• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Do watt meters accurately measure the power drawn by class D amps?

Destination: Moon

Senior Member
Joined
Feb 28, 2021
Messages
431
Likes
280
Location
Western USA
I plugged my little SMSL DA9 into a relatively cheap watt meter that then plugs into the wall. It's accurate for other devices like Li battery chargers etc. Reads to 0.1 decimals but don't know the accuracy.

I'm asking because at even higher than normal listening levels it shows the amp only drawing about 3.5 watts or less! The wattage varies as the music volume rises and falls from moment to moment but I'm s surprised at how little power it's using. Speakers are Elac DBR 62
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
11,951
Likes
13,531
Location
Riverview FL
It measures an average, which infers not an instantaneous value.

A couple of watts of power going to the speakers is likely to be rather loud,

The power required to go louder rises quickly, though.

---

Repeat the test with a sine tone...

Multiply the meter reading by 1.414 to approximate the peak value.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
11,951
Likes
13,531
Location
Riverview FL
OP
D

Destination: Moon

Senior Member
Joined
Feb 28, 2021
Messages
431
Likes
280
Location
Western USA
It's so little power and it's easily 70 plus db in a large room at 14 feet. I'm wondering, do I really need more than 50w per channel (I've been considering a new amp) if all I use is 4 or 5 +2???
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
11,951
Likes
13,531
Location
Riverview FL
It's so little power and it's easily 70 plus db in a large room at 14 feet. I'm wondering, do I really need more than 50w per channel if all I use is 4 or 5 +2???

Your speakers:
Sensitivity: 86db @ 2.83v/1m

Add up to 12dB SPL for four speakers if all playing the same signal at the same time.

Estimate of power consumed by one speaker at different levels:

1628124351246.png


70+ dB SPL is almost nothing for power requirement.
 
OP
D

Destination: Moon

Senior Member
Joined
Feb 28, 2021
Messages
431
Likes
280
Location
Western USA
Your speakers:
Sensitivity: 86db @ 2.83v/1m

Add up to 12dB SPL for four speakers if all playing the same signal at the same time.

Estimate of power consumed by one speaker at different levels:

View attachment 145538

70+ dB SPL is almost nothing for power requirement.

That's awesome! What's the SPL mean if there isn't a distance specified?

Edit:. Opps see that it's 1 meter.... How does that translate at 4 meters?
So that 70DB at 4M would be what at 1M?
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
11,951
Likes
13,531
Location
Riverview FL
That's awesome! What's the SPL mean if there isn't a distance specified?

The sensitivity specification for speakers is usually referenced to a meter or so.

Yours:
Sensitivity: 86db @ 2.83v/1m

86dB SPL is measured at 1 meter (1m) when the standard test voltage is applied.

2.83Vrms makes 1 watt with an 8 ohm speaker, 2 watts with a 4 ohm speaker, and something in-between with a 6 ohm speaker (yours). All "more or less" as speaker impedance varies with frequency.

---

Without walls, the sound level would drop about 6 dB with each doubling of distance.

In a room, the drop is much less as the sound that would have passed you by and gone unheard is reflected at you from the room surfaces, adding to or cancelling some of the direct sound, depending on the phase of the reflected waves relative to the direct wave..
 
Last edited:

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
6,416
Likes
12,248
Location
Monument, CO
That's awesome! What's the SPL mean if there isn't a distance specified?

Edit:. Opps see that it's 1 meter.... How does that translate at 4 meters?
So that 70DB at 4M would be what at 1M?

Take Ray's chart and move 6 dB to the right on the SPL for each doubling of distance from 1 m. For 2 m, add 6 dB to your listening SPL, so look at 76 dB instead of 70 dB and see what the power is. For 4 m, add 6 dB to get to 2 m, and another 6 dB to get to 4 m, so look at the power at 70 + 12 = 82 dB SPL, etc. It is still a very low number.

Class D amps are very efficient, though less so at very low power levels, but say it is 80% efficient so multiply by 1/0.8 = 1.25 to get the average wall power. The peaks will be significantly higher, but your meter is likely just measuring average power, and the power supply inside your amps supplies the peaks so the wall does not usually "see" the very fast peaks.

You can also look at something like this: _Peak SPL Calculator

Ray, would you mind posting (attaching) your spreadsheet? Seems like a cool thing to have...
 

DVDdoug

Major Contributor
Joined
May 27, 2021
Messages
1,133
Likes
1,389
It's just not that helpful to measure the power into the amp... You want to know the power out, into the speakers. :( It's unfortunate that so few amps have peak power meters. Clipping indicators are more common but still rare.

I plugged my little SMSL DA9 into a relatively cheap watt meter that then plugs into the wall.
Of course all of the energy going to the speakers has to come from the power outlet. And of course, the amplifier is not 100% efficient so if you are "pulling" 5W (average) there will be less than 5W (average) going to the speakers. Class D amplifiers can be very efficient but I think they are the most efficient at high-outputs. (They will consume some power with silence, so at that point they are 0% efficient.)

Like Ray says, you're really interested in the peak power which can be 10 times the average or more or less depending on the program/music you're listening to. (It's the peaks that clip.) And even if the Watt meter could read the peak power out of the wall, the power supply in the amplifier stores energy so it won't "track" the amplifier's peak output power.

I don't recommend testing your amp with high-power constant test-tones (which would have the same peak and average power)... You could blow your speakers. The power ratings for speakers are based occasional peaks with regular program material, and with most of that energy going to the woofer.
 
Last edited:
OP
D

Destination: Moon

Senior Member
Joined
Feb 28, 2021
Messages
431
Likes
280
Location
Western USA
So if power output of an amp is expressed as RMS does that mean most of the transient spikes in power are encompassed in the rating?
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
6,416
Likes
12,248
Location
Monument, CO
So if power output of an amp is expressed as RMS does that mean most of the transient spikes in power are encompassed in the rating?

Aarrgghhhh... There's a lot to unpack in that simple statement.
  1. Watts RMS is not a thing. Average power, in watts, is the product of RMS voltage (volts) and RMS current (amperes). Manufacturers all too often use Wrms but that is a nonsensical unit. RMS x RMS does not equal RMS. /rant/ Not your fault!
  2. The RMS, root-mean-square, value of a voltage or current is equivalent to the DC value. In other words, if you put 10 Vrms across a resistor, and 10 Vdc, it will get just as hot. In that sense, a waveform that has very large peaks (transients) that occur only now and then, and a waveform with no transients that is steady-state, can have the same RMS value. So in that sense the transients are "encompassed" in the RMS value, but the RMS value does not tell you what the peak values are -- you need to know more about the waveform.
  3. You cannot conflate input and output power; again, you need more information.
    • You need to know the amplifier's efficiency, which tells you how much input power is needed for given output. That can range from 25% or so for a single-ended class A amplifier, to 50% for a push-pull class A amp, to about 60% or so for a class AB amp, to 90% or more for a class D amp. Your audiophile push-pull class A amplifier wastes 50% of the power coming into it as heat; the class D amplifier only wastes 10%. But the efficiency varies with power output so even there you need more information.
    • The input power is roughly the output power divided by the efficiency, so a class A amp putting out 100 W will draw about 200 W (100 W/0.5) from the wall, wasting 100 W as heat. A class D amp will draw about 100/0.9 ~ 111 W, wasting 11 W as heat.
  4. For music, the ratio of peak to average level is about 17 dB (some more, some much less). That is a factor of 50 in power, so if you only need 1 W for your average listening level, you need about 50 W to handle the biggest peaks without clipping. For movies the number can be even higher; I have seen 20 dB (100x) to 30 dB (1000x!) cited informally. The good news is that the average power for most people is far, far less than what they think. But the peaks are probably bigger, relatively, than what they think... And note the big peaks tend to be loud cymbal crashes, drum strikes, blaring trumpets (I like those! ;) ), or explosions and gunshots that probably don't really sound much different if clipped so you probably do not really need 1000 W to watch your big action movies.
  5. Note that, as Ray said, sound in free space falls off at 6 dB for every doubling in distance, but in a room the sound waves are constrained and reflect back to the listener so the fall-off may be slower.
    • In power, 6 dB is a factor of 4, and 3 dB is a factor of 2, so a few dB can make up for a lot of power.
    • In the midrange frequencies, a 1 dB change is just noticeable to most of us. If you bump up the volume "just a little" that is probably about 3 dB (taking twice the power). It takes 10 dB, and 10x the power, to sound twice as loud.
    • Similarly, a speaker rated 90 dB/W/m will require half the power to sound as loud as a speaker rated 87 dB/W/m, so if you like it loud buy speakers with high sensitivity ratings.
HTH - Don

More info on units: https://www.audiosciencereview.com/forum/index.php?threads/units-symbols-and-terms-oh-my.1923/
More info on RMS: https://en.wikipedia.org/wiki/Root_mean_square
 
Last edited:
OP
D

Destination: Moon

Senior Member
Joined
Feb 28, 2021
Messages
431
Likes
280
Location
Western USA
Don, thanks so much for that explanation. I had most of it from everyone's response but you summed it all up really well. I wasn't ignoring efficiency. But I figured If I'm measuring 4 watts going into the amp, then the actual power going out to the speakers was even less and, /2..,... I've been saying 4 watts but the actual draw from the wall was really 2.5 watts just after a loud section of music at my normal listening level. I'm still surprised how little (ave) power it takes to fill the room with sound! And for reference, my old Sony TA555 draws about 80 watts in similar circumstances

And I share your enthusiasm for horns! The Crusaders are my all time favorite
 
Top Bottom