid input signal indefinitely ?

They must survive any kind of rated input voltage range + defined margin and they must survive overloading (current or thermal) protected by a proper function of protection circuits.

IEC overload test requests this:

This is in fact +10dB overload.

3.3.2. Method of measurement
a) The amplifier is brought under standard measuring conditions. (= 10dB below rated power)
b) The source e.m.f. is increased by 20 dB during a time interval of less than one-quarter
period of the input signal and kept at this value for approximately 1 s.
c) The source e.m.f. is then reduced to its initial value during a time interval of less than onequarter
period of the input signal.
d) The time that passes before both the positive and the negative output peak voltages have
reached their final value, within 1 dB unless otherwise specified, is measured by means of
a suitable calibrated oscilloscope.

 

[amirm]

This is tested with respect IEC standard and covers much more tests than a mere 1kHz power test.

We are not discussing IEC. We are discussing FTC regulation. Your review starts with this:

"Hypex NC252MP class D amplifier review

This test review was performed to comply with FTC: 16 CFR Part 432 (July 5, 2024) - Power Output Claims for Amplifiers Utilized in Home Entertainment Products"

There is not one test in there that follows FTC regulation as stated. You don't even use 8 ohm load for heaven's sake.

 

[amirm]

would not this be the proper design margin for longevity ?

It would be but this regulation doesn't get you there. Let's say I run the Fosi amp per FTC and it shuts down at 1 minute instead of 5. It is self protecting so longevity is there to be had. To pass the regulation, they would have to stick a fan in there which likely will fail over time. The fan would come on and allow you to push the amp even more. How is this leading to more design margin?

Keep in mind that the purpose of FTC is NOT at all to improve reliability. It aims to solve an advertising/marketing problem. To that end, no research was performed to see if passing this test does any good for reliability of amps.

 

[amirm]

also one must factor in component variations etc so some design margin is needed ? J

This is all baked in already. TI for example has done more testing and design verification of the IC than you can possibly imagine compared to what goes on in a discrete amplifier design. They are ready for people to screw things up in the implementation in every possible way so the IC comes with very robust protection and extra margin. Sure there are some components around it but the FTC test is not going to magically test their margin.

Same is true of Hypex/Purifie modules. They likely have gone through sophisticated modelling as far as cooling and design variation with monte carlo simulations and such. They need these modules to not fail in the market or they face huge warranty costs.

To get more reliable design, you need to outdo the above entities with extensive analysis and large scale failure analysis. Running a 5 minute stress test is just not going to do it when it doesn't represent actual usage scenario.

 

[amirm]

Just like an elevator does not fall to the bottom if 11people instead of 10 people steps in ?

That's what the protection circuit does today. Government says that is not good enough. Said elevator needs to be tested with 50 people jammed in there even though it will never have that kind of usage.

 

[Sokel]

It would be but this regulation doesn't get you there. Let's say I run the Fosi amp per FTC

Apparently they do sell it with fans too,just external ones,or fans separately described as "fan for V3 mono" or similar.
Market will always find a way

https://www.aliexpress.com/item/1005007310872364.html

 

[Old_School_Brad]

It would be but this regulation doesn't get you there. Let's say I run the Fosi amp per FTC and it shuts down at 1 minute instead of 5. It is self protecting so longevity is there to be had. To pass the regulation, they would have to stick a fan in there which likely will fail over time. The fan would come on and allow you to push the amp even more. How is this leading to more design margin?

Fans wouldn't be necessary. In this hypothetical scenario, the amplifier fail the test at its rated power but would comply at a lower power level, which would then be designated as its FTC-compliant rating -with the unknown margin that amounts to under various loads and music signals in a real use case.

 

[pma]

Importantly, even with that giant case and amp he has built, he could not go past 4 minutes! So even if you forced manufacturers to overbuild a class D amplifier this much, and have the genius of @pma behind putting it together, you still cannot pass the FTC test at 1 kHz. Heaven help you at 20 Hz!

The "giant case" (3U 19" is nothing giant) does not help, because the design flaw is in the on-board SMPS. The heatsink has remained cold (maybe slightly warmer than before the test) during the test. The problem is that SMPS is undersized to supply amplifier rated at 2x250W/4ohm. The supply overheats and collapses. There is no reliable heat transfer from the SMPS parts to the module Al base plate (and thus to the heatsink). I have already explained it before, obviously without effect. The amplifier could be rated as 2 x 200W/4ohm, then it is able to make it with this proper heatsink.

One more note, a traditional class AB amp, placed in the similar case and with same rating 2x250W/4R and same power measured by stepped level test, passed the 5 minute test at rated power easily, though of course producing more heat than the class D Hypex. But, the heat is effectively dissipated and not concentrated in the heat isles.

A250W amplifier test review

 

[Mnyb]

Valid signal is music. Not 20 to 20 kHz constant sweep for 5 minutes. To be sure, you can design an amplifier that can do this for 100 hours or whatever You just have to be willing to pay for it the extra measure. And put up with larger enclosure, fan noise, etc. So it doesn't come for free. Currently the market doesn't demand this as folks are quite happy with what they are getting.

Yes that's what i meant would modern chip amps survive indefinitely with full scale input music signals that reaches the 2v or 4 v required for full output but not clipping ?
If you're a bit cheap on the thermal design as integrator it might not work ? Even if the chip goes into protection and saves the amp it's not the desired outcome, it should keep playing music .

I also realize that it would be ridiculously time consuming to actually try to test these things yourself Infion or Hypex have done thier job . It's possible that the integrator has done a poor thermal design thou ?

I *think* everything is ok. So as i gambled on active speakers that's probably has these kind of amps . And these amps seems to work in the field for most users .

 

[Chr1]

I agree with Old_School_Brad. It should really just be a case of lowering the power specifications. The makers of small desktop amps shouldn't need to completely redesign for thermal management. They should however have to specify the realistic continuous power output that they are designed for. I agree that it's good for amplifiers to go into protection rather than fry, but consumers should be given a clearer indication of an amps capability. Seems to me that the problem is not so much with the makers of the chips, though it would appear that there are some issues there too.

 

[restorer-john]

Those of us who care to test amplifiers properly will continue to do so. We will report the ones that cannot meet their advertised ratings, either here, somewhere else, to the regulatory bodies that may be interested, or the punters that are buyers. Amir can do whatever he likes. I don't care.

ASR is just a fringe enthusiasts' website. It has zero presence outside our little obsessive bubble of an echo chamber. We are fringe, let's be honest. Arguments that seem important here, are nothing in the real world. Nobody actually cares until the chit they bought fails inexplicably.

That said, regulatory bodies do care. They care about their funding, their work and their regulations, and they love to be informed about organizations/people/companies trying to skirt, deceive, and blatantly lie their way around the regulations they created. They take it as a personal impost. I know this. If a website was actively recommending products that couldn't meet their advertised specifications, especially when the reviewer was more than capable of determining that, there would potentially be a a big problem for that poor reviewer and his LLC.

 

[JustJones]

This amplifier came out after the updated FTC regulations became effective. It doesn't conform to the regulation. Report them.

Those of us who care to test amplifiers properly will continue to do so. We will report the ones that cannot meet their advertised ratings, either here, somewhere else, to the regulatory bodies that may be interested, or the punters that are buyers. Amir can do whatever he likes. I don't care.

ASR is just a fringe enthusiasts' website. It has zero presence outside our little obsessive bubble of an echo chamber. We are fringe, let's be honest. Arguments that seem important here, are nothing in the real world. Nobody actually cares until the chit they bought fails inexplicably.

That said, regulatory bodies do care. They care about their funding, their work and their regulations, and they love to be informed about organizations/people/companies trying to skirt, deceive, and blatantly lie their way around the regulations they created. They take it as a personal impost. I know this. If a website was actively recommending products that couldn't meet their advertised specifications, especially when the reviewer was more than capable of determining that, there would potentially be a a big problem for that poor reviewer and his LLC.

P282 Stereo Power Amplifier – March Audio

The P262 is a Purifi based stereo power amplifier. This simply elegant product is milled from solid aluminium with a tough anodized treatment. Exceptional fit, finish and build quality. Class leading technical and subjective audio performance.

marchaudio.com