• 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!

Cheap class D amps. Andrew Robinsons take.

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,512
Location
Seattle Area
I watched the video. If he had just read one of my reviews he would know that these have load dependency so their frequency response does depend on the load. Alas, he picked a Klipsch speaker that has exaggerated highs and very high sensitivity in that region to boot. So no wonder he hears a hiss.

He says the Arylic A50/B50 are better. Well, I tested the A50 and it is one of the worst amplifiers I have tested: https://www.audiosciencereview.com/...s/arylic-a50-review-wireless-amplifier.19944/

index.php


index.php


Reading between the lines, it seems that they had sent him amps and he has been recommending them for a while. Hence my issue that I post about just now. You never know what you are hearing is a fact or advertising driven combined with poor subjective analysis.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,512
Location
Seattle Area
I will add though that he is right in one respect: if you are seriously into this hobby you need to step up for load independent amplifiers and those with more power. For that reason, I would never use one outside of workshop, etc. But if all you have is $80, these are fantastic choices to have compared to junk that used to be sold just a couple of years ago.
 

TonyJZX

Major Contributor
Joined
Aug 20, 2021
Messages
1,893
Likes
1,821
but why would you need such powerful amps if your selection of speakers is of high sensitivity and fairly low load... i mean as long as you dont own stuff that dips down to 2-3 ohm

i do get that once you're fully into the autism spectrum where you own $10,000 speakers then yeah, why are you using $200 amps?

the amp I'm listening to now is a $99 clearance H/K affair that truly is decent but would I trust it at 95dB plus? no
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,512
Location
Seattle Area
but why would you need such powerful amps if your selection of speakers is of high sensitivity and fairly low load...
What high sensitivity speakers? Just about every speaker I test has a sensitivity of 85 to 86 dB which is quite low. You need hundreds of watts to push them to proper reference level (peaks).
 

JeremyFife

Addicted to Fun and Learning
Joined
Jan 8, 2022
Messages
707
Likes
794
Location
Scotland
What high sensitivity speakers? Just about every speaker I test has a sensitivity of 85 to 86 dB which is quite low. You need hundreds of watts to push them to proper reference level (peaks).
I'm slightly confused by this statement "need 100's of watts").
My understanding, from this site, is that at normal listening levels, normal speakers (around 85dB, 8-4 ohm range, listening at 2-3 m) you don't need anything like that much power.

Does "reference level peak" mean something theoretical - a level to measure at but not for actual listening? Or is this sort of headroom necessary for 'big', dynamic music like some jazz and full orchestra classical?

Should I be planning an amp upgrade?

Curious
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,228
Likes
17,810
Location
Netherlands
I'm slightly confused by this statement "need 100's of watts").
My understanding, from this site, is that at normal listening levels, normal speakers (around 85dB, 8-4 ohm range, listening at 2-3 m) you don't need anything like that much power.

Does "reference level peak" mean something theoretical - a level to measure at but not for actual listening? Or is this sort of headroom necessary for 'big', dynamic music like some jazz and full orchestra classical?

Should I be planning an amp upgrade?

Curious
An 85 dB speaker needs 450W to reach 105 dB (“reference level”) at 3M distance.

I would not envy the poor little 2-way bookshelf handling that kind of power, even with a 15 dB crest factor.
 

JeremyFife

Addicted to Fun and Learning
Joined
Jan 8, 2022
Messages
707
Likes
794
Location
Scotland
An 85 dB speaker needs 450W to reach 105 dB (“reference level”) at 3M distance.

I would not envy the poor little 2-way bookshelf handling that kind of power, even with a 15 dB crest factor.
I guess that's my question... why do I want reference level output? If I listen loud, for me, at 85dB (6dB headroom, Crown calculator) then I need something like 30W ... unless I've really missed something, which is possible!
Is 'reference level ' something to aspire to for the sake of it or are there practical applications?
 

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
I guess that's my question... why do I want reference level output? If I listen loud, for me, at 85dB (6dB headroom, Crown calculator) then I need something like 30W ... unless I've really missed something, which is possible!
Is 'reference level ' something to aspire to for the sake of it or are there practical applications?
85dB SPL is average (rms) and when assuming a conservative 20dB crest factor (eg some classical music, rock pop is more like 10 to 15) one would get to 105dB peak and require the 450w. However if you don’t listen at 85 average, I certainly don’t, and only consider 15dB crest, the power requirements get exponentially less of course.
 

JeremyFife

Addicted to Fun and Learning
Joined
Jan 8, 2022
Messages
707
Likes
794
Location
Scotland
85dB SPL is average (rms) and when assuming a conservative 20dB crest factor (eg some classical music, rock pop is more like 10 to 15) one would get to 105dB peak and require the 450w. However if you don’t listen at 85 average, I certainly don’t, and only consider 15dB crest, the power requirements get exponentially less of course.
Thanks. Fair comment - I don't listen that loud either (more like 70dB SPL). The only time I'd run 450W into my speakers would be if I wanted a new pair:)
I'm trying to understand @amirm 's comment about reference level and why I might want that
 

notsodeadlizard

Senior Member
Joined
Feb 18, 2023
Messages
403
Likes
362
One thing is not entirely clear to me: how Andrew Robinsons keeps from laughing when he "reviews" an amplifier with a declared output power of 300 watts and a standard 150 "on paper" watt power supply.
I would have laughed.
And nothing more is needed at all.
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
One thing is not entirely clear to me: how Andrew Robinsons keeps from laughing when he "reviews" an amplifier with a declared output power of 300 watts and a standard 150 "on paper" watt power supply.
I would have laughed.
And nothing more is needed at all.

He knows that the governments and there physicist lie at us. We dont need more power plants. We just have to buy some cheap power amps. And we can quadruple the power of the power stations easy. ;)

Buuuut we should not support that crazy power claims of some manufactures.
Yes give them a headless until they get it.
 
Last edited:

Talisman

Addicted to Fun and Learning
Forum Donor
Joined
Mar 27, 2022
Messages
897
Likes
2,547
Location
Milano Italy
One thing is not entirely clear to me: how Andrew Robinsons keeps from laughing when he "reviews" an amplifier with a declared output power of 300 watts and a standard 150 "on paper" watt power supply.
I would have laughed.
And nothing more is needed at all.
Some potency claims are really absurd by some manufacturers.
However we must stop calculating the peak power with the calculator and the data plate of the power supply.
Many capacitors are able to accumulate enough energy for short bursts that are much higher than the continuous power that can be supplied by the power supply, and since music has a very large peak component, they are able to manage 25/50% more than the continuous power supply data for transients that are not it is as absurd as it seems to be for many
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
Some potency claims are really absurd by some manufacturers.
However we must stop calculating the peak power with the calculator and the data plate of the power supply.
Many capacitors are able to accumulate enough energy for short bursts that are much higher than the continuous power that can be supplied by the power supply, and since music has a very large peak component, they are able to manage 25/50% more than the continuous power supply data for transients that are not it is as absurd as it seems to be for many

Then give me two powers, clear is rms out can never be more than rms in at least if you not inventet the perpetum mobile. How ever you measure peaks, you have to descripe that peaks. What makes something like 200W complety useless couse you had to add, for 1nS or 1ms or 20mS. If not then we are at something completly useless like 200W pmpo. All that burst powers are BS listen to some techno with a long deep sine bass. Than all that peak power is as usefull as a nailfungus.
 
Last edited:

Bjorn

Major Contributor
Audio Company
Forum Donor
Joined
Feb 22, 2017
Messages
1,286
Likes
2,562
Location
Norway
I'm slightly confused by this statement "need 100's of watts").
My understanding, from this site, is that at normal listening levels, normal speakers (around 85dB, 8-4 ohm range, listening at 2-3 m) you don't need anything like that much power.

Does "reference level peak" mean something theoretical - a level to measure at but not for actual listening? Or is this sort of headroom necessary for 'big', dynamic music like some jazz and full orchestra classical?

Should I be planning an amp upgrade?

Curious
Keep in mind that the rated sensitivity is an average mainly for the midrange area.

Especially in the lows, the sensitivity drops significantly. So unless one is using a subwoofer one needs hundreds of watts to reach decent levels.
 

Talisman

Addicted to Fun and Learning
Forum Donor
Joined
Mar 27, 2022
Messages
897
Likes
2,547
Location
Milano Italy
Then give me two powers, clear is rms out can never be more than rms in at least if you not inventet the perpetum mobile. How ever you measure peaks, you have to descripe that peaks. What makes something like 200W complety useless couse you had to add, for 1nS or 1ms or 20mS.
I can agree with you, but that's a different story, more about the total absence of universal standards to describe the power of an amplifier on a commercial level.
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
I can agree with you, but that's a different story, more about the total absence of universal standards to describe the power of an amplifier on a commercial level.

Its called rms power. Its just if you like to sell a amp with a cheap psu for a high price you can use what ever peak power you like.
 

Talisman

Addicted to Fun and Learning
Forum Donor
Joined
Mar 27, 2022
Messages
897
Likes
2,547
Location
Milano Italy
Its called rms power. Its just if you like to sell a amp with a cheap psu for a high price you can use what ever peak power you like.
ok, at what frequency? Because class D modules such as Hypex for example cannot even remotely express their rated power at high frequencies, even if in the real world it is not a problem, given that high frequencies contain very little energy, the problem remains that that value "20/20,000" is not valid
The problem is more complex than a simple formula, and as always requires reading graphs and understanding the measurements well. It will never be a unique number that tells us the whole story.
So let's go back to the problem, there is no single and universally adopted standard to commercially describe the power of an amplifier, so everyone writes what suits them best
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
ok, at what frequency? Because class D modules such as Hypex for example cannot even remotely express their rated power at high frequencies, even if in the real world it is not a problem, given that high frequencies contain very little energy, the problem remains that that value "20/20,000" is not valid
The problem is more complex than a simple formula, and as always requires reading graphs and understanding the measurements well. It will never be a unique number that tells us the whole story.
So let's go back to the problem, there is no single and universally adopted standard to commercially describe the power of an amplifier, so everyone writes what suits them best

You are right rms power should not be given at 1khz. It should be measured from 20-20khz and then the lowest number should be given. Yes its a complex proplem thats why you better not open that cans of worms called peak power.

If you like to relay on something than stay in peaks under your rms power. All the rest is guessing and good luck. Not realy usefull.
 
Last edited:
Top Bottom