How many WATTS of power do we really need for amplifier / receiver?

KayMan2

Hi everyone,
.
I have been shopping around for a receiver to drive my massive-sized RTi12 / CSi5 / Surrounds home-theater. After auditioning multiple receivers in **this thread** , I was perplexed by the variety in wattage. Denon topped the chart at 140 watts per channel, while Arcam was the bottom at 75 wpc: why such a huge difference in comparable products? If these were cars, it would be as if a 458 horse-power Ferrari engine costs the same as a 250 hp Nissan engine.

It didn't make any sense, so I set out to calculate exactly how many watts do we need. Here is a simplified calculation that you can follow along and adjust for YOUR own setup:

.
1) Setting a Target:

We need to select a goal for how loud do we want the listening room to be.

12-gauge shotgun?
Rock concert front-row?
Jet-engine territory?

Well, here are the accepted standards for recommended permissible exposure time for continuous time weighted average noise, according to CDC:
**Acceptable Noise Levels for Safety**

Basically you can watch a movie at 90 decibels, or an audio album at 95 dB without risking long-term hearing loss. I mostly watch movies, so I picked 90 dB as a target loudness level. You are free to pick a higher number... the calculations below will still apply.

.
2) Adjusting Target for Headroom:

To make sure the audio is not distorted when you push the Amplifier to the limit, we need to allow for some headroom.

3 dB is an accepted headroom recommendation at speaker-level by European Broadcasting Union. Here is **a graph** .

So my new target loudness adjusted for headroom is 90 dB + 3 dB = 93 dB.
Make a note of this number, and now we will fork out in a different direction in the calculation.

.
3) Speaker Sensitivity

Polk, Bowers and Wilkins sensitivity = ~90 dB
Klipsch = 100+ dB

"Sensitivity" is how loud a speaker will be, when you apply ONE watt of power and sit ONE meter away from the speaker.

.
4) Adjusting Sensitivity for Distance

Every time listening distance doubles, volume level drops by ~6 dB. Message me if you want a more complicated explanation of this.

So now calculate "Distance adjusted sensitivity" of your speakers: in my case, I am ~4 meters away from the speakers (14 feet), so in my set up the volume will drop be 12 dB.
The distance-adjusted speaker sensitivity will be 90 dB - 12 dB = 78 dB.

.
5) Increasing the volume

Now every time the wattage doubles, loudness goes up 3. So here is a table for Polk speakers:

1 watt per channel of RMS power = 78 dB (This is the baseline number we calculated in step 4)
2 watts = 81 dB
4 watts = 84 dB
...
...
32 watts = 90 dB
64 watts = 93 dB
128 watts = 96 dB


So we are reaching our target loudness number we calculated in step (2) at just 64 watts! This is after factoring in headroom and room size, so this is the maximum power I will ever need for my set-up.

.
Two special notes:

1. The biggest factor in loudness levels is room size. When distance doubles, power is reduced by a factor of four! If I had a much larger room to fill, I would personally switch out my Polk speakers with more sensitive Klipsch, instead of buying large mono-block amps. 10 dB more sensitivity for these speakers means that they need 8-times less power to drive!!
2. If your speakers are positioned within 4 feet of a wall, add +3 dB to the final number in step (4). This is because walls reflect the sound waves, increasing the loudness level in the room.

Hope this helps :-)

K

BlueFox

How much money do you really need?

heiney9

Wow, you can cut and paste from a textbook :rolleyes:

KayMan2

BlueFox wrote: »

How much money do you really need?

Haha, I also subscribe to the "more is better" philosophy when it comes to money. But this calculation is intended to help out folks like me who are on a budget. Picking a target saves money in Amp costs, so you don't end up buying more than you need!

thsmith

The Denon 3808 forsale here is fleamarket is a bargain.

steveinaz

1. Room size
2. Desired levels
3. Speaker efficiency
4. Speaker nominal Impeadance

decal

If these were cars, it would be as if a 458 horse-power Ferrari engine costs the same as a 250 hp Nissan engine.

Not a very good analogy.

obieone

To paraphrase an old redneck "Sound is a question of money, how fast do you want to go?"

kuntasensei

But... this amp goes to 11...

madmax

KayMan2 wrote: »

The biggest factor in loudness levels is room size. When distance doubles, power is reduced by a factor of four! If I had a much larger room to fill, I would personally switch out my Polk speakers with more sensitive Klipsch, instead of buying large mono-block amps. 10 dB more sensitivity for these speakers means that they need 8-times less power to drive!!
K

Funny thinig about this one. I've tried several speakers in both my old small livingroom and my now huge livingroom. I've noticed the much larger room (about 6 times bigger) makes each set of speakers sound much bigger and full sounding than the smaller room, and with less wattage. Both rooms have about the same amount of livelyness. There are probably other factors at play but I have not figured out what they are.

KayMan2

Found a tool that automates the calculation!!

http://www.crownaudio.com/apps_htm/designtools/elect-pwr-req.htm


Edit: And here is a second tool, which factors in multiple speakers and wall reflection energy:

http://myhometheater.homestead.com/splcalculator.html

Hope this helps other newbies like me!

NewHTguy

Kayman,

I think your example might help those who are mostly concerned with not doing damage to their speakers. However, sound quality is what drives my purchases and more power from an amp that has relatively low distortion will usually provide better SQ. I say this as someone who doesn't believe everything I read on a message board. Rather when I first connected my Parasound to my Rti10s, there was an improvement over my Onk 875, which happens to be a fairly capable receiver. Not the sort of improvement you hear when you are in another room, or when the windows are open. But the improvement was there. Was it worth it? For me the answer was definitely yes.

You say that if you had a larger room you would switch to Klipsch. Frankly I couldn't listen to Klipsch in any size room because they are too bright for me. You seem to think that decibels are all that matter. If that is the case, you could have saved some money by buying lower quality speakers.

Joe08867

200 watts is all you need.

KayMan2

NewHTguy wrote: »

Kayman,

I think your example might help those who are mostly concerned with not doing damage to their speakers. However, sound quality is what drives my purchases and more power from an amp that has relatively low distortion will usually provide better SQ. I say this as someone who doesn't believe everything I read on a message board. Rather when I first connected my Parasound to my Rti10s, there was an improvement over my Onk 875, which happens to be a fairly capable receiver. Not the sort of improvement you hear when you are in another room, or when the windows are open. But the improvement was there. Was it worth it? For me the answer was definitely yes.

You say that if you had a larger room you would switch to Klipsch. Frankly I couldn't listen to Klipsch in any size room because they are too bright for me. You seem to think that decibels are all that matter. If that is the case, you could have saved some money by buying lower quality speakers.

Couldn't agree with you more - loudness does not correlate with tonality, detail, and other subjective sound measures. There is no instrument as good as one's own ears.

I posted my calculations because I was not sure if a 75-watt receiver will be able to drive RTi12s. This number crunching was just a way to verify this.

Drenis

1 watt is all you safely need.

heiney9

KayMan2 wrote: »

Found a tool that automates the calculation!!

http://www.crownaudio.com/apps_htm/designtools/elect-pwr-req.htm


Edit: And here is a second tool, which factors in multiple speakers and wall reflection energy:

http://myhometheater.homestead.com/splcalculator.html

Hope this helps other newbies like me!

KayMan2 wrote: »

Couldn't agree with you more - loudness does not correlate with tonality, detail, and other subjective sound measures. There is no instrument as good as one's own ears.

I posted my calculations because I was not sure if a 75-watt receiver will be able to drive RTi12s. This number crunching was just a way to verify this.

The numbers and specs really mean nothing and are useless since there are many more variables to consider which aren't accounted for in you calculations. As for the theory, it's solid.....but translation into a real world setting, with a real live rig, with real a music source, it means very little.

H9

KayMan2

heiney9 wrote: »

The numbers and specs really mean nothing and are useless since there are many more variables to consider which aren't accounted for in you calculations. As for the theory, it's solid.....but translation into a real world setting, with a real live rig, with real a music source, it means very little.

H9

Sure - this post was meant to be a watts-to-loudness calculation only, not a quality assessment of amps. To give an analogy, I can probably sing as loud as Justin Bieber :eek:, but who sounds better is a subjective thing!

That is why I am showroom hopping for past few days and listening to different receivers (captured in a separate thread).

pitdogg2

KayMan2 wrote: »

I posted my calculations because I was not sure if a 75-watt receiver will be able to drive RTi12s. This number crunching was just a way to verify this.

Another thing to consider is that posted specs at most are a dream of some manufactures. Denon's 140 watts may only be 50wpc in 5 ch. during lots of sound activity. where Arcam may well be much higher current(amps) and a true 75 watts x 5. NAD list some of their AVR's as 65w.p.c. but in 2 channel they are 165w.p.c. catchy the drifty;)

heiney9

Watts -to- loudness means nothing based on rudimentary calculations. Again, they don't take into account all the differening (real) environments. Nothing subjective at all about my earlier post.

Your formula relys on all values of the equation to be constant in order to produce the outcome. Nothing about sound is constant in the real world. Certainly a very loose guide but hardly concrete when sitting down to listen the exact same rig in different environments. Do you listen to a constant test tone? I know I don't, so the values in the formula's don't correlate to anything in reality. Of course, unless you listen to test tones.

H9

heiney9

I do agree that watts does equal loudness as it can be shown mathematically as watts increase so does the loudness. But it's like saying horsepower equals speed, without knowing any of the other constants involved.

H9

cnh

Welcome to Club Polk!

What everyone said above, +1!

However, if you "really" want everybody to hear your set up (past, present and future). You have to have,

1.21 Jigawatts per channel! That'll do ya!

cnh

F1nut

This thread should be titled, "How not to shop for gear."

nooshinjohn

I can't remember who was responsible for the study, but I do remember that it took some 2700 watts to accurately reproduce the "snip" sound of a pair of scissors.

kawizx9r

cnh wrote: »

Welcome to Club Polk!

What everyone said above, +1!

However, if you "really" want everybody to hear your set up (past, present and future). You have to have,

1.21 Jigawatts per channel! That'll do ya!

cnh

I'm still waiting for you to send me that flux capacitor!

mhardy6647

I need three and a half, per channel.

If you have speakers that are ca. 104 dB/watt @ 1 meter; you probably do, too.

mrbigbluelight

You know, you really oughta get rid of all that old fashioned junk; throw it away, kharma, or maybe even .... kharma ? Just thinking out loud.

Lasareath wrote: »

That's why my amp puts out 96Amps peak to peak current.

Well, if you want to cheat the specs, sure, go with P-P. Me & my Onkyo will stick with RMS. Sounds better.