David Lamkins picked up his first guitar a long time ago. As best he can recall the year was 1967: the year of the Summer of Love. Four decades later David has conjured up an amalgam of folk, rock and jazz solo guitar music for the occasional intimate Portland audience.
LCW on Bandcamp
location: Portland, OR USA

Facets: amplifiers, loudness, physics, power, sound engineering, speakers, @musings info

More power!

Now that I've started playing my guitar through a modeling processor my amplification has changed as well. I'm using a "full-range, flat response" (FRFR) amplification system. This simply means that the amplification system is a sound reproduction system that doesn't add any significant coloration of its own to the sound. Examples of FRFR systems include PAs and keyboard amplifiers.

My FRFR system is a 300-watt portable PA. Yes, that's right... I wanted to play at lower volumes, so I went from a 60-watt tube amp to a 300-watt PA system.

How can that possibly make any sense?

Don't get taken in by wattage ratings. The perceived loudness of an amp depends upon the power, the overall frequency response of the system (including the speakers) and the efficiency of the speakers.

Guitar speakers are very efficient and emphasize the midrange frequencies to which our ears are most sensitive.

Full-range PA speakers can spec out as efficient as guitar speakers. However, they have a flatter, more balanced frequency response than guitar speakers. The midrange peak that gives guitar speakers their "punch" and "cut" doesn't exist on full-range speakers.

It has been a while since I've looked at home stereo speakers, but my recollection is that they're not as efficient as PA and guitar speakers. Also, they're almost certainly not designed for continuous high-powered operation.

Speaker efficiency is measured in dB/W at 1M from the speaker under certain standardized test conditions. The rating is an average of the efficiency as measured at many frequencies. The efficiency of a guitar speaker at very low or very high audio frequencies is negligible, but the efficiency at certain midrange frequencies (typically around 2KHz) is much higher than the average efficiency rating of the speaker.

Even within the range of frequencies put out by a guitar (from 80 Hz up to the speaker's high limit, typically somewhere around 5 KHz) the output for a given input will vary by frequency over a range of 10 dB or more. To put that into perspective, a 10 dB change in speaker efficiency is equivalent to a 10-times factor in amplifier power!

Look at this another way... The difference between the rated (average over all frequencies) efficiency of a guitar speaker and the actual efficiency at the (approximately) 2 KHz peak response can be in the range of 5 to 10 dB. That's like driving a flat-response speaker with a flat-response power amp that's three times (5 dB) to ten times (10 dB) more powerful than your guitar amp, using a graphic EQ with 2KHz boosted by the magnitude of the guitar speaker's peak.

Still not clear? OK, let's try this. To get comparable volumes from a FRFR system, you're probably going to need five to ten times the amount of power that you'd expect from your guitar amp. And that's only if you play clean all the time. If you push the output stage on a modern tube guitar amp, it's possible to get highly-distorted power at roughly twice the rated output of the amp. This isn't true of all tube amps: it depends upon a lot of factors. But it's a good rule of thumb, and the reason why we tend to use speakers rated at one-and-a-half to two times the power rating of a tube guitar amp. So let's revise that earlier statement: depending upon how hard you push your tube amp, a comparable FRFR rig is going to need anywhere from five to twenty times the power rating of your tube guitar amp in order to get a comparable volume level in the room!

Let's put some numbers to that... If you're used to playing through a fifty-watt guitar amp, you're going to need a FRFR system rated somewhere between 250 watts and 1,000 watts! Does that "higher powered" FRFR system still look like a great value for the money?

My personal experience bears this out. I replaced my Fender Vibro-King with a GT-8 and a Yamaha STAGEPAS 300 PA. The Vibro-King is rated at 60 watts. The PA is rated at 300 watts. The PA, even when pushed to just below clipping, is noticeably quieter than the Vibro-King. That works out OK for me since I didn't need or use the Vibro-King's flat-out volume. But if I did want significantly more volume I'd need a much bigger PA. Doubling the perceived volume takes a 6 dB to 10 dB increase in power, which is a factor of four to ten. (Yes, there's a range. Remember that we're dealing with perceptions and not with actual measurements.)

Does that means I'd have to use a 1,200 watt PA if I wanted to double my volume? Yes and no... If all other things were equal, yes. But the STAGEPAS is an integrated system and I don't know the efficiency of its speakers. I can infer something about their efficiency by working backwards from the specs that tell me about the maximum output power (300 watts) and maximum SPL (112 dB at 1M). Let's do some approximation based upon the following:

First I need to know the dB difference between 1 watt and 300 watts. There's a factor of three and two factors of ten. Decibels are additive, so 300 watts is 25 dB more than 1 watt.

Now I can infer the efficiency of the speaker by assuming that Yamaha's "maximum SPL" rating of 112 dB at 1M is based upon the full 300-watt output power. So if we reduced the output power to 1 watt we'd expect that the SPL would drop by 25 dB to 87 dB. And that figure, 87 dB/1 W at 1M, is the efficiency of the STAGEPAS speakers. Not great, but not unexpected for a compact full-range speaker...

A larger, heavier PA speaker would typically be 10 dB more efficient than the STAGEPAS speakers. Remember that 6 dB to 10 dB is equivalent to a perceived doubling of volume. So I could double the volume of my rig by keeping my 300-watt Yamaha amp and substituting a pair of speakers rated at 93 to 97 dB/W efficiency. Of course that would probably more than double the weight of my PA while quadrupling its size. Win some, lose some...

The point of this rather lengthy exercise, though, is to show you how wattage alone is insufficient to predict the volume of a FRFR rig. Using conventional power amps and PA speakers, you're going to need at least a 250-watt amp to get the same perceived volume that you'd get from a 50-watt tube amp. That's not a bad thing, it's just the way the physics works out.

Anecdotally, a guitarist I used to play with ran his modeler through a pair of the big Roland keyboard amps (180 watts per side, IIRC) and he could hold his own quite nicely; that was an insanely loud band, too.

The upside of all this is that clean solid-state amplification is lightweight and inexpensive. Furthermore, because both your amp and speaker are full-range devices, all of your tonal coloration comes from the modeler. This is the way that modelers are designed to be used (assuming that you're using the amp sims and not just the effects).

February 26 2007 18:21:50 GMT