I was watching a YouTube channel that I’ve begun to like and one video was about “umpth” watts. He did some math and talked about how meat and potato wattage output ratings tend to be much lower than in specs. I already kinda knew that but his numbers were quite a bit lower.
This morning while having tea and poking around Parts Express, I was reading q&a on this Wondom 100w mono board. One person said they were only able to measure ~25w rms. In another query a staff member explained that the 87w @ 1% thd wasn’t listed as rms and was only short bursts. He said ~40w was more realistic for an rms expectation. I totally should have known better and been more critical in my reading of specifications.
The question now is what kind of actual rms power might one expect from a class d board, assuming the power supply is adequate? Also, we’re talking about brands like wondom, sure, ice, Dayton, ect. I realize that a true 15w is actually pretty good and will get most speakers loud but I’m concerned about subwoofer power. In particular, I’m planning on two small ~.25cf sealed enclosures for two Dayton Epique 5.5’s.
Depending on enclosure volume winsld predicts around 150w to reach near xmax, which apparently the 5.5’s can do fairly well behaved. Eq cut here and there for room corrections and seems every bit of 200w might be needed from time to time. I’ve personally noticed that my Anarchy 554’s don’t seem to hit anywhere near xmax off what should be ~70 on low bass.
This Wondom 2x160 is on the Anarchy 5.5’s with 19a 24v mean well. Dayton KABD 2x50 is providing the dsp’d line out.
https://store.sure-electronics.com/product/660
I’d like to give this KABD 4x100 a go bridged for 2x200 but I’m now concerned that it won’t be giving an actual 200x2 rms.
https://www.parts-express.com/Dayton-Audio-KABD-4100-4-x-100W-Bluetooth-Amp-Board-with-DSP-325-434
If this whole much less rms power than advertised thing is correct then do I need some 600X1 boards to get closer to a real 200w? As I’ve been looking around I don’t see any rms ratings on boards…
This morning while having tea and poking around Parts Express, I was reading q&a on this Wondom 100w mono board. One person said they were only able to measure ~25w rms. In another query a staff member explained that the 87w @ 1% thd wasn’t listed as rms and was only short bursts. He said ~40w was more realistic for an rms expectation. I totally should have known better and been more critical in my reading of specifications.
The question now is what kind of actual rms power might one expect from a class d board, assuming the power supply is adequate? Also, we’re talking about brands like wondom, sure, ice, Dayton, ect. I realize that a true 15w is actually pretty good and will get most speakers loud but I’m concerned about subwoofer power. In particular, I’m planning on two small ~.25cf sealed enclosures for two Dayton Epique 5.5’s.
Depending on enclosure volume winsld predicts around 150w to reach near xmax, which apparently the 5.5’s can do fairly well behaved. Eq cut here and there for room corrections and seems every bit of 200w might be needed from time to time. I’ve personally noticed that my Anarchy 554’s don’t seem to hit anywhere near xmax off what should be ~70 on low bass.
This Wondom 2x160 is on the Anarchy 5.5’s with 19a 24v mean well. Dayton KABD 2x50 is providing the dsp’d line out.
https://store.sure-electronics.com/product/660
I’d like to give this KABD 4x100 a go bridged for 2x200 but I’m now concerned that it won’t be giving an actual 200x2 rms.
https://www.parts-express.com/Dayton-Audio-KABD-4100-4-x-100W-Bluetooth-Amp-Board-with-DSP-325-434
If this whole much less rms power than advertised thing is correct then do I need some 600X1 boards to get closer to a real 200w? As I’ve been looking around I don’t see any rms ratings on boards…
It can be quite shocking opening up some cheap power amps.
I bought a 100w+100w amp and the transformer was tiny.
Quite often the rating is for peak or short burst.
I bought a 100w+100w amp and the transformer was tiny.
Quite often the rating is for peak or short burst.
This is a pretty complex topic, but a good first question is always "How much power do I need?" (and I said "need", not "want"!).
Basically for a "realistic" power specification you want to look in the datasheets for figures measured at 1% THD or less and across the whole frequency range (instead of only at 1 kHz). You'll find that in the ICEpower spec sheets and some other brands as well. If your board doesn't show anything of use, it is typically also in the raw IC datasheets from TI and the like.
Basically for a "realistic" power specification you want to look in the datasheets for figures measured at 1% THD or less and across the whole frequency range (instead of only at 1 kHz). You'll find that in the ICEpower spec sheets and some other brands as well. If your board doesn't show anything of use, it is typically also in the raw IC datasheets from TI and the like.
Here’s the spec sheet for the wondom 160x2. Shows a basically flat response but only at 1w. The parameter sheet says 111w at 1% thd though. Might it really be a true 111w rms?
There are many ways to cheat on the power spec. Here are a few:
In my view the honest approach is to measure the THD+N vs output power and quote the RMS output power at the onset of clipping as the maximum output power. That's my approach. But some would argue that the output power should be specified at the point where the THD becomes audible. So now we're not just talking the onset of clipping, we're talking audible distortion. How much is audible can be debated. In controlled studies, values from 0.01% to 1% have been noted. So you'll often see amps specified at 1 % or 10 % THD+N. The reason for this is simple. As you can see in the plot below, which I measured on a power amp, there is a significant difference in output power for the onset of clipping, 0.1% THD+N (top of the graph), 1%, and 10% THD+N.
I call this a 200 W amplifier. Some may call it a 250 W amplifier as that's where the THD+N reaches 0.1 %. But in the commercial world you'll find many others who would label this a 300 W amp (1 % THD+N) or maybe even a 350 W amp (10 % THD+N).
The FTC got rid of the amplifier testing standard last year so now we're back to the Wild West. Honest manufacturers will give you the THD+N curve as well as the point where they've defined the maximum output power. You can then make an informed purchasing decision. If the manufacturer won't give you those data I'd just walk away. It's not like there's a shortage of manufacturers in the audio market.
Tom
- Pick the largest power number on the front page of the data sheet of the amp chip. This seems to be a common thing on eBay/Amazon/etc. An amp might deliver 400 W into 2 Ω, so it'll be sold as a "400 W amp". Never mind that it'll never deliver anywhere close to that with a more reasonable load (or with the power supply that it's sold with).
- Measure the maximum peak output voltage the amp can deliver in a short burst with a lab power supply, calculate the power from there and call it "peak music power".
- Same as #2 but use the RMS power instead. This one is starting to look like an honest approach.
In my view the honest approach is to measure the THD+N vs output power and quote the RMS output power at the onset of clipping as the maximum output power. That's my approach. But some would argue that the output power should be specified at the point where the THD becomes audible. So now we're not just talking the onset of clipping, we're talking audible distortion. How much is audible can be debated. In controlled studies, values from 0.01% to 1% have been noted. So you'll often see amps specified at 1 % or 10 % THD+N. The reason for this is simple. As you can see in the plot below, which I measured on a power amp, there is a significant difference in output power for the onset of clipping, 0.1% THD+N (top of the graph), 1%, and 10% THD+N.
I call this a 200 W amplifier. Some may call it a 250 W amplifier as that's where the THD+N reaches 0.1 %. But in the commercial world you'll find many others who would label this a 300 W amp (1 % THD+N) or maybe even a 350 W amp (10 % THD+N).
The FTC got rid of the amplifier testing standard last year so now we're back to the Wild West. Honest manufacturers will give you the THD+N curve as well as the point where they've defined the maximum output power. You can then make an informed purchasing decision. If the manufacturer won't give you those data I'd just walk away. It's not like there's a shortage of manufacturers in the audio market.
Tom
tomchr - I agree that's the best way, although you have to be a bit careful about those graphs as they sometimes have a flattish section before clipping, how exactly to you define the precise rise-point of clipping? Perhaps something like 50% above the lowest point, or 0.1%, whichever is lower (ie anything worse than 0.1% doesn't count as an amplifier 🙂 That 50% allows for some rippling in the THD at moderate powers.
So for your graph the 0.0003% point would count as the clipping point.
How you deal with manufacturing spread is another tricksy detail to consider. And the spread in mains voltages come to that...
So for your graph the 0.0003% point would count as the clipping point.
How you deal with manufacturing spread is another tricksy detail to consider. And the spread in mains voltages come to that...
I’ve tried to gather information on the audibility of distortion products amongst the various ranges over the last few years. What I’m getting at is that I personally only ever need as many watts as fingers on my left hand above the midbass. In those frequencies where we’re apparently quite sensitive to distortions we’re also running amplifiers in the pristine low power range.
Where rms power concerns me is below 60hz and particularly below 40hz, yes I said below 40hz. I enjoy various forms of synthesized bass, and many times said synth bass is a droning tone lasting longer than brief bursts. I also like to design with small sealed subwoofers, aka Dayton Epique 5.5” which apparently remains quite well behaved up to near 14mm xmax. I’ve heard it a handful of times that we can tolerate quite a bit of distortion in the lower bass region and 1% thd doesn’t worry me one bit below ~100 hz.
Even more respectable music artists like Pink Floyd can offer some deep bass filled passages. Movies especially can deliver headroom taxing bass effects. I guess what I’m asking really boils down to this, is the 111wpc that my Wondom boards state the same 111 watts that winsld models with?
Where rms power concerns me is below 60hz and particularly below 40hz, yes I said below 40hz. I enjoy various forms of synthesized bass, and many times said synth bass is a droning tone lasting longer than brief bursts. I also like to design with small sealed subwoofers, aka Dayton Epique 5.5” which apparently remains quite well behaved up to near 14mm xmax. I’ve heard it a handful of times that we can tolerate quite a bit of distortion in the lower bass region and 1% thd doesn’t worry me one bit below ~100 hz.
Even more respectable music artists like Pink Floyd can offer some deep bass filled passages. Movies especially can deliver headroom taxing bass effects. I guess what I’m asking really boils down to this, is the 111wpc that my Wondom boards state the same 111 watts that winsld models with?
If 40hz content is important to you, yes, you’ll certainly need reasonable power, but more importantly, you should consider real bass drivers, not small diameter drivers that have greater than average throw. Big drivers couple the air in a way that small drivers simply cannot. Also they will be more efficient, requiring smaller amplifiers.
I have a small living room where fitting even ~.5 cf cabinets would be a little big, the parts express .23cf knockdowns (11.5h 7.5w 8.5d are about the biggest I’m interested in using. Going with small-ish drivers in a .23cf bass reflex alignment has been kinda difficult as well. Seems almost every 5” class driver wants a half cube to get low. Considering the design constraints, I feel like throwing power at the problem is my best option here. And it’s just cool to me having a a set of bookshelves the size of a 12 pack shaking the windows.
I built this .5cf enclosure tuned to 27hz around a tang band w5 and mounted it under my tv above the bottom shelf. It seems fairly efficient and plays quite low but it interferes with the function of the shelves. Bare in mind that there’s usually not wires sprawling around everywhere…
My other option is to use the .04cf cabinets I’m building for the rear surround as front mains with a maybe dual .4cf subwoofer under each side. With close to half a cubic foot I could start getting useful output from a pr with ~75g. A higher q eq boost of 3db right at tuning seems to lift the low bass just enough to make the complexity worthwhile. But, when I really think about it as practicality as I can, it seems throwing power at a small high xmax subwoofer is the path of least resistance.
I have a 500w 24v open frame smps, boost converters and plenty of other electronic stuff to supply full electricity requirements for this endeavor. I’m just not sure if my wondom 160x2 is actually giving me 111wpc or if my Dayton kabd 4x100 @6ohm is giving me around 100wpc at 4ohm. The kabd says 50x4 at 1%thd at an 8 ohm load so I reckon that it “should” do 100x4 at 4ohm 1%thd.
I built this .5cf enclosure tuned to 27hz around a tang band w5 and mounted it under my tv above the bottom shelf. It seems fairly efficient and plays quite low but it interferes with the function of the shelves. Bare in mind that there’s usually not wires sprawling around everywhere…
My other option is to use the .04cf cabinets I’m building for the rear surround as front mains with a maybe dual .4cf subwoofer under each side. With close to half a cubic foot I could start getting useful output from a pr with ~75g. A higher q eq boost of 3db right at tuning seems to lift the low bass just enough to make the complexity worthwhile. But, when I really think about it as practicality as I can, it seems throwing power at a small high xmax subwoofer is the path of least resistance.
I have a 500w 24v open frame smps, boost converters and plenty of other electronic stuff to supply full electricity requirements for this endeavor. I’m just not sure if my wondom 160x2 is actually giving me 111wpc or if my Dayton kabd 4x100 @6ohm is giving me around 100wpc at 4ohm. The kabd says 50x4 at 1%thd at an 8 ohm load so I reckon that it “should” do 100x4 at 4ohm 1%thd.