How Many Watts am I using?

[Bucknekked]

This is a long standing question that drives a ton of "should I buy a new amp?" type of questions. I think a lot of people have no real idea (me included) what their current rig is outputting for power and whether or not more power would make any audible difference. So here goes for the amplifier gurus:

QUESTION: How can I measure the wattage my AVR is putting out right now as I play music? What's the technique? Is there a way to guesstimate it?

Observation: Before you tell me to go to Gene's article on how he measures amplifier power (FOUND HERE), I went there and couldn't make heads or tails of it. I need SIMPLE. Not necessarily technically accurate, but SIMPLE.

Here's what I have done so far:
I took my trusty Simpson Voltmeter (yeah, its analog), set it to VAC on the lowest scale
Hooked it up across my speaker input leads (high impedance meter, should not affect the load)
Measured the VAC at two listening volumes : 80db and 90db at my listening position
At 80db the meter barely registered : maybe .25 VAC (one quarter volt ac)
At 90db the meter was dancing with the music at 1.3 to 1.5vac
My listening position is 60 inches on the nose from the speaker
Speakers are Klipsch RB25's and they are rated at 8 ohms

How can I turn that simple measure in to a guesstimate about output power on my amp?
The only formula I have seen gives me too small a number. Rather than die a painful arithmetic death, I know someone on AH knows how to do this in their sleep. And no, I don't have an oscilloscope these days. I sold it years ago.

 

[highfigh]

This is a long standing question that drives a ton of "should I buy a new amp?" type of questions. I think a lot of people have no real idea (me included) what their current rig is outputting for power and whether or not more power would make any audible difference. So here goes for the amplifier gurus:

QUESTION: How can I measure the wattage my AVR is putting out right now as I play music? What's the technique? Is there a way to guesstimate it?

You
Observation: Before you tell me to go to Gene's article on how he measures amplifier power (FOUND HERE), I went there and couldn't make heads or tails of it. I need SIMPLE. Not necessarily technically accurate, but SIMPLE.

Here's what I have done so far:
I took my trusty Simpson Voltmeter (yeah, its analog), set it to VAC on the lowest scale
Hooked it up across my speaker input leads (high impedance meter, should not affect the load)
Measured the VAC at two listening volumes : 80db and 90db at my listening position
At 80db the meter barely registered : maybe .25 VAC (one quarter volt ac)
At 90db the meter was dancing with the music at 1.3 to 1.5vac
My listening position is 60 inches on the nose from the speaker
Speakers are Klipsch RB25's and they are rated at 8 ohms

How can I turn that simple measure in to a guesstimate about output power on my amp?
The only formula I have seen gives me too small a number. Rather than die a painful arithmetic death, I know someone on AH knows how to do this in their sleep. And no, I don't have an oscilloscope these days. I sold it years ago.

A VOM or DMM isn't going to be accurate at most frequencies and when you have multiple frequencies at the same time, it just kind of takes its ball and goes home.

If you want to know exactly, you'll need to use an oscilloscope- it doesn't care if you send one frequency, full bandwidth, pink or white noise, etc.

Personally, I wouldn't worry about needing more power unless the volume control indicator is close to 12:00 on a regular basis and you're listening to the radio (I use that as an example because it's often the loudest source). Best practice, IMO, is to make sure your amplifier's maximum output is at least double what you use, so you'll have decent headroom.

BTW, and this is really for anyone who may try it- I hope you know not to use a Simpson or any other analog VOM to read resistance when it could be connected to an amplifier's output or to any other sensitive input jacks or devices. For those who don't know why, set the VOM to read Resistance X1 and connect it to a 12V light bulb.

 

[lovinthehd]

Even considering Klipsch's heavy hand on the sensitivity number of 94dB, and your slightly further distance than 1m, you're probably using in the neighborhood of 2 watts?

 

[Bucknekked]

A VOM or DMM isn't going to be accurate at most frequenciesBTW, and this is really for anyone who may try it- I hope you know not to use a Simpson or any other analog VOM to read resistance when it could be connected to an amplifier's output or to any other sensitive input jacks or devices. For those who don't know why, set the VOM to read Resistance X1 and connect it to a 12V light bulb.

highfigh:
good comments. I'm pretty much figuring I won't be too accurate because I'm probably using the audio equivalent of a shovel to drive a nail (not the best tool for the job). But its the ballpark I'm interested in.
I have the itch for more power, but the sneakin' suspicion that I'm not even close to using 10% of what I have. Plus, I'd really like to know what I am actually consuming.

One thing I learned today after owning this meter for over 30 years, it has a db scale on it. I have looked at this meter for 3 decades and never seen the green scale in db. Because I'm interested in audio measurements now, its sorta leaped out at me today. from the manual:

I had no idea this meter would measure db. I have no idea how to do that ! I am actually stumped.
Just when you think you know something, up jumps reality and I'm back to a newb again.

 

[Bucknekked]

Even considering Klipsch's heavy hand on the sensitivity number of 94dB, and your slightly further distance than 1m, you're probably using in the neighborhood of 2 watts?

lovinthehd
I would like to see the arithmetic, but, trying to use some of the formulas I've seen I keep coming up in the 1 to 2 watts range, or less. That seems WAY too low a number to me, but, like you suggested, it also fits.
Using the make believe Klipsch number of 94db (not too far from my 90db measure) and their make believe number of 1 watt at 1 meter, that would be in the ballpark with my numbers. I'm 5 feet away, and I don't know exactly my wattage expenditure, but I can peg my db on top of theirs in my test. Even if they are off by several factors (like 3x or 4x or 5x) in what they state, I'm still under 10 watts at 90some-odd-db.

I would love to know if this is even a reasonable thing to do (measuring wattage), and if this isn't it, what is. If an oscilloscope is the answer, I wonder who I might know that would have one I could borrow for a few hours? O'scopes have sorta fallen out of favor.

Thanks for the quick opinion on the wattage. I am still going to hold out that one of our AH genius's will whip out a slide rule and give me a calculation, or tell me how I can get one made. Yes, that was a shameless name drop about slide rules. I have an entire collection of them. I remember when they were high tech.

 

[lovinthehd]

lovinthehd
I would like to see the arithmetic, but, trying to use some of the formulas I've seen I keep coming up in the 1 to 2 watts range, or less. That seems WAY too low a number to me, but, like you suggested, it also fits.
Using the make believe Klipsch number of 94db (not too far from my 90db measure) and their make believe number of 1 watt at 1 meter, that would be in the ballpark with my numbers. I'm 5 feet away, and I don't know exactly my wattage expenditure, but I can peg my db on top of theirs in my test. Even if they are off by several factors (like 3x or 4x or 5x) in what they state, I'm still under 10 watts at 90some-odd-db.

I would love to know if this is even a reasonable thing to do (measuring wattage), and if this isn't it, what is. If an oscilloscope is the answer, I wonder who I might know that would have one I could borrow for a few hours? O'scopes have sorta fallen out of favor.

Thanks for the quick opinion on the wattage. I am still going to hold out that one of our AH genius's will whip out a slide rule and give me a calculation, or tell me how I can get one made. Yes, that was a shameless name drop about slide rules. I have an entire collection of them. I remember when they were high tech.

LOL I have a slide rule still myself, don't think I've used it in 40 years or so, though

Have you played with an spl calculator before? http://myhometheater.homestead.com/splcalculator.html

 

[Verdinut]

Hi guys!
I have a similar question to Bucknekked's question on amplifier power being used.

I listen mostly to classical music including opera on CDs, SACDs. DVD-As as well as Blu-ray discs. Dynamic range can get over 100 dB, as with a SACD, a DVD-Audio disc or a DTS-Master soundtrack on a Blu-ray disc.

My question is: When I play a disc, the Marantz SR5010 playback volume display is usually set at an average of 65-70. If I didn't set any dynamic compression on the unit, would someone have an idea as to what percentage of the rated power is being output by the left and right front channels for a stereo recording? I assume that this receiver could put out a maximum approximate continuous 160 watts per channel at 4 ohms (2 channels driven).

I intend to purchase three QSC RMX850a power amps to drive my 3 front 4 ohm three-way speaker enclosures. Please note that neither the high frequency transducers or the subwoofers (being used as woofers in the same boxes) have an impedance going lower than 4 ohms, which makes them easier to drive. At present, I am using the SR5010 to bi-amp the front L & R speakers and an external SONY TA-3200F for the center channel.

I was just curious about the average wattage used when the display indicates 70 for instance. Does it mean that the receiver is using as much as 70% on the average of its continuous power supply capacity?

 

[PENG]

LOL I have a slide rule still myself, don't think I've used it in 40 years or so, though

Have you played with an spl calculator before? http://myhometheater.homestead.com/splcalculator.html

That is one of the best calculator found online but if you are trying to find the target spl you have to do it by trial and error. You also have to know how to adapt it for different load impedance. Still, it is accurate (I checked it against my own spreadsheet) and is easy enough to use. I like the fact that it allows for factors such as room reinforcement and number of speakers.

It would be nice if can post this in such a way that we don't have to do it every other week.

 

[PENG]

lovinthehd
I would like to see the arithmetic, but, trying to use some of the formulas I've seen I keep coming up in the 1 to 2 watts range, or less. That seems WAY too low a number to me, but, like you suggested, it also fits.
Using the make believe Klipsch number of 94db (not too far from my 90db measure) and their make believe number of 1 watt at 1 meter, that would be in the ballpark with my numbers. I'm 5 feet away, and I don't know exactly my wattage expenditure, but I can peg my db on top of theirs in my test. Even if they are off by several factors (like 3x or 4x or 5x) in what they state, I'm still under 10 watts at 90some-odd-db.

You are doing a great job and getting results that are more accurate than you may think. I don't use slide rules and scientific calculators any more because I find it much easier to use spreadsheets such as Microsoft Excel's.

In this case, the online calculator lovinthehd posted can do it easily, though I did unnecessary verify it with the spreadsheet just for fun.

Estimated wattage using the online calculator based on the given data

- 5 feet away
- Speaker sensitivity 94 dB/2.83V/1m and nominal impedance 8 ohm is equivalent to 94 dB/1W/1m

The online calculator shows you need 0.5W to yield 87.3 dB, or 1W to yield 90.3 dB.

That's based on one speaker placed away from the wall. If the speaker is 2 to 4 feet from a wall, you may get 90.3 dB with 0.5W.

For 2 speakers, you should get 90.3 dB with only 0.25W.

Estimated wattage by measurements and power formula:

Measured voltage at 90 dB: 1.5V

Power = 1.5X1.5/8 = 0.195 VA (not watt because the average phase angle is unknown).

So the calculated value using your experimental result is lower, but if the average impedance of the speaker during test was actually closer to 6 ohms, and you did have both L/R speakers on, then the values would have been much closer. Also, as you know already, you can't expect the measured voltage to be accurate because your multi meter is not designed to measure non sinusoidal waveform accurately, though in terms of RMS value it should get you in the ball park.

Note that so far we are dealing with RMS voltage and AVERAGE power, if pure test tones were use, it will be 1.414 times higher for the voltage peak and 2 times higher for the power peak.

If pink noise or random music is used, say the kind of music waveform that has crest factor of 15 dB, then the 0.25W average will have peaks of 8W, still low. For the same test, if you sit another 5 ft back, that 8W peak will become 32W.

 

[lovinthehd]

That is one of the best calculator found online but if you are trying to find the target spl you have to do it by trial and error. You also have to know how to adapt it for different load impedance. Still, it is accurate (I checked it against my own spreadsheet) and is easy enough to use. I like the fact that it allows for factors such as room reinforcement and number of speakers.

It would be nice if can post this in such a way that we don't have to do it every other week.

You are doing a great job and getting results that are more accurate than you may think. I don't use slide rules and scientific calculators any more because I find it much easier to use spreadsheets such as Microsoft Excel's.

In this case, the online calculator lovinthehd posted can do it easily, though I did unnecessary verify it with the spreadsheet just for fun.

Estimated wattage using the online calculator based on the given data

- 5 feet away
- Speaker sensitivity 94 dB/2.83V/1m and nominal impedance 8 ohm is equivalent to 94 dB/1W/1m

The online calculator shows you need 0.5W to yield 87.3 dB, or 1W to yield 90.3 dB.

That's based on one speaker placed away from the wall. If the speaker is 2 to 4 feet from a wall, you may get 90.3 dB with 0.5W.

For 2 speakers, you should get 90.3 dB with only 0.25W.

Estimated wattage by measurements and power formula:

Measured voltage at 90 dB: 1.5V

Power = 1.5X1.5/8 = 0.195 VA (not watt because the average phase angle is unknown).

So the calculated value using your experimental result is lower, but if the average impedance of the speaker during test was actually closer to 6 ohms, and you did have both L/R speakers on, then the values would have been much closer.

Note that so far we are dealing with RMS voltage and AVERAGE power, if pure test tones were use, it will be 1.414 times higher for the voltage peak and 2 times higher for the power peak.

If pink noise or random music is used, say the kind of music waveform that has crest factor of 15 dB, then the 0.25W average will have peaks of 8W, still low. For the same test, if you sit another 5 ft back, that 8W peak will become 32W.

I'd also discount the 94dB a few dB....this is a good read http://www.avsforum.com/forum/89-speakers/2128818-can-klipsch-speaker-actually-meet-s-specs-no-can-t.html

 

[PENG]

I'd also discount the 94dB a few dB....this is a good read http://www.avsforum.com/forum/89-speakers/2128818-can-klipsch-speaker-actually-meet-s-specs-no-can-t.html

Yes, I would too, maybe 3 dB or so. This whole exercise is about "ball park" estimate so I didn't bother mentioning it as I had in other posts about Klipsch speakers. We have no idea what kind of room reinforcement, and the test signal he's using. So the overall result could have been skewed more than +/1 3 dB for sure. The fact remains, 90 dB average is very loud, even 84 dB if he sits another 5 ft back, is much louder than what I normally listen to. So with those speakers, he probably won't need any more than 100W average/200W peak capable amp.

 

[Bucknekked]

You are doing a great job and getting results that are more accurate than you may think. I don't use slide rules and scientific calculators any more because I find it much easier to use spreadsheets such as Microsoft Excel's.

So the calculated value using your experimental result is lower, but if the average impedance of the speaker during test was actually closer to 6 ohms, and you did have both L/R speakers on, then the values would have been much closer.

I want to express my thanks to Peng and lovinthehd for helping me learn how to accomplish this basic task. Its one thing to simply answer a question, its another (more helpful) to teach how a task can be performed.
The only thing they could have done better was use a slide rule.

On the topic of two speakers or one, I did have both speakers running when I did my db measurements. It never occurred to me that just running one might have been better. Peng took that in to consideration in his explanation. I believe you said that if I had 2 running the measurements fell even closer to the mark as to what you would expect.

I tried the online calculator and found it to be pretty cool. I bookmarked it for future reference. I'll be doing this again for sure when the Salks arrive. Its the imminent (sometime in April) arrival of the Salks that kicked me in to gear to check this out. The Salks, and my natural curiosity to know how much power I am actually using.

Thanks again. I will write a separate post about what the answers mean to me, in context. I don't want to write a book in one post.

 

[Bucknekked]

What did I learn from the experiment to see how many watts I am actually using when I sit and listen to music? I typically listen at the 80db level for the average evening. When I want some zz Top and I want it loud, the 90db setting comes in to play. When I want it louder than that, its rare, I can get there too. Remember, my room is only a 10x10x10 sealed cube. I am not playing Carnegie hall here.

At 80db, I am probably not burning more than 1w out of my AVR. Night in and night out, I use the 80db listening level. 1 watt. At 90db, a pretty loud setting, I'm still not burning much more than maybe 2 to 3 watts (peaks will be more etc etc).

What that tells me is that a desire for more watts (a seemingly endless want) is simply not warranted or supported by the facts. I am a rock n roll guy, learning to appreciate jazz and the blues. Needing "more watts" has been a thought in my head for as along as I can remember. A cherished notion.

I think, all in all, the numbers tell me that if I buy more watts, I will just be heating the room with most of them. I can't approach amplifier clipping because I will get driven out of the room first. It would seem, if I am understanding the results, more power in the amp/avr is like a muscle car in Manhattan.

If this is true, its must be frustrating to answer all the "I need a new amp for more power" questions that the AH forum sees week in and week out. I see a lot of folks start their posts with "I need more power to drive my new xyz speakers, what should I buy?" or "I want these new xyz speakers and I know my AVR isn't as good as a real amplifier".

I still want more power. Its still in my head. But now, I think I will hold on to the cash for something else.
Maybe, I can get some more slide rules for my collection. Perhaps there's one out there for sound pressure or wattage.

 

[PENG]

On the topic of two speakers or one, I did have both speakers running when I did my db measurements. It never occurred to me that just running one might have been better. Peng took that in to consideration in his explanation. I believe you said that if I had 2 running the measurements fell even closer to the mark as to what you would expect.

I tried the online calculator and found it to be pretty cool. I bookmarked it for future reference. I'll be doing this again for sure when the Salks arrive. Its the imminent (sometime in April) arrival of the Salks that kicked me in to gear to check this out. The Salks, and my natural curiosity to know how much power I am actually using.

Yes, and you can see from that online calculator that for 2 speakers you gain 3 dB and if you select yes to select "Near a wall" you gain another 3 dB. Each time you gain 3 dB of spl, it means your power requirement is halved. So a gain of 6 dB total from having 2 speakers with a wall 2-4 ft behind them will drop your power requirement from 1W to 0.25W.

Next time you do it, I suggest you use test tones that are sine waves, and you can download them. 1kHz would be good. You can also use pink noise, still better than random music that won't give you steady reading on your spl meter and the Simpson meter. The Simpson meter will be quite accurate if you use sine wave test tones and the results you get should be very closed to what you get from the online calculator.

If you can get a higher end Fluke digital meter, you will also get more meaningful results even with a music signal as long as the music spl is reasonably steady, not like classical music. That is because a good Fluke meter measures true RMS and can deal with higher crest factors. The Simpson is really mainly for sine waves.

 

[Verdinut]

You are doing a great job and getting results that are more accurate than you may think. I don't use slide rules and scientific calculators any more because I find it much easier to use spreadsheets such as Microsoft Excel's.

In this case, the online calculator lovinthehd posted can do it easily, though I did unnecessary verify it with the spreadsheet just for fun.

Estimated wattage using the online calculator based on the given data

- 5 feet away
- Speaker sensitivity 94 dB/2.83V/1m and nominal impedance 8 ohm is equivalent to 94 dB/1W/1m

The online calculator shows you need 0.5W to yield 87.3 dB, or 1W to yield 90.3 dB.

That's based on one speaker placed away from the wall. If the speaker is 2 to 4 feet from a wall, you may get 90.3 dB with 0.5W.

For 2 speakers, you should get 90.3 dB with only 0.25W.

Estimated wattage by measurements and power formula:

Measured voltage at 90 dB: 1.5V

Power = 1.5X1.5/8 = 0.195 VA (not watt because the average phase angle is unknown).

So the calculated value using your experimental result is lower, but if the average impedance of the speaker during test was actually closer to 6 ohms, and you did have both L/R speakers on, then the values would have been much closer. Also, as you know already, you can't expect the measured voltage to be accurate because your multi meter is not designed to measure non sinusoidal waveform accurately, though in terms of RMS value it should get you in the ball park.

Note that so far we are dealing with RMS voltage and AVERAGE power, if pure test tones were use, it will be 1.414 times higher for the voltage peak and 2 times higher for the power peak.

If pink noise or random music is used, say the kind of music waveform that has crest factor of 15 dB, then the 0.25W average will have peaks of 8W, still low. For the same test, if you sit another 5 ft back, that 8W peak will become 32W.

Hi Peng,

Isn't a transducer's sensitivity at 1 watt/1m usually 2-3 dB less than that quoted for an input of 2.83v/1m?

 

[highfigh]

highfigh:
good comments. I'm pretty much figuring I won't be too accurate because I'm probably using the audio equivalent of a shovel to drive a nail (not the best tool for the job). But its the ballpark I'm interested in.
I have the itch for more power, but the sneakin' suspicion that I'm not even close to using 10% of what I have. Plus, I'd really like to know what I am actually consuming.

One thing I learned today after owning this meter for over 30 years, it has a db scale on it. I have looked at this meter for 3 decades and never seen the green scale in db. Because I'm interested in audio measurements now, its sorta leaped out at me today. from the manual:



I had no idea this meter would measure db. I have no idea how to do that ! I am actually stumped.
Just when you think you know something, up jumps reality and I'm back to a newb again.

Most meters that test for voltage are designed for line voltage, so their usable range doesn't need to be more than about 30Hz-200Hz, if even that wide.

The dB scale is probably just an easy visual guide- if the meter has a needle that can be set to one spot as a reference, you would measure the voltage and use a bit of math to determine the amount of change. Hard to say without knowing the model.

 

[lovinthehd]

Hi Peng,

Isn't a transducer's sensitivity at 1 watt/1m usually 2-3 dB less than that quoted for an input of 2.83v/1m?

1 watt is 2.83V at 8 ohm....

 

[Bucknekked]

Yes, and you can see from that online calculator that for 2 speakers you gain 3 dB and if you select yes to select "Near a wall" you gain another 3 dB. Each time you gain 3 dB of spl, it means your power requirement is halved. So a gain of 6 dB total from having 2 speakers with a wall 2-4 ft behind them will drop your power requirement from 1W to 0.25W.

Next time you do it, I suggest you use test tones that are sine waves, and you can download them. 1kHz would be good. You can also use pink noise, still better than random music that won't give you steady reading on your spl meter and the Simpson meter. The Simpson meter will be quite accurate if you use sine wave test tones and the results you get should be very closed to what you get from the online calculator.

If you can get a higher end Fluke digital meter, you will also get more meaningful results even with a music signal as long as the music spl is reasonably steady, not like classical music. That is because a good Fluke meter measures true RMS and can deal with higher crest factors. The Simpson is really mainly for sine waves.

how about REW to generate the test tones? I could chose any tone at any SPL using REW. I can also use REW to select just one speaker, then the other, then both.

My son has some of the new Fluke meters. I could see what he has if that might make a difference.

I appreciate the learning assistance here. This is a fun weekend project. Maybe from my wifes point of view its a pointless waste of time, but, hey, she's not the audio nut. I am.

 

[highfigh]

What did I learn from the experiment to see how many watts I am actually using when I sit and listen to music? I typically listen at the 80db level for the average evening. When I want some zz Top and I want it loud, the 90db setting comes in to play. When I want it louder than that, its rare, I can get there too. Remember, my room is only a 10x10x10 sealed cube. I am not playing Carnegie hall here.

At 80db, I am probably not burning more than 1w out of my AVR. Night in and night out, I use the 80db listening level. 1 watt. At 90db, a pretty loud setting, I'm still not burning much more than maybe 2 to 3 watts (peaks will be more etc etc).

What that tells me is that a desire for more watts (a seemingly endless want) is simply not warranted or supported by the facts. I am a rock n roll guy, learning to appreciate jazz and the blues. Needing "more watts" has been a thought in my head for as along as I can remember. A cherished notion.

I think, all in all, the numbers tell me that if I buy more watts, I will just be heating the room with most of them. I can't approach amplifier clipping because I will get driven out of the room first. It would seem, if I am understanding the results, more power in the amp/avr is like a muscle car in Manhattan.

If this is true, its must be frustrating to answer all the "I need a new amp for more power" questions that the AH forum sees week in and week out. I see a lot of folks start their posts with "I need more power to drive my new xyz speakers, what should I buy?" or "I want these new xyz speakers and I know my AVR isn't as good as a real amplifier".

I still want more power. Its still in my head. But now, I think I will hold on to the cash for something else.
Maybe, I can get some more slide rules for my collection. Perhaps there's one out there for sound pressure or wattage.

If you can experiment with speaker placement, I strongly recommend it. Mark the present location and move each one in small increments to find their sweet spot. I did that a couple of years ago and it made a huge difference in the sound of my system, the apparent location of vocals and instruments, the frequency response and it actually sounds better at low SPL than it did at ANY level before I did this.

At the moment, the Pandora station I'm listening to is playing 'Girl From The North Country' from Joe Cocker's Mad Dogs & Englishmen and I would be very surprised if my amp is pounding it out at more than 1/2 Watt/channel but the details are far from lost. Another interesting thing is that the high and low frequencies don't lack anything- the current song is 'Sucker Row', by Mark Knopfler and the cymbals & bass guitar come through very nicely, as do the electric guitars (not a loud song), drums and vocals. The next song is 'The Lucky Man' by Allison Krauss & Union Station- everything is crystal clear.

 

[PENG]

Hi Peng,

Isn't a transducer's sensitivity at 1 watt/1m usually 2-3 dB less than that quoted for an input of 2.83v/1m?

Not if the load is 8 ohm, 3dB less if the load is 4 ohm.