How do you match the power ratings?

[Wilder Amplification]

the norm is to use amps with higher power capacity than the speaker ratings for best clean/clear sound.

Finally someone who gets it.

Yeah - I know there are exceptions and there are some really tech-savvy DJ's out there, somewhere, I just apparently don't get to meet many of them.

Of course not...they're smart enough not to blow anything up.

Now guitar amps, on the other hand, are a whole different animal where speaker/amp design rules are commonly ignored or intentionally violated to achieve tonal bliss. Pretty much whatever works for ya...

And in the case of tube guitar amplifiers, you'll never end up with DC on the speaker since they're transformer coupled. The OT primary winding on the other hand is a whole different story altogether.

[Chuck H]

So far I've been arguing the way "I" do things based on my crude upbringing with console speakers and what was considered common knowledge at the time. FWIW the OP IS TALKING ABOUT A PA SYSTEM not home audio or a state of the art entertainment center. And I'll bet six beer (import or whatever your country of origin is) that standard practice for sound REINFORCEMENT systems is still, and shall probably remain contrary to over rating the amplifier for the speaker system. WHY??? because bonehead musicians are using this stuff in a crude and basic manner.

And FWIW I'll put my old Marantz tube amp and Pioneer console speakers up for a challenge against any modern surround system. Oh, the speakers are rated a little over the amp, yes.

Chuck

[J M Fahey]

1) PLEASE stick to one (one) definition during this discussion, changing them ex-post-facto is the dirtiest trick of them all.

And in the case of tube guitar amplifiers, you'll never end up with DC on the speaker since they're transformer coupled.

30 seconds ago, you defined "DC" as the flat tops of clipped sinewaves.

the flat line on top of a clipped wave form (that flat line is DC).

Tube amps *can* also clip audio waves (sinusoidal or not) and those "square waves" can reach speakers, so what are you talking about now?
2) Anyway, those flat tops are not DC, but AC, by definition, even if squarewave AC.
Any voltage that either changes value continuously *or* stays at a given value ... for a few milliseconds , and then not only changes but gets opposite polarity ... for a few milliseconds and goes back and forth hundreds or thousands of times a second, *is* AC.
Besides the basic definition, we have some extra proof:
Proof 1: it gets transmitted to the load by capacitors (as in single supply SS amps) or through transformers (as in Tube amps and a few SS ones and a very important use of them: switching power supplies which exactly do that: transmit squarewaves through transformers).
Proof 2: those clipped sounds can be *heard*. Last time I checked we humans could not hear DC signals ..... nor could cone speakers reproduce them for us.
At most, we can feel static pressure on our eardrums in flight, underwater, or in a fast elevator.
3)

During the duration of the clip, the speaker literally stops moving,

No, it does not.
If you apply a voltage, whether a fixed one or a variable one, the voice coil will apply a force to a free-to-move mass , which will accelerate , in a direction defined by the relative polarities of the fixed magnetic field provided by the speaker magnet and the signal-dependent magnetic flux created by the current traveling along the voice-coil wire.
When the last one reverses polarity a few milliseconds (or even microseconds) later, the acceleration will change direction.
Nothing "static" there, by no means. We have movement all time long, no matter what the electrical waveshape !!!
Even "true" DC, as that coming from a battery, will cause acceleration here !!!, even as counterintuitive it looks on first sight.
The *only* way to have essentially no movement there, even for a short time, is having the voice coil mechanically hit the backplate, which is a very destructive phenomenon, and the probable death cause of many of those DJ subs, specially if said DJs use amplifiers capable of delivering far above the rated power handling , even more if they go trigger happy with equalizers and subharmonic generators.
Ooooooooops ! I just said a powerful amplifier can destroy lower rated speakers !!!
As a fresh and very close example, please read http://music-electronics-forum.com/t...ghlight=mackie where there are *many* examples of big trouble when a speaker is mated to a much higher power amplifier.
Agreeing with you, everybody says the sound is chest-thumping; unfortunately it also agrees with most others that the beast is as unreliable as a wet pack of matches.
End notes:
1) Please note I am not quoting other Net Pages to be backed up by Gurus' opinions , but referring straight to Physics Laws.
2) I have never belittled your former PA operator experience (nor referred to mine), but would like to clear 2 points:

.yet never hit clip and never took out speakers.

As to never killing a speaker, no doubt about that, if you say so. Thanks God many (most) are sensible enough to avoid that, even if at least for economic reasons.
As to >"never hit clip"<, how do you know?
The most mind-opening idea I ever had was to carry my MS-215 battery powered MiniScope everywhere, clipping its leads to anything, specially speaker terminals, and squint trying to decipher waveforms on its tiny 2" screen.
Going to PsychoAcoustics now:
Experiments show that sound systems clipping up to 10% of the time still sound quite clean.
Please notice that I did not say "10% distortion", but straight clipping (30% to 40%), 10% of all time.
3) Please do not take this as a personal attack and do not overreact to a perceived insult which does not exist.
My motivation is simply to avoid somebody who reads this post taking an unsafe decision when choosing his next speakers.
Anyway, I think that no harm will be done , because somebody "who knows" , ... well ... already "knows", and somebody which has doubts, after reading it will run away and avoid this topic like the Plague.
If anything I wrote causes you even the slightlest discomfort, please receive my humblest apologies.

[Wilder Amplification]

In regards to capacitor coupled/transformer coupled SS amps...they have been a thing of the past for a few decades now. Pretty much all SS power amplifiers made in the last 20 years are all direct/DC coupled to the speaker.

This is another reason why I mention adjusting the power amplifier knobs for proper gain structure (i.e. set them so that the amplifiers approach maximum clean output at exactly the same time that the mixer approaches maximum clean output). This makes it so that the amp's inputs are not so sensitive to upstream noise and transient spikes and also forces you to maximize clean gain at the front end. This essentially minimizes the noise floor while maximizing clean signal, which amounts to a maximized S/N ratio and amounts to a very clean and crisp sounding system with lots of punch. Set your global limiters up properly and the system will be VERY reliable.

So far I've been arguing the way "I" do things based on my crude upbringing with console speakers and what was considered common knowledge at the time. FWIW the OP IS TALKING ABOUT A PA SYSTEM not home audio or a state of the art entertainment center. And I'll bet six beer (import or whatever your country of origin is) that standard practice for sound REINFORCEMENT systems is still, and shall probably remain contrary to over rating the amplifier for the speaker system. WHY??? because bonehead musicians are using this stuff in a crude and basic manner.

And FWIW I'll put my old Marantz tube amp and Pioneer console speakers up for a challenge against any modern surround system. Oh, the speakers are rated a little over the amp, yes.

Chuck

So you basically were discounting everything I was stating all in an effort to "dumb it down" for "bonehead musicians"? It is this very dumbing down of this stuff that is the very reason WHY they are bone heads in the first place!

Why not teach them the CORRECT way to run a PA system and stop the trend of bonehead musicians blowing stuff up because they either don't understand signal path management/gain structure, OR they don't have enough available clean power to get the clean/tight/big/punchy sound they're after and are constantly pushing the amps into severe clipping in an attempt to get power that simply isn't there?

They're only "boneheads" because they don't know any better and haven't learned the "correct" way of doing things. Moreover, it's been pounded into their heads to "let your ears tell you what's 'right' " while not realizing that the human ear is a VERY crude listening device. Your ears can play tricks on you and they cannot hear EVERYTHING that's going on electrically within a system.

I never did understand those who think with the "Let your ears tell you what's 'right' " mentality. There are electrically harmful things that happen in a PA system that the human ear simply cannot detect (the primary reason why most pro audio gear has signal level and clipping indicators).

Case in point...ripple from an SMPS somehow gets injected into the audio. You will NEVER hear that because ripple hum in an SMPS power supply is well and above both the human hearing range AND the speaker's usable audio frequency range. The voice coil will still see it though and if it's powerful enough it will overheat the coil...but you would never know it was there because we simply cannot hear that high. Your ears ARE NOT the "be all end all" of how a PA system should be dialed in UNLESS everything in the system is "correct" ELECTRICALLY (i.e. available power, gain structure, etc etc).

Some people perceive upper frequency harsh clipping as "clarity"...these are the people who are constantly damaging horn drivers from clipping the horns...yet they continue to set it up the way they do because they want what they THINK is "clarity".

It's NOT just about "if it sounds good it is good" because "good" is a very subjective term. Make sure everything follows the standard electrically first. Once that's been verified, THEN concentrate on the sound.

Having said all of this...yes I know the OP was speaking of a live PA system. But...with more than enough power, the right speakers and mics and good sounding instruments you can get live sound with LOTS of punch and clarity with very little distortion and why would you NOT want that? I saw it mentioned earlier about "wanting a distorted sound for rock bands"...distortion in rock comes from the guitar amps being distorted. When the whole band is distorted it sounds like pure garage rock and you might as well not mic the band at all. Running things clean and tight with LOTS of power headroom sounds much more professional...ESPECIALLY with rock bands (think "baseball bat to the chest kick drum"...and absolute requirement for hard rock).

I see lots of bands who are "too loud" yet you can't really "feel" their sound. Harsh sounding upper frequency clipping is most of what causes people to get the "too loud" perception. When you have more power available than can be used, a good gain structure, the right mics and speakers you don't have this problem at all. I've mixed lots "louder" than most bands could get away with yet have never gotten the complaint that we were "too loud" because it was clean and crisp with lots of tight and punchy bass. We were "felt" more than we were "heard" and it simply takes lots of reserve power to make that happen. When you're bouncing the speaker off the amp's rails it not only sounds like garbage but it's not good for the speaker. With more clean power available than what can be used, you don't run into this problem.

[pdf64]

'The MOSFETs in a power amplifier are a metering device. They meter the amount of CURRENT that the amp's +/- rails can draw through the speaker based on how much input signal is applied to the MOSFET gates.'
My understanding is that audio power amps generally have a voltage amplifier architecture - it's the output voltage that is being controlled, rather than the current. Additional protection circuits might act to limit that current to a value that doesn't damage the output devices. But below that level, the output current isn't controlled.

[martin manning]

FWIW the OP IS TALKING ABOUT A PA SYSTEM...

I don't think that's clear at all:

...why are guitar/bass amplifiers and PA systems different? Additionally, how would you match the power ratings for a power amp run through a bass cabinet?

Seems like he is interested in understanding the differences between the different applications, but really wants to know how to set up a bass amp and cabinet. This is all interesting discussion (really!), but it'd be nice if he would clarify his request a little.

[Wilder Amplification]

My understanding is that audio power amps generally have a voltage amplifier architecture - it's the output voltage that is being controlled, rather than the current. Additional protection circuits might act to limit that current to a value that doesn't damage the output devices. But below that level, the output current isn't controlled.

Resistance to current flow creates voltage across that said resistance just as resistance to water flow creates pressure. The value of current you have flowing through a given value of resistance determines the voltage you will have across said resistance (E = I X R...simple Ohm's Law).

In order to have a measureable voltage you must have current flowing through a resistance. You can't think of it as a "one or the other" type of thing. That's like the hot rod community and how they talk about building a motor for "torque" or "horsepower" which doesn't make any sense since horsepower is derived from torque and RPM.

That being said, the +/- rails in an amplifier pull current through a load (i.e. the speaker). The MOSFETS directly control this load current. The current flowing through the load's resistance/impedance generates a voltage drop across said load. The product of the load current and the voltage dropped across the load is your output power.

[pdf64]

'In order to have a measureable voltage you must have current flowing through a resistance.'
A voltage (potential difference) can exist in the absence of current flow, eg a battery - the voltage doesn't just appear when a load is connected.
For a voltage amplifier with an input of X and gain A, the output voltage will be AX, the output voltage doesn't alter for any load from infinity ohms downwards, within the electrical constraints of the supply rails and open loop output impedance.
Intuitively we can see that's the case, as otherwise a power amp's output voltage would track the speaker impedance, and so rise significantly at mechanical resonance, where we might see the impedance rise to 10 x nominal.

[Wilder Amplification]

A voltage (potential difference) can exist in the absence of current flow, eg a battery - the voltage doesn't just appear when a load is connected.

Notice I stated "measurable voltage". You cannot see a voltage with a meter unless current is flowing. Current flows through the internal resistance of the meter itself and the meter calculates the voltage based on the amount of current flowing through its fixed internal resistance.

the output voltage doesn't alter for any load from infinity ohms downwards, within the electrical constraints of the supply rails and open loop output impedance.

The amount of CURRENT the rails must pull through the load to achieve the same output voltage across the load changes with load impedance. This means that more input signal is required to cause our MOSFET current metering devices to "open up more" to source more current to the load.

Intuitively we can see that's the case, as otherwise a power amp's output voltage would track the speaker impedance, and so rise significantly at mechanical resonance, where we might see the impedance rise to 10 x nominal.

The amount of CURRENT an amplifier must source to the load to achieve maximum output voltage does change with speaker impedance. It will have to source less current to the load to achieve the same output voltage across the load on a higher load impedance and vice versa on a lower load impedance.

Example...let's say you have a 16 ohm load on an amplifier that is capable of generating a 40 volt maximum swing regardless of load impedance. The rails will have to pull 2.5 Amps through that 16 ohm load in order to pull the load up to the full rail voltage (E = I x R...2.5 Amps x 16 Ohms = 40 Volts).

Now let's take that same amplifier and put a 4 ohm load on it. Now the rails will have to pull 10 Amps through the load to pull the load up to the rail voltage (E = I x R...10 Amps x 4 Ohms = 40 Volts).

So the maximum output voltage has not changed...however the amount of current required from the amp to swing the load up to full output voltage HAS changed. Provided the amp can source that amount of current to the load, the output voltage will remain the same but the required amount of current to pull the load up to that voltage will increase/decrease with load impedance.

[g1]

This current/voltage thing seems a little chicken & eggy to me.
Yes Jon, "Resistance to current flow creates voltage across that said resistance just as resistance to water flow creates pressure. The value of current you have flowing through a given value of resistance determines the voltage you will have across said resistance (E = I X R...simple Ohm's Law)."
However I can also state: The amount of current flowing through a resistance is determined by the voltage applied to the said resistance, just as water pressure creates flow dependant on resistance to the flow. (I = E/R).
Perhaps some of the problem is the old "bipolar are current controlled devices, tubes & Fets are voltage controlled".

Anyway, back to the original topic. Seems to me the 2 schools of thought are due to application differences. If no clipping or overpowering can be guaranteed then I will go with Jon that you should have some power amp headroom for peaks etc.
If there is any chance of clipping/overpowering (someone else running the system) then speakers that can handle all the amps power are a safer bet.
What would be great here is if anyone can provide numbers/ratios of amp to speaker power in various powered speaker cabs.

[Wilder Amplification]

I guess for me it makes it much easier to understand when you think of things in terms of current flow and voltage being a byproduct of a resistance imposed on said current flow. Current flow is precisely what electricity is the study of.

All too often I see people speak of "voltage" and "impedance" while current flow seems to be completely forgotten about. Again it's the whole "Torque vs Horsepower" debate of the hot rod community yet you cannot have a measurable quantity of one without the other.

How often do we see "impedance" misunderstood? Quite often...yet if you think of things as you would a normal power supply and focus on current and voltage, it becomes quite clear exactly what impedance is and does.

Take for instance impedance matching as it applies to pickups. You always hear "If you plug a 'high impedance' pickup into a 'low impedance' input you'll 'load' the pickup. A buffer will fix the problem". But how many people know WHY this happens? However if you explain it as -

"The pickup is like a small power supply. It supplies an output voltage but can source very little current. When you plug a high impedance pickup into a low impedance input, the low impedance of the input that you're connecting it to is literally trying to draw current that the pickup cannot source without maintaining good regulation, which causes the output voltage to drop. By placing an active buffer in line between the pickup and the input, you isolate the pickup from the low impedance input and the input instead draws current from the buffer's output, which via its power supply can source the current while maintaining regulation."

It actually makes more sense that way.

Another case in point is when people speak of "impedance matching" as it applies to tube amps and any other sort of transformer coupled amplifier. When you think of things in terms of impedance it's confusing. Yet when you realize that an output transformer is just a step down power transformer that operates at audio frequencies and the different taps just supply voltage and current, it becomes much easier to understand. If you put too low of a load on the thing, you will pull too much current and burn the thing up just like you would a power transformer.

When dealing with amplifiers it becomes much more clear if you look at amplifiers as an adjustable power supply. After all, that's exactly what they are.

I know that some people like to think in terms of voltage while others think in terms of current. However I think in terms of both. It's not about "applying a voltage and the voltage pushing current through" as most people's methods would imply. It's about a difference in charges that exists between two points that is attempting to balance out the charge differential when a current flow path (i.e. a "circuit") is completed between said two points. The voltage exists merely as a byproduct of the circuit's resistance to current flow. With zero ohms of resistance you would have no voltage because completing the circuit between the two points with zero ohms of resistance places the two points at the same potential (i.e. a charge differential between the two points would no longer exists as connecting them together with a zero ohm resistance would balance out the charge between said two points). With infinite ohms of resistance you would have the charge differential but no measurable current. With a finite value of resistance you always have both voltage and current and is why you should not hold preference to one while completely discounting the other.

[g1]

Great explanation Jon, especially the last sentence! I was wrongly getting the impression that you were "holding preference" to current and discounting voltage.
I think more reasons current is discounted is due to the fact that we rarely measure current (due to the need to break into the circuit) and also because so much electrical stuff uses the term "current draw".

Back again to the original topic, I forgot to mention that I think much of the disagreement here is due to not stipulating whether we are talking about speaker power in a) program OR b) RMS wattage.
I had thought that part of the idea behind the program power rating for speakers was so that people would have amps capable of more power than the speaker could really handle RMS wise, thus power amp headroom when speaker program power matches amps RMS power (in a convoluted sort of way).

[defaced]

Uh guys, the OP hasn't viewed this topic since it was posted a month ago. Just thought I'd let ya know you did all that ego flexing for nothing.

[Wilder Amplification]

Back again to the original topic, I forgot to mention that I think much of the disagreement here is due to not stipulating whether we are talking about speaker power in a) program OR b) RMS wattage.
I had thought that part of the idea behind the program power rating for speakers was so that people would have amps capable of more power than the speaker could really handle RMS wise, thus power amp headroom when speaker program power matches amps RMS power (in a convoluted sort of way).

RMS Power is another one of those misnomers. AFAIK, there is no such thing as "RMS Power". The text books don't even mention "RMS Power".

Some people refer to "Average Power" (Pavg) as "RMS Power" because you use RMS Voltage and RMS current to calculate it. However, Vrms x Irms = Pavg whereas Vpk x Ipk = Ppk and average power = 1/2 peak power.

Example...you have 120Vrms @ 1A rms -

120V x 1A = 120 Watts

Well if you figure the peak value of each -

120V x 1.414 = 169.68 Peak
1 x 1.414 = 1.414 Amps Peak

169.68Vpk x 1.414 Apk = 240 Wpk

And we know 240 Watts is double that of 120 Watts so this proves that Wavg = 1/2 Wpk, yet Wavg is calculated from RMS voltage and RMS current. But this certainly doesn't make it "RMS Power".

Now for a little brain teaser...the interesting thing is that if you multiply Vpk by Irms or multiply Vrms by Ipk, you end up with a power figure that is in fact 70.7% of the peak power and 141.4% of the average power -

120Vrms x 1.414Apk = 169.68 Watts

169.68Vpk x 1A rms = 169.68 Watts

240 Wpk x .707 = 169.68 Watts

120 Wavg x 1.414 = 169.68 Watts

So is it THAT power figure that people are really referring to when they speak of "RMS Power"? Or is it really "Average Power" they're referring to and are just misusing the terminology? Or perhaps maybe "RMS Power" is just a unit of measurement created by the Pro and Car Audio guys to make them sound like they know what they're talking about or something. But as far as I've seen in any textbook or any sort of technical literature that speaks of power figures, there's only "Peak" and "Average" power, and average power is really what counts.

[Steve Conner]

Oh, stop it, the peak/average/RMS power thing is quite confusing enough

Average power is what people mean by "RMS power" and it's the most physically meaningful power measurement.

Multiplying RMS voltage by peak current, etc, is meaningless.

Multiplying peak voltage by peak current gives you peak or "instantaneous" power. However, it's energy that burns things up, not power. (Which is another way of saying the above, that average power is the meaningful thing.)

You can have huge instantaneous power, but if it only lasts for a minuscule time, then very little energy is dissipated and not a lot happens. Taking that to its conclusion, instantaneous power means nothing because it's specified over an infinitely small time interval. You have to qualify it with a time, like "A 1.21 gigawatt pulse lasting for 3 nanoseconds"

On the PA/bass guitar front, I'd concur. If it was my own rig, I'd feel confident using amps a bit bigger than the speakers were rated for. If I was making it for someone else, I'd use smaller ones: they cost less and are less likely to blow up the speakers, although that's no guarantee. If you drive woofers down at a frequency where the cabinet doesn't load them properly, you can shred the cone with surprisingly little power. (energy?) Horn-loaded cabinets are notorious for it, because the bandpass effect of the horn filters out the distortion harmonics caused by over-excursion, so the sound guy can't hear how badly he's overdriving the woofers until it's too late. A high-pass filter is the fix for that.

One final thing is efficiency. The efficiency of a speaker is almost more important than its power rating. An efficient speaker will make the required amount of noise with little power, so the sound guy won't have to overdrive it and burn it out. Efficiency and power handling are completely independent, you can get speakers with any amount of either.