How many watts am I really using?

[clintonb]

I read some posting about measuring how many watts an amp is putting out. They said if you use a pure sine wave and use a resistor instead of a speaker, you could simply measure the voltage across the resistor and then calculate the power using the formula power = voltage * voltage / resistance.

My 50-watt amp volume was still set at the same location it was at my last gig. I plugged in my guitar and played my A-string to check the volume. Then I plugged in my signal generator and set the pitch and volume so it matched my guitar. Since the guitar string has a bunch of overtones, I turned down the tone knob so it sounded more pure like the sine wave. I then replaced the 8ohm speaker with a 20-watt 8-ohm resistor and measured about 2 volts across the resistor. That comes out to about 1/2 a watt! So out of 50 watts I'm only using a half watt?! I wonder if I measured this correctly.

Then I started thinking about speaker sensitivity ratings. My Warehouse Guitar Speakers Veteran 30 produces 100dB for 1 watt at 1 meter (I think these measurements are for 1000 Hz signal). 100dB is not considered a safe volume level. So it appears it doesn't really take that many watts to be too loud at 1 meter with that speaker. I have no idea how quickly the sound pressure level decreases with distance.

Since we mic our amps on stage, I wonder how many watts I really need for gigging? I've done gigs with one of my amps in it's triode mode which supposedly produces around 6 or 7 watts and it seems to keep up just fine. I wonder if I could gig with one of those really small amps like the Zvex Nano or Rock Block or Blackheart Giant Ant, etc.

Are the wattage ratings for guitar amps based on the loudest it can get or is it based on how loud it gets before it distorts?

I have no idea what sound pressure levels I play at. But I saw this article that says you should dial in your tones while playing at 85dB: Ever wonder why your amp sounds so good at home yet not-so-good at a gig? - How To Get Dialed In - Epinions.com

Things I'm curious about:
- How loud do others play at gigs?
- What's the lowest power amp do others use at gigs?
- At what point does a guitar amp "open up"? Is it purely at a certain point in volume (85dB perhaps?).

[corworld]

I'm not sure if you used a signal generator or your guitar but for testing purposes, you should use a 1k sine simply to be able to compare your findings. To find the wattage of the amp you need to measure the clean wattage, so it is best to have a scope to watch the sine wave.

The speaker and cabinet make a big difference in the perceived volume in the room. A db meter would help you find your gig volume and be able to recreate that.

Another thing to try would be to hook up your speaker and measure the watts with it plugged in.

Most musicians don't need really big amps unless they play in big rooms. 50 watts is fine for all but the biggest rooms.

[clintonb]

I'm not sure if you used a signal generator or your guitar but for testing purposes, you should use a 1k sine simply to be able to compare your findings.

I was using a signal generator when I was measuring the wattage.

To find the wattage of the amp you need to measure the clean wattage, so it is best to have a scope to watch the sine wave.

So, in essence you are saying that guitar amp wattage ratings are based on a clean sound.

The speaker and cabinet make a big difference in the perceived volume in the room. A db meter would help you find your gig volume and be able to recreate that.

I'll have to buy a meter. I'm curious how loud I really play.

Another thing to try would be to hook up your speaker and measure the watts with it plugged in.

Actually, I kept reading that you shouldn't measure it with the speaker connected. I thought they said it affected the readings. But then again, it might be because they were measuring maximum wattage.

[MWJB]

To find out how many watts you are really using simply connect a true RMS AC voltmeter accross the speaker terminals & play. The reading will jump about some but try and identify an average. Square this figure, then divide by speaker ohmage and you get the wattage that you actually use, clean or not & not the RMS (clean wattage) from the manufacturer.

Most folks/manufacturers use a resistive load to measure W RMS, partly because it saves you going deaf. As you noted yourself, when you set the signal to what you thought was an appropriate level you only saw 0.5W.

You shot yourself in the foot with your "test" by resetting the knobs from what they were at when you last used the amp.

[Steve Conner]

It all depends on the gig. I've had coffee shop gigs where I couldn't turn a 3 watt amp up. And I've had rehearsals with a metal band where drumstick splinters were flying around, I was struggling to hear a 50 watt amp cranked into a 4x12" over the rest of the noise, and my ears were ringing the next day in spite of having worn earplugs.

My homebuilt amps both have a meter on the front to read the power tube cathode current, and since they're both class-AB fixed bias, I can estimate how much power the amp is putting out from the meter reading. At the above gig I remember it was pegged somewhere over 300mA. But I've also had sound guys start to whine as soon as the needle moved off its idle reading.

Guitar signals are not sine waves, and we get into issues of headroom, crest factor and so on. The average power may only be 1W, but it could take a 30W amp to reproduce the pick attacks cleanly. Crank up the distortion, and it all turns to square waves which are much easier to deal with.

[pdf64]

I think that your methodology of turning down the tone, and using a sine wave, are making your bench test unrepresentitive of your stage situation.
Because as you noticed, your guitar sounds nothing like a sine wave - there's loads of harmonics and dynamics.
To make an assessment of what amp power you are using, best just to play your guitar, tone set as normal, with the amp into the load resistor, and measure the maximum voltage across the load resistor with a true RMS meter, ideally one that captures the max level and responds into the kHz range.
Do it with and without whatever overdrive / booster pedals you would normally, and use the result which gives the highest maximum level.
Best to use a load resistor at least twice the power rating of your amp, I'll guess that you're going to be seeing a lot more that 2V.
That article had very little value IMO. It doesn't mention Fletcher Munson curves, which is likely the main reason why the perception of tonal balance changes with SPL.
It acknowledges that at stage volume (110dB), treble and bass settings will likely need to be reduced, compared to their setting at 85dB. So what's the point of buying a sound meter and setting them at 85dB in the first place?
Amp power rating are based on their maximum continuous output before/at clipping. A 100 watt amp may put out a continuous 140 watt clipped signal, and an even higher momentary (peak) signal.
Stage SPL will usually be set by how loud the drummer plays, 110dB is usually assumed to be a reasonable approximation. The article gives good info about how amp power and (total) speaker efficiency combine to give the SPL. But it seems to equate speaker distortion with power amp distortion, and to avoid turning your amp volume too high. It therefore seems to apply only to guitarists who get their tone from pedals and pre-amp overdrive, not those those for whom power amp overdrive plays a crucial part.
I would guess that with a quiet to average drummer, a 15 watt amp into 2x12" speakers, or 30 watts into 1x12" speaker, would be the minimum to get anywhere near a cleanish sound. Double those if your need clean headroom or have loud drummer, and double again for clean headroom and a loud drummer.
If you think that your stage volume levels are damaging your hearing (which it likely will do), then use ear plugs (flat frequency response type), or use isolating in ear monitoring. Peter.

[cbarrow7625]

I think your test setup included at least one mistake and one omission (possibly a second mistake).

1.) If you are going to measure the power out of an amp with a single sine wave you cannot have tone controls in line with the signal. In a guitar amp, at some frequencies the signal can be 10-15dB down from average depending on the tone control settings. You either need to a.) inject signal at the PI with a single sine wave to get a true power amp reading or (if you want the tone controls in the circuit) b.) measure the power with sine waves at multiple frequencies and average the readings properly to get an average of the amp output across the entire frequency range. The better method is to use a good automated measurement system with a swept sine wave, pink noise, MLS or other steady-state signal to capture the frequency response of the entire amp (including tone controls) so that you can see the power output at different frequencies and/or average the full output at all frequencies.

2.) Omission or mistake? I don't know. You didn't mention what type of meter you used. As previously stated, a true RMS meter is a must. But for this type of test, you also need a meter that is designed to work properly at all audio frequencies. Most inexpensive meters are only designed to work up to about 400Hz. They have a built-in internal low-pass that will make any frequencies above 400Hz measure lower than they actually are. Your 2 volt measurement may have been much higher than you actually measured due to the limitation of your meter(not to mention the EQ may have made the measurement at that frequency lower than the average). Do you know if this meter works at all audio frequencies? What type / model is it? The answer can often be found in the spec sheet for the meter.

As for the loudness of your speaker; for a point source loudspeaker, like a guitar speaker, the sound pressure drops off at 6dB per doubling of distance (known as the inverse square rule). So your 100dB 1w/1m speaker (with 1 watt applied) sounds like 94dB at 2m, 88dB at 4m, 82dB at 8m, etc. If you want to find out how loud that speaker will be with 50 or 100 Watts applied try using this online calculator:

SPL Calculator

A 100dB 1w/1m speaker will produce 120dB of output at 1 meter with 100watts applied. 116dB @ 2m, 112dB @ 4m, etc.

Hope this helps.

[clintonb]

Thanks for all the info everyone.

I have one multimeter that measures true RMS, but it seems to be broken. So I had to use my other meter. The article I read said that you don't need a true RMS if you test with a pure sine wave, otherwise you do.

One reason I was wondering about how many watts I was outputting was because I'm getting a 100-watt amp (I usually play a 15-watt or 50-watt amp) and I was curious if I could use it with my 1x12 cabinet whose speaker has has a 60-watt rating. Years ago, I had a Reverend Hellhound 1x12 combo amp that put out 60 watts, yet the speaker power rating was only 50 watts. Reverend said that was OK as long as you didn't turn the amp too loud. So I figured I could use a 1x12 speaker cabinet as long as I didn't turn it up too loud. I wanted to see how many watts I was actually outputting to see if I could safely use my 1x12 speaker cabinet.

[MWJB]

Not a good idea, rule of thumb is to use a speaker that is rated for twice the amp's output. What if you get to a gig and the drummer & everyone else feel like playing louder than usual...? You risk burning up the speaker's voice coil, which will show the OT an open load, which will kill your power tubes in an instant, possibly taking the OT with them.

Why not use the 1x12" cab in conjuction with another speaker cab? Observe manufacturers guide on speaker loads.

[cbarrow7625]

Yeah, +1 on that. Especially since we guitarists like to shove low crest factor square waves down to the speaker. There is more power in a square wave than a sine wave and there is more voice coil heating from low crest factor music than from the 6dB crest factor pink noise that is supposed to be used to rate the speakers in the first place.

2x power rating might be excessive for a guitar speaker / amp combo but if not blowing up the speaker is your #1 priority, then it is a good rule to follow.

[R.G.]

As a bit of commentary on the original issue, you may in fact only be using 1-2W most of the time.

Speakers convert electricity into sound pressure levels. The efficiency of the conversion varies. The stated efficiency in speaker specs is two numbers: the sound pressure level with one watt of drive and the maximum power input.

Most guitar speakers are very efficient - for the class of speakers, which as a group are quite inefficient overall. Typical guitar speakers produce from 96 to over 100db SPL with 1 watt of drive. For comparison, an air-driven jackhammer is about 100dbSPL if you're 1 meter from it. A jet engine 100meters away will be 110db to 140db.

So you're getting jackhammer-loud with your first watt. After that, the impression of "twice as loud" tends to need 10x the power or about 20db of SPL; so 10W should (very roughly) get you into the jet-engine loud range at at distance of 1M from the speaker. 100W would be maybe four times as loud perceptually, with 100x the power.

It's far simpler to get louder by changing speakers. A speaker with 6db higher SPL at 1W/1M is the same as using a 2x more powerful amplifier.

That's from one speaker. Using multiple speakers to spread the sound out can produce a much louder sound field at some distance from the speakers.

[cbarrow7625]

R.G., you don't need correcting very often (never that I can remember actually) but 10x power is a 10dB increase, not 20dB. A 10x voltage increase is 20dB.

10dB is typically considered "twice as loud". 10W is twice as loud as 1W. 100W is twice as loud as 10W.

[R.G.]

R.G., you don't need correcting very often (never that I can remember actually) but 10x power is a 10dB increase, not 20dB. A 10x voltage increase is 20dB.

10dB is typically considered "twice as loud". 10W is twice as loud as 1W. 100W is twice as loud as 10W.

Yep, you are right. 10x power is 10db, 10x voltage is 20db. I can only plead temporary insanity brought on by working mostly with voltage. 8-)

[Sweetfinger]

regarding wattage- I used to sell a lot of the Bruce Zinky "Lil Smokey" cigarette pack amps. They're powered by an LM386 which probably delivers half a watt at best through a tiny little speaker. The selling point was to plug it into a fairly efficient 4X12 cab, at which point they were shockingly loud! You could actually gig with one if you had to.