hypex power ratings

[March Audio]

Manufacturers want to loosen test conditions for their product to look better.

No manufacturers, such as myself want the test to be relevant and meaningful.

 

[March Audio]

There is no universal and defined MUSIC, to be used as a standard. And not only 1kHz is requested, but sine waves from 20Hz to 20kHz. As John has already stated.
.

There is no music that comes even vaguely close to requiring the same power output at 20kHz as it does at 1kHz.

Please go and find the worst, most taxing music example you can and we can examine it.

 

[davidr3032]

@davidr3032

ATI aren't wrong. They are rating their product in accordance with the US rules.

This discussion has been had many times on ASR (and other places). Basically, the Hypex power ratings are not compliant with US FTC amplifier regulations, but, as they are only selling components (modules), not complete amplifiers, they can get away with it. It is up to the assembler/builder/manufacturer of the completed product to advertise their product in accordance with those regulations if sold in the US.

The EU regulations are much looser, and, with Hypex in particular, their specifications are cherry picked to look good. Not uncommon practice, many manufacturers do that. They are still an excellent product for many people, don't get me wrong.

The 20Hz-20kHz FTC rating means the minimum continuous average power at a specified distortion or lower can be achieved at any frequency across the specified bandwidth on a continuous basis. It doesn't mean all frequencies at once.

We all know that typical music has lower power demands at high frequencies, but that is irrelevant. It's just an excuse for covering up the inherent problems and un-levelling the playing field. In the case of the Hypex modules, it is more about the dissipation and destruction of the output filter capacitors at high power/high frequencies. They simply cannot deliver their full rated power at very high frequencies (20kHz) for more than a few seconds.

Once classD amplifiers improve with even faster switching devices, filter related dissipation issues will become less of an issue and we'll begin to see full power, full bandwidth ratings. I look forward to that.

In the meantime, you are doing well to be one of the few who read through the specs and into the nitty gritty. Keep up the independent thinking.

Thanks Restorer John. You were mentioned earlier in this thread and I was hoping you would chime in at some point. What you say makes a lot of sense and over the past few days I have come to the same conclusions. Though I'm sure you have a lot more knowledge on this than I do. I only starting looking into hypex based amps a few days ago. So since thay have been out for a few years, I'm sure i have gone over the same ground as others. Doubtless I won't be the only one asking these questions as class d amplifiers become more and more popular.

I think it would be better if the manufacturers of hypex based amplifiers and other class d amps gave more realistic power ratings like the way ATI do. Then they could always supplemented those with further specs explaining what class d amps are capable of. The specs on these amps are impressive with low distortion and high damping factors etc. They have a lot going for them. So plenty of selling points without the misleading power ratings

If its true that in terms of power ratings, they don't perform well with common testing methods this can be explained in the manufacturer blurp

I think hypex themselves can morally get away with the way they summarise the power ratings. As they are not manufacturers of amplifiers and give lots of specs and graphs with more detailed info. So I dont think its such a big deal. However i think it is a big deal when the manufacturers of the hypex based amps use these summerized figures as the power ratings for their amps. Thats where people can be mislead

Like you say, they aren't operating on a level playing field. I have done the maths and their goal mouth is exactly 42.5 degrees up a slope. And I have the graphs to prove it (though I do use a different method for measuring angles than the standard way)

 

[davidr3032]

Yes they can output full power at 20kHz, but as I keep pointing out they will never need to to. No amplifier needs to.

OK thanks thats very interesting. As most amplifiers can't operate full power at 20kHz. Usually there's a drop of power at such high frequencies. Why do you think ATI have rated it 200w 20hz-20khz while you rate it at 330w?

 

[March Audio]

I think it would be better if the manufacturers of hypex based amplifiers and other class d amps gave more realistic power ratings like the way ATI do. Then they could always supplemented those with further specs explaining what class d amps are capable of. The specs on these amps are impressive with low distortion and high damping factors etc. They have a lot going for them. So plenty of selling points without the misleading power ratings

They are not misleading, its been explained and demonstrated that they do what they say on the tin. They can sustain the quoted power 20 to 20kHz.

I showed a video earlier a NC252 based amp doing exactly that conforming with the 5 minute FTC test.

 

[March Audio]

OK thanks thats very interesting. As most amplifiers can't operate full power at 20kHz. Usually there's a drop of power at such high frequencies. Why do you think ATI have rated it 200w 20hz-20khz while you rate it at 330w?

Usually? Is there?

You need to speak to ATI to answer that. Perhaps they dont use an adequate power supply, I dont know.

Can I ask why you keep focussing on this particular point (20kHz) when its clearly not important or required?

 

[davidr3032]

You need to speak to ATI to answer that. pehaps they dont use an adequate power supply, I dont know.

Yeah quite a difference from your specs. 200w compared to 330w. Are you suggesting that ATI have made a catastrophic mistake with their design?

 

[davidr3032]

Usually? Is there?

You need to speak to ATI to answer that. Perhaps they dont use an adequate power supply, I dont know.

Can I ask why you keep focussing on this particular point (20kHz) when its clearly not important or required?

Because a power rating of an amp 20hz-20khz should include the frequency of 20khz

 

[March Audio]

Because a power rating of an amp 20hz-20khz should include the frequency of 20khz

No, thats just the requirement of the FTC. It has been clearly explained why it is of no relevance at higher frequencies, but just to make the point again, the Hypex amps have no issues with doing so.

BTW I just looked at the ATI website and they use their own linear power supplies, not the Hypex SMPS which the Hypex specs are generated with.

This is Amirs test of a Nord NC500 based amp. If you look at the graph it clearly matches the Hypex data.

Oh BTW there is a very good reason for only using 6kHz in a THD test. The 2nd and 3rd harmonics of 6kHz fall at 12 and 18kHz, ie within the audible range. If you test at 20kHz the 2nd and 3rd harmonics are at 40 and 60kHz which are well above the auditory, or speaker reproduction range.

Again its about having meaningful and relevant tests.

 

[March Audio]

Yeah quite a difference from your specs. 200w compared to 330w. Are you suggesting that ATI have made a catastrophic mistake with their design?

As I said, I dont know why they cant achieve the full Hypex performance spec, but I am suspicious about the use of a linear PSU. It would need to be very large to provide the 1kW needed for a stereo NC500 design.

 

[davidr3032]

No, thats just the requirement of the FTC. It has been clearly explained why it is of no relevance at higher frequencies, but just to make the point again, the Hypex amps have no issues with doing so.

BTW I just looked at the ATI website and they use their own linear power supplies, not the Hypex SMPS which the Hypex specs are generated with.

View attachment 106619

Its not just the requirement of the FTC its also the requirement of the end of the sentence "the power rating of the amp is 330w 20hz-20khz"

So you think ATI are using power supplies that are reducing the amps potential power rating by a huge amount? Thats quite a controversial conclusion

 

[March Audio]

Its not just the requirement of the FTC its also the requirement of the end of the sentence "the power rating of the amp is 330w 20hz-20khz"

So you think ATI are using power supplies that are reducing the amps potential power rating by a huge amount? Thats quite a controversial conclusion

Oh BTW there is a very good reason for only using 6kHz in a THD test. The 2nd and 3rd harmonics of 6kHz fall at 12 and 18kHz, ie within the audible range.

If you test at 20kHz the 2nd and 3rd harmonics are at 40 and 60kHz which are well above the auditory, or speaker reproduction range, so again another reason its irrelevant.

Again its about having meaningful and relevant tests.

Not controversial at all. Its a logical and reasonable conclusion.

Please look above at Amirs independant test of a Nord NC500 based stereo amp that uses a Hypex SMPS and has no issue matching the Hypex spec.

 

[pma]

Once classD amplifiers improve with even faster switching devices, filter related dissipation issues will become less of an issue and we'll begin to see full power, full bandwidth ratings. I look forward to that.

John, biggest stress to LC filter capacitors is because they must be filtering the switching frequency of 450-600kHz (rectangular shape) with full scale swing, reduced by L of the filter. Just calculate Ic = C x dv/dt. That's why the filter capacitors must be properly chosen.

 

[davidr3032]

No, thats just the requirement of the FTC. It has been clearly explained why it is of no relevance at higher frequencies, but just to make the point again, the Hypex amps have no issues with doing so.

BTW I just looked at the ATI website and they use their own linear power supplies, not the Hypex SMPS which the Hypex specs are generated with.

View attachment 106619

This is Amirs test of a Nord NC500 based amp. If you look at the graph it clearly matches the Hypex data.

View attachment 106620

View attachment 106621

Oh BTW there is a very good reason for only using 6kHz in a THD test. The 2nd and 3rd harmonics of 6kHz fall at 12 and 18kHz, ie within the audible range. If you test at 20kHz the 2nd and 3rd harmonics are at 40 and 60kHz which are well above the auditory, or speaker reproduction range.

Again its about having meaningful and relevant tests.

I love Amirs testing reports and its great to have such a talented knowledgeable independent reviewer to check things against the manufacturers specs. However he was not testing the Nord nc500 amp to find its power rating here. Nor does he speculate what the power rating is. It would be incorrect to conclude from these graphs that the power rating of the amp is 700w into 4ohm. He is not saying it is and is not trying to comply to the FTC standard. But like I say thats not what he is trying to achieve here.

He states the power before clipping is 445w into 4ohm. I dont think it would be right to assign the amp a higher power rating than this. And if he made the amp himself I dont think he would

However manufacturers of these amps do assign a higher power rating of 445w. So the rating they give of 700w is well after clipping

If manufacturers don't comply to a standard then the playing field is not level. And power ratings for the average consumer become confusing and they can be easily mislead

 

[March Audio]

Sorry but of course he was testing the power rating.

His conclusion of 445 watts was at an arbitrary thd + n level of 0.003%. The level he chooses varies on each and every test. It should be consistent.

This very low level is way before clipping. If you go down that road why not choose 0.0001%?

The FTC test is *not* a technical standard. It is for compliance with advertising rules in the USA *only*. I have explained some of its technical flaws. Please take this on board.

Amirs data is almost identical to the Hypex data. The Hypex data is not misleading as you have stated. It is accurate.

With the input of multiple posters in this thread hopefully you have now been given a better understanding of how to read the data.

You can choose to believe whatever you wish. I hope you find an amp which suits your needs.

 

[davidr3032]

Sorry but of course he was testing the power rating.

His conclusion of 445 watts was at an arbitrary thd + n level of 0.003%. The level he chooses varies on each and every test. It should be consistent.

This is way before clipping. If you go down that road why not choose 0.0001%?

The FTC test is *not* a technical standard. It is for advertising rules in the USA *only*. I have explained why it is technically flawed. Please take this on board.

Amirs data is almost identical to the Hypex data. The Hypex data is not misleading as you have stated. It is accurate.

With the input of multiple posters in this thread hopefully you have now been given a better understanding of how to read the data.

You can choose to believe whatever you wish.

Yes well exactly. Therefore the correct power rating is not 700w into 4ohm. Thats one thing we can say for sure. And its wrong for manufacturers to quote it as so. If it is 445w then surely manufacturers who quote 700w are misleading people. By a significant amount

But Amir isn't specifically trying to find the amps overall power rating here. He's merely simply stating the amp power before clipping is 445w. Which is great info to know

 

[RayDunzl]

How is it "unlevelling the playing field" when say a 100 watt (at 1kHz) amplifier will never ever be required to output 100 watts at 20kHz?

The amp will be required to wiggle the output voltage at 20kHz near or at peak voltage when the 20kHz signal rides lower frequencies near the power limits.

Just an observation, I have no further argument other than that.

 

[March Audio]

Yes well exactly. Therefore the correct power rating is not 700w into 4ohm. Thats one thing we can say for sure. And its wrong for manufacturers to quote it as so. If it is 445w then surely manufacturers who quote 700w are misleading people. By a significant amount

But Amir isn't specifically trying to find the amps overall power rating here. He's merely simply stating the amp power before clipping is 445w. Which is great info to know

Yes well exactly what?

It is perfectly correct to say the power output is 700 watts into 4 ohms at 1% thd + n. Its all on the graph and easy to read.

No Amir states the power is 445 Watts at 0.003% thd + n. Yes this is a long way before any reasonable definition of clipping. He could have equally stated that it was 200 watts at 0.002% which is also below clipping.

I will leave you to it, choose whatever power you think is apropriate.

 

[davidr3032]

After some research on linear power supplies I can accept that the ATI hypex amplifiers that use them would result in their amps being less powerful than other nc500 based amps. This makes sense to me. So it would seem that although the nord nc500dm I was asking about is less powerful than the 400w power rating its given. The figure of 250w can't be used to speculate as to the power rating of the Nord nc500 amps for example. Two different power supplies. Two different power outputs

So ive learned something at least
As to the power rating of the nord nc500dm I know its not 400w. But I also know its very powerful. 330w seems high to me as according to Amirs measurements it would be clipping. So I'm just going to tell myself its somewhere near 300 watts or so. I've bought a nord nc500dm now so I will just try it and see how it sounds. And not turn things up too much in case I fly across the room like Marty Mcfly

 

[March Audio]

It will be required to wiggle the output voltage at 20kHz near or at peak voltage when the 20kHz signal rides lower frequencies near the power limits.

Just an observation, I have no further argument other than that.

True but as the previous examples showed this might only be fractions of a watt.