Printer Friendly Forum Posts - Under pwer vs. Over power

Ok, I've seen this topic discussed alot on this forum, and people continue to offer up both as suggestions.

Can we clear this up. Do you get an amp capable of putting out more power then your speakers can handle, or less power?

I personally think that having a more powerful amp would be the way to go because if you use an under powered amp you would be more inclined to turn up the gain more which would also increase the distortion. Whereas with a more powerful amp you could leave the gain lower which would produce less distortion then the under powered amp at higher gains.

What are your opinions on this subject?

-------------

That is how I see it. I've always been told more power is better than not enough for this reason.

-------------
'85 Toy

IT DOES NOT MATTER. As long as you know what you're doing and you set up your system correctly, you can use any amp with any speaker.

The key to that, of course, is "know what you're doing." This means fully understanding both your amp's and your loudspeaker's capabilities.

FOR MOST CAR STEREO APPLICATIONS, matching the listed RMS rating for speaker and amp is the commonly acceptable approach since most people do not really understand what they are doing. Then, at least they have a chance of not destroying their gear and of getting the most out of it.

For professional applications, I always recomend a 2X to 3X safety factor between amplifier rating and loudspeaker rating, and always based on the amplifier rating. Meaning, if your amps are ratied at 250 watts per channel, your loudspeakers need to be capable of handling 500 watts or greater.

And, just one more time for clarity, remember this: THERE IS NO SUCH THING AS UNDER-POWERING A LOUDSPEAKER. Any amplifier properly set up can be used with any loudspeaker.

The only non-mechanical thing that can damage a loudspeaker is heat. The only way an amplifier will generate heat inside a loudspeaker is through TOO MUCH POWER. The ways an amplifier can supply too much power is through CLIPPING (which generates heat both from straight rail voltage being passed through and by causing the mechanical motion of the voice coil to stop) or by simply producing higher power output than the loudspeaker's thermal limits.

-------------
Support the12volt.com

DYohn] wrote:

nd, just one more time for clarity, remember this: THERE IS NO SUCH THING AS UNDER-POWERING A LOUDSPEAKER. Any amplifier properly set up can be used with any loudspeaker.

Yeah, I know there is no such thing as under powering a speaker. Like you said, any amp can power any speaker

What I was/am thinking is that is you use a 50w RMS amp on a 100w RMS driver, the owner of the components would be inclined to turn the gain on the amp to 100% to acheive the full potential of the components with what is had. As we/I already know, this would make it very easy to cause the amp to clip if the pre-amp is not properly modualted.

I guess I'm saying that with an amp capable of putting out more power then the speaker can handle, the owner would be less inclined to turn the gain up to high, which would help to prevent clipping.

Does this make any sene, or do I just sound like another one of these dumb people posted_image

-------------

Poormanq45 wrote:

What I was/am thinking is that is you use a 50w RMS amp on a 100w RMS driver, the owner of the components would be inclined to turn the gain on the amp to 100% to acheive the full potential of the components with what is had. As we/I already know, this would make it very easy to cause the amp to clip if the pre-amp is not properly modualted.

This is exactly what I meant when I said the average person does not know what they are doing. "Achieving the full potential of the components" is a very dumb thing to think about but it is how most people think. It drives me mad sometimes. "Turing the amp gain to 100%" will, indeed, create clipping and is a very dumb thing to do in ANY case. This should never, never be done. It will destroy speakers and maybe destroy the amp. It causes severe clipping. There is no way around it. This statement indicates the owner does not understand amplifier gain. Amplifier input gain is NOT a volume control!!!!!

Bottom line: if the amplifier gain is set correctly, matched to the head unit output, the amp will never clip and it will put out its full capability as a clean signal. If the speakers receive this clean, unclipped signal, they will operate at their "full potential" no matter what the power ratings are. The difference in SPL between 50 watts and 100 watts is 3db. 3db is about 1/3 louder. It is not significant. Most speakers can produce their "full potential" as far as frequency response with 2 watts input. Many speakers can achieve their "full potential" of SPL with 10 watts input.

-------------
Support the12volt.com

Amen!!!! Why do more people not understand this concept?!

-------------
New Project: 2003 Pathfinder

There are many misconceptions floating around concerning this topic..people pick up on them and believe them. That's why this topic is so misunderstood. If you're not 100% positive you know what you're doing...play it safe and match RMS ratings..less power is fine as well.

-------------
Dave

Pioneer Premier DEH-P660
15" Kicker CVR
Profile AP1000M
It's Loud

I know that less power is fine:

Ok, this may sound like a dumb question, but what EXACTLY does the gain controller on an amplifier do? I always thought that it controlled the sensitivity, but obviously that is not correct. So what is it?

-------------

Poormanq45 wrote:

but what EXACTLY does the gain controller on an amplifier do? I always thought that it controlled the sensitivity, but obviously that is not correct. So what is it?

It controls the sensitivity.

That is such an awsome post Dyohn. It's amazing how many people ignore our recommendation to them to run a 700-1000 watt amp and instead pick up a 1500 watt one because of the ratings on our subwoofers. Heading up the warranty department, I see a lot of subs come back because of this. In fact we are deflating our ratings a little bit on the new lines just to add a "dummy" factor into it. Apparently using the pro sound standard for rating doesn't work in the car audio market.

I do want to make one small clarification though. A speakers voice coil will blow with too much unclipped power as well. The speaker doesn't care if the signal is clipped or not, if the power levels exceed it's thermal capabilities it will blow. This is because to a speaker a clipped signal is no different than an audio signal. It only cares about the power level of that signal. There is of course a time factor involved as well. Music is dynamic and it's peaks in power levels might not be long enough to heat up a coil beyond it's limits. That's how SPL competitors can send 10 kilwatts into subs without blowing them. One of our competitors (Scottie Johnson) does this in his vehicle and has never blown a voice coil. But these people know what they are doing.

Steven Kephart

Adire Audio

-------------

Thanks Steven. Yes, exactly the point I was trying to make. The power presented to a voice coil is disapated as heat. The more power for longer durations = more heat whether the power is clipped or not; less cone movement = more heat since it's the cone movement that cools the loudspeaker. Clipped power = worst of both since not only will it pass too high a voltage, but since it is in effect a DC signal cone movement stops and even more heat results.

Subjecting a speaker to 50KW in a 100msec burst is a completely different thing than the average user "cranking their gains." Like I said, if the user knows what they are doing, anything is possible! posted_image Thanks for weighing in on this one.

-------------
Support the12volt.com

Here is my point, and I think it is what poormanq45 is talking about also. I have seen lots of people buy 70, 100, 150 watt rms amps. They have speakers capable of handling 400 or 500 watts rms. You hook everything up correct and they leave. A couple of days later they hear somebody with the same speakers that they have, but hooked up to a 500 watt or more amp. Usually the first thing they are going to do is "turn up the gain on the amp". This is their quick fix to get more sound. More sound and more distortion. Then they come back and say their system sounds like crap. I've seen this more times than I can remember. Most people think the gain control IS a volume control. That is why they are not installing stereos. They are far more likely to do this with less power rms than with more power. Or at least that is what I think he was talking about.

-------------
'85 Toy

Yes, this happens, people are idiots and they will do this no matter what size gear they have. Installers should, in my opinion, properly adjust the system and tell the system owner that if they adjust the gains after they leave the shop their warranty is void... or atleast YOU won't honor it. I've even used red wax on gain and crossover controls to detect if they have been moved.

-------------
Support the12volt.com

Why are they far more likely to do this with less power? Higher amplifier power doesn't make the speakers any more efficient relative to the other speakers in the system.

Here's my take. SPLWannaBe wants his bass VERY HOT compared to the rest of the system. Rather than turning his gain DOWN on the mains, he turns it UP on the subwoofer. This is just dumb. He'd do this regardless of whether he had 10,000 watts or 2.

In sum: not thinking is what makes people more likely to jack up the gain - REGARDLESS OF AMPLIFIER POWER.

Lesson: THINK!

-------------
New Project: 2003 Pathfinder

Exactly.

-------------
Support the12volt.com

DYohn] wrote:

p> Clipped power = worst of both since not only will it pass too high a voltage, but since it is in effect a DC signal cone movement stops and even more heat results.

I wasn't sure if you had read the other thread about this topic we just had. I had Dan come in and comment about the DC signal and he had some interesting information. Here's the link: https://www.the12volt.com/installbay/forum_posts.asp?tid=51115&PN=2&tpn=3

Steven Kephart

Adire Audio

-------------

Ah yes, of course in the strictest sense there is no DC present in even a square wave (100% clipping) since it is a Fourier transform, and even in it's purest, single frequency form is constantly alternating back and forth between two extremes. What I said was a clipped signal is "in effect" a DC signal, since when the wave is in the flat portion (clipped) it is unchanging like DC. As the frequency of the signal decreases, and as the tone is more mono-tone (a singal frequency) the net signal approaches two DC signals, one at X+ and one at X-.

My reasoning behind stating that a clipped signal makes an over-power situation worse (or creates an over-power situation like Mr. Wiggins explained in the other thread) is the loudspeaker behaves for each half-cycle as if the signal was DC. It stops moving at the extreme of the signal-created excursion stroke and thus the heat generated by the current flow must be disipated solely through the conduction and convection properties of the voice coil winding (and mounting system) itself. There is no air flow created through system movement at these extremes, which can accelerate failure due to heat. Of course it will as instantaneously as possible shift to the oposite extreme for every other half cycle.

I don't think we're contradicting one another, just looking at the behavior slightly differently. Dan's more of a pure theoretical engineer; I'm more of an applications guy.

-------------
Support the12volt.com

supradude wrote:

Here is my point, and I think it is what poormanq45 is talking about also. I have seen lots of people buy 70, 100, 150 watt rms amps. They have speakers capable of handling 400 or 500 watts rms. You hook everything up correct and they leave. A couple of days later they hear somebody with the same speakers that they have, but hooked up to a 500 watt or more amp. Usually the first thing they are going to do is "turn up the gain on the amp". This is their quick fix to get more sound. More sound and more distortion. Then they come back and say their system sounds like crap. I've seen this more times than I can remember. Most people think the gain control IS a volume control. That is why they are not installing stereos. They are far more likely to do this with less power rms than with more power. Or at least that is what I think he was talking about.

Thank you, that is exactly what I was talking about.

Also, if the gain controls the sensitivity, is there a problem with setting the gain to ~100% and using the Headunit to actaully control the volume and amplifier output? I think this is feasable, but I'm guessing that setting the gain lower is just an added layer of protection, correct?

Steven Kephart: Thanks for your informative post, but I think you kind of belittled me posted_image

. I know that heat is what actually "kills" a voice coil. All it is is a semi-conductor. All semi-conductors have a percentage of inefficieny. This loss usually appears in the form of heat. This increased heat, at a certain point, sort of starts a "snow ball" effect, as the heat increases, the resistance increases, which further increases the heat, and so on until something melts.

-------------

The purpose of the gain control is to match the input sensitivity of the amplifier to the output voltage of the head unit. This is what ALLOWS the HU to control the volume and to have the full range of the amplifier available. I really suggest you do some reading. Here's a decent place to start.

A voice coil is NOT a "semi-conductor." A voice coil is a pure electrical conductor. It is made from wire.

-------------
Support the12volt.com

Sorry, I used the wrong term. I used semi-conductor, when I actually meant conductor. i was just thinking of it in reference to a super-conductor with an efficiency of ~%100.

Oops, Edited for stupidity

So then, what happens if you set the amp for a HU that puts out 8vrms, but the HU you're using actually puts out 2vrms?

-------------

Ok, just excuse my post right above this one.

SO then, as you turn the gain "up", or usually clockwise, you are setting it for a LOWer input vrms?

-------------

DYohn] wrote:

p>Ah yes, of course in the strictest sense there is no DC present in even a square wave (100% clipping) since it is a Fourier transform, and even in it's purest, single frequency form is constantly alternating back and forth between two extremes. What I said was a clipped signal is "in effect" a DC signal, since when the wave is in the flat portion (clipped) it is unchanging like DC. As the frequency of the signal decreases, and as the tone is more mono-tone (a singal frequency) the net signal approaches two DC signals, one at X+ and one at X-.
My reasoning behind stating that a clipped signal makes an over-power situation worse (or creates an over-power situation like Mr. Wiggins explained in the other thread) is the loudspeaker behaves for each half-cycle as if the signal was DC. It stops moving at the extreme of the signal-created excursion stroke and thus the heat generated by the current flow must be disipated solely through the conduction and convection properties of the voice coil winding (and mounting system) itself. There is no air flow created through system movement at these extremes, which can accelerate failure due to heat. Of course it will as instantaneously as possible shift to the oposite extreme for every other half cycle.
I don't think we're contradicting one another, just looking at the behavior slightly differently. Dan's more of a pure theoretical engineer; I'm more of an applications guy.

Wow, thanks for the explanation. Now I know I am out of my league here, so please don't take this as me disagreeing. I am just playing devils advocate to learn.

But is the cone held in place long enough that there is lack of air movement? I can see this maybe at a REALLY low frequency like say 5 Hz. But even at 20 Hz the change to polar extremes is pretty quick. In fact that video I linked in that other thread was taken with the subwoofer playing at about 15 Hz. And yet it is moving back and forth at a decent rate. I suppose one difference between a sine wave and a square wave is that acceleration is attempting to change at a much greater rate with a square wave. Or is that incorrect?

Steven Kephart

Adire Audio

Now don't get me wrong. I think explaining it to be similar to DC offers a great visual aid for those learning.

-------------

Steven Kephart wrote:

But is the cone held in place long enough that there is lack of air movement? I can see this maybe at a REALLY low frequency like say 5 Hz. But even at 20 Hz the change to polar extremes is pretty quick. In fact that video I linked in that other thread was taken with the subwoofer playing at about 15 Hz. And yet it is moving back and forth at a decent rate. I suppose one difference between a sine wave and a square wave is that acceleration is attempting to change at a much greater rate with a square wave. Or is that incorrect?

That's a very good question. In my experience, well-made woofers can probably handle the extra heat much better than more poorly designed and built ones. Cheaper speakers use cheaper components, of course. You are correct that a square wave attempts to change state (phase) instantly. Since infinite acceleration is impossible in any system with mass, the loudspeaker cannot react at the same rate as electricity can. In this case, a higher frequency square wave can cause the voice coil to "twitch" as it tries to alternate between extreems and it cannot keep up. It simply heats up in this case. Tweeters are prone to damage from clipping in this manner.

-------------
Support the12volt.com

Poormanq45 wrote:

So then, what happens if you set the amp for a HU that puts out 8vrms, but the HU you're using actually puts out 2vrms?

In this case the amp will receive a lower signal than it is adjusted for and will not increase output at the same rate as the HU (it won't get as loud). You want the maximums matched so that the amplifier output increases at the same rate as the HU preamp output.

Poormanq45 wrote:

SO then, as you turn the gain "up", or usually clockwise, you are setting it for a LOWer input vrms?

This is generally correct.

-------------
Support the12volt.com

DYohn] wrote:

p> You are correct that a square wave attempts to change state (phase) instantly. Since infinite acceleration is impossible in any system with mass, the loudspeaker cannot react at the same rate as electricity can. In this case, a higher frequency square wave can cause the voice coil to "twitch" as it tries to alternate between extreems and it cannot keep up. It simply heats up in this case. Tweeters are prone to damage from clipping in this manner.

Forgive me for going off on a tangent that has little to do with the original topic. But this is a topic I am still trying to fully comprehend. I think the mistake that most people make, including many engineers we have found, is that they think in terms of acceleration, rather than change of acceleration. From what I have been told, speakers are constant acceleration devices. Now with a squre wave as far as I can tell the rate in change of acceleration is much higher because it is going directly from infinity to zero, rather than a smooth transition with a sine wave. Now mass will effect a square wave just like a sine wave. It won't effect the rate in change of acceleration because mass is constant. It will only limit the peaks of those waves (SPL). However the inductance within a speaker will effect the rate in change of acceleration since inductance tries to keep current from changing. This should make for an interesting looking wave form produced by a speaker with high inductance trying to produce a square wave. posted_image

But then maybe I'm wrong. Maybe in such an extreme instance like a square wave, mass does start to effect the transient response for the reasons you mention. But then as I mentioned above, inductance is effecting it as well. I would guess one or the other would dominate. But which one?

Steven Kephart

Adire Audio

-------------

I am not a loudspeaker designer, but from my understanding of physics and electricity you must take into account both the inertia of the mass being accelerated (any change of momentum or directions requiring acceleration in order for the change to occur) AND you must take into effect the electro-magnetic behavior of the voice coil windings as current carrying conductors in a magnetic field. Both Newton and Faraday have a say in this one, I suspect.

I think both mass and inductance/impedence have an impact on transient reaction and response, and both work together to determine the ability of a loudspeaker to reproduce any soundwave. Which has a greater impact probably depends more on the specific parameters at each measurment than on any generally predictable behavior. I'd be willing to bet the contribution from mass and the contribution from inductance, reactance, resistance, flux density, etc. changes over the loudspeaker's response curve and as input levels change; at some times one contrinutes more and at other times other factors do.

This is conjecture on my part, but it seem to fit with my basic understanding of the phenomenon occurring in a speaker voice coil. I could of course be wrong!

-------------
Support the12volt.com

steven wrote:

But is the cone held in place long enough that there is lack of air movement? I can see this maybe at a REALLY low frequency like say 5 Hz. But even at 20 Hz the change to polar extremes is pretty quick.

Steven Kephart

Adire Audio

Nah, IIRC, a halving of frequency results in a quadrupling of excursion. If this is indeed correct, then the cone would be moving back and forth at a slower rate, but it would be moving much farther. So the air movement would still be "There"

-------------

Poormanq45 wrote:

steven wrote:

But is the cone held in place long enough that there is lack of air movement? I can see this maybe at a REALLY low frequency like say 5 Hz. But even at 20 Hz the change to polar extremes is pretty quick.

Steven Kephart

Adire Audio

Nah, IIRC, a halving of frequency results in a quadrupling of excursion. If this is indeed correct, then the cone would be moving back and forth at a slower rate, but it would be moving much farther. So the air movement would still be "There"

That is correct, and a VERY good point. However it wouldn't be moving back and forth at a slower rate because that would change the frequency produced. posted_image It just means that for every doubling of static movement, there is 4 times more air moving past the coil (or is that even correct?). So the question still stands. However this does become even more complex when we add enclosure effects into the mix.

Steven Kephart

Adire Audio

-------------

Ok, I know I'm a newbie guys, but I'm trying to learn the in's and out's of the gain control before I bust into turning it all over the place....I've read all the posts in this topic and also have read the BCAE article. Here's my question. My new Xenon 200.4 has (2) input sensitivities on it. One for front, one for rear. They go from 200 millivolts to 8 volts. Was that article saying on BCAE that the 200 millivolts side was actually the high side? Also, if my HU is rated at 4 volts out, should I set my gains halfway to match my HU to my amp? Hope these aren't dumb questions, I just don't want to be one of those fools out there...I'd like to actually understand what I'm doing before I do it.