AVS Forum banner

Downsides/dangers of using less amp watts than the speaker is rated for ?

10K views 57 replies 16 participants last post by  johnson636 
#1 ·
I just built a SI HT 18 sub in a sealed box.


I wanted to hear/test it out, so I took the lead off my shaker amp and connected it to the sub.......that amp does 120 watts per channel.


All went well and it sounded good.........goes low/shakes everything and gets pretty loud, considering it can take 600 watts rms.

If I crank it way way up and it hits a super low note....like 10hz and under the clip light flashes for that instant and then goes away running only 120 watts to it. I know clipping is very bad.


But my thought is that it performs soo well on 120 watts, I could add a second one and run both from the inuke 1000.

The Inuke will do 330 ish real world watts to each channel.

Any danger..other than clipping from doing this ?

I understand DB max output will suffer, but with only 120 watts I can hear it 75 foot outside the house and it is plenty loud inside. The original plan was to bridge the Inuke 1000 so it would do approx 670 watts into this single 18, but I am not sure I need that much power.


Whats everyones thoughts on running 330 watts to a 600 watt rms speaker......?
 
See less See more
#2 ·
I just built a SI HT 18 sub in a sealed box.
I wanted to hear/test it out, so I took the lead off my shaker amp and connected it to the sub.......that amp does 120 watts per channel.
More power means higher output with less distortion. That translates into better sound. If you have the capability to measure the response of the sub, test it with both the 120 watt amp and any iNuke. You can then decide for yourself what to drive it with.
 
#4 ·
Whats everyones thoughts on running 330 watts to a 600 watt rms speaker......?
Think of it this way, if you have a 1,000 watt amp and only used 100 watts from it to get the output your required, would you need a 1,000 watt amp?

I have my garage speakers and they will easily handle 300 watts RMS until the sun burns out. I run anything from a chip amp at 7 watts per channel or a Crown PA amp with the limiter set at 250 watts per channel. Granted, the speakers are very efficient so 7 watts fills the garage with sound (and I have subs) 250 watts with the limiter light flashing will cause hearing damage rapidly. I know have a 150 watt amp that does the garage speakers full time as the PA amp now pushes a pair of subs.

If you don't clip the amplifier, all is well. The iNuke 1000 does 100 watts RMS into 2 channels at 8 ohms, 200 watts into 4 ohms and 300 watts into 2 ohms RMS. Bridged is 400 into 8 ohms and 600 into 4 ohms. Always look at the RMS ratings and skip the peak stuff... The iNuke 3000 DSP punches 300 watts into 8 ohms and 600 watts into 4 ohms--will be worth spending the extra 75 bucks to get triple the power.

Good luck! :cool:
 
#5 ·
Clipping will have no effect on a sub, as nearly all of the energy present in a clipped signal that's not contained in a clean signal lies well above the sub woofer pass band. The only limitations of an amp rated for less than the speaker is it won't go as loud as it's capable of and won't have any headroom. But if it's loud enough it's loud enough, no matter what the amp size. If you had the ability to see how little power your system is actually producing on average you'd be surprised, if not shocked.
 
#8 ·
Bill, I think I am confused more now about clipping based on what you just said. I saw this video explaining the effects of clipping a sub and thought it made sense, but seems to contradicts your first sentence. I believe the video is more in reference to using a sine wave at a specific frequency more so than music with multiple frequencies. Is this person wrong about clipping subs? Could you point me in the right direction if it is wrong? Thanks.


Eric


 
#6 ·
I already have the Inuke 1000dsp for 80 bucks used...just have been too lazy to do rack surgery and install it.......I will try one channel and bridged and see what it likes.....:D


I was just worried about underpowering damage issues, but I see that is pretty much a non issue with sub.


It is shame I can only do 2ohms or 8 ohms with 2 of these subs, no 4 ohm bridged to run both of them with the amp bridged.
 
#7 ·
FWIW, I currently have a Version 1 SI HT18D4 wired in series to 8ohms and running off an old pair of Marantz MA-500's in bridged mode which puts out about 360 watts and have been doing so for a little over a year now and have had no problems. I have shut the amps down a couple of times from excessive bass in movies but I have learned what settings to make to keep that from happening any more and the sub is going on strong. So I think you should be just fine.
 
  • Like
Reactions: unretarded
#9 · (Edited)
He's completely wrong. This explains to some extent why clipping is harmless to subs, but can easily toast tweeters:
http://www.bcae1.com/2ltlpwr.htm

The person in the video does not know what a square wave is. A square wave is a fundamental plus all of the harmonics of that fundamental at equal power. That power still does not exceed the power of the fundamental. Moreover, no sub woofer driver would ever see all of the power of the harmonics anyway, as they're filtered out by the inductance of the voice coil. The same is true of woofers. Only tweeters and mid ranges receive more than normal power input when a signal is clipped.

I didn't bother to look at all of the video, but if it's the usual under-powering mythology he probably says that the driver will stop moving at the peak plateaus, causing the coil to overheat. That never happens. He might even say that because of the plateaus that the signal is DC. That's also not true. DC does not alternate through zero.

Finally, drivers never see the square wave as seen on an oscilloscope. When you see someone using an oscilloscope trace to show why clipping kills subs or woofers you know they're wrong.

That's not to say that a 50w amp can't blow a 100w woofer. It can, simply because that 50w is measured at low THD and constant power. Just about any amp can deliver at least twice its rated power at high THD in short bursts, so it can toast a driver rated higher than the amp rating. Since the high power transient is accompanied by high THD the assumption is often made that it was the high THD that killed the driver, but it wasn't. It was over powering.
 
#15 ·
After being on here every day for almost a year and reading a few hours each day.......I 100% trust what Bill is saying....thanks.



I also understand wanting him to explain as some of this stuff takes a while to grasp.


One thing I know to be true, is at almost 50 years old I have come to understand most of what people accept as general understanding, whatever the subject, is based in rumor,hearsay and sprinkled with a few facts......:D
 
#31 ·
One thing I know to be true, is at almost 50 years old I have come to understand most of what people accept as general understanding, whatever the subject, is based in rumor,hearsay and sprinkled with a few facts......:D
One of the greatest statements ever! It is completely true. A lot of people just spout off what they have heard and never researched it or know if it is even true or why.

P.S. Good to see someone else do a budget theater!
 
#17 · (Edited)
You are all wrong... :p

Clipping or Watts doesn't "necessarily" kill speakers, nor even the lack of either.

But thermal and mechanical failures DO. (heat and bottoming out etc)

The 5 most common mechanical failures are (in no particular order):
-The dust cap flies off
-The thing bottoms out, causing things to smash together like a car accident.
-The coil is accelerated with such force that the cone bends or the coil flies through the cone/cap.
-The spider glue comes apart
-The coil heats up and expands, melts it or the glue melts etc etc.

As for heat. That is cause by power being applied over a given amount of time. (There is a TIME component.)
There is flash heating, rapid heating, or long-term heating.
*All of this heat is caused by: current being resisted.

Each brand/model will have different characteristics and thus different BTU/Time curves, or Watt/Time curves I suppose, aka an oversimplified term called "power handling".

Databass regularly clips 20kW into every driver tested above 40hz, but the TIME duration is a fraction of a second.
The amount of TIME the power is applied for MATTERS A LOT.
Are we talking 20kW for 20ms, 2kW for 10 seconds, or 200watts-average applied for hours on end... It is impossible to say, that depends on what the user is doing to it. The songs, the amplifier used, the driver's cooling abilities, and how hard it is pushed.

A tweeter might be rated for 10watts, but certainly not @ 10Hz unclipped; perhaps not even 1 watt @ 10Hz (even if it didn't bottom out or clip once.)

Even subwoofers aren't rated for full power @ 1Hz, or even full excursion at 1Hz for serious lengths of time. You aren't clipping it or exceeding its "supposedly rated" wattage, it doesn't have to be either to kill it, but you ARE exceeding its thermal-handling. That is what kills it: ALWAYS
It's a lot more complicated than just "clipping or not clipping" or "x watts or Y watts".

*Technically, if you could reduce the coil to 0K, it would become of a super conductor, and the ohm's would drop to 0 and therefore offer no resistance to any amount of current, and thus no run-away heat buildup.
The amplifier wouldn't enjoy the 0-ohms unless also cooled to 0K, but that's another ball of wax.
 
#18 ·
Hey BTH... Duh, and what was your point again in that tirade of random (mostly off topic) factoids? :p

The subject is the dangers (if any) of a clipped waveform. Of course duration is a factor, so it's a good point to bring up even though it is obvious. Thermal runaway many times goes hand in hand with a clipped signal in both driver's coils and amplifiers even when the driver and amp are closely matched in the system. That was my point and I don't think that I'm wrong.
 
#19 ·
There are 3 common types of clipping.
Input, output, and source (today that would be digital).

Digital clipping happens when all the byte's bits are ones.

Input clipping is typically when the user is being drunk with a pre-amp or trying to send xlr signals into rca inputs.

Output clipping can happen at any voltage or current. When there is no more VA to give, then clipping results. Which will cause high frequency harmonics, which in the mostly converted to heat (especially for a sub).

If the coil and amp can handle the extra heat, then it won't be hurt by any of those 3 clipping types.

If you could water cool or LOX your coil and amp, it could probably handle full clipping all day long. But passive air cooling in a tight space, not so much...

The lack of VA is typically either the SP (rectifier, transformer etc) or output transistors. The watts or VA depends on the load's ohms and the beefyness of said components, as well as input voltage and amp sensitivity/gain.
 
#20 ·
Notice how i didn't say RMS.

Impluses can be clipped just as easily as signals like sinewaves or complex music can.

Their values don't matter... unless you are trying to calculate BTU's or Joules... then you'd need to know the applied-wattage and expected power-handling and compare them in RT. No end users do that. Easier to just use a laser thermometer and guess.
 
#24 ·
The level on the clipped signal is higher than the unclipped signal... what are you trying to even show here? If you raise the level, there will obviously be more heat, whether the signal clips not.
:confused:

There's so much nonsense, cross talk, and people misunderstanding basic concepts in this thread it's painful to read.
 
#29 ·
A square wave is going to produce more total energy over time than a regular sine wave at the same amplitude. More energy over the same time = more heat, and heat can kill any speaker. But I suppose this is really only a problem when the amount of power being fed to the speaker is already close to the limits of what the coil can handle, so if you are grossly underpowered already, this shouldn't have much effect.

I can see where a large woofer with a lot of mass in the cone is not going to just stop when the wave squares off, but it would slow down. This would have the effect of focusing all the energy on one small section of the coil for a slightly longer period of time, rather than spreading it out more evenly over more of the coil. So yeah, a clipped output (or a square wave) would make a voice coil hotter in certain spots, but it won't stop moving completely so it's really just a matter of less efficiency in cooling capacity. Again, if you are underpowering it, this shouldn't be a problem, it should be able to handle the extra heat.

I would think the most danger in clipping is in the mechanical end of things, specifically at the ends of the voice coil. The magnetic field of a speaker is focused on a small section of the voice coil. When the coil is centered in the magnetic field, any power going to that coil will convert to mechanical energy (and heat) within that small focused area, and the rest of the coil is just adding resistance from the copper wire. When the woofer extends out to the limit in either direction, it is possible that less of the coil will be in that magnetic field, meaning that any power going to it will be focused on an even smaller section of the coil. Less coil with the same amount of energy means more heat and less power handling capability. If you are clipping the signal, you are putting max output from the amp to the speaker for a longer time while it is at the lowest possible capability of handling that power without damage. Plus heating the end of a coil has more risk in warping the coil than heating the middle of it. Warp the coil and you start to get rub, which can lead to a short, which will turn into catastrophic failure.

But again, if you are underpowering the speaker, what is the chance that you are pushing excursion to the point where you are losing the ability to handle the power due to the coil moving too far out of the magnetic field?

All this is VERY dependent on the design and quality of the speaker. Some speakers you could run at 90% rated power with a full square wave and never have an issue. Others might fry themselves at 50% rated power with a clipped or square wave.

So while I wouldn't say it is a hard rule that underpowering a speaker makes clipping a non-factor, I would say that it depends heavily on the design of the speaker and exactly how much power you put to it. There is certainly less risk in blowing up a speaker with less than the rated power though. And other than that, there is no downside to running a speaker with less power than it is rated at.
 
#30 ·
I can see where a large woofer with a lot of mass in the cone is not going to just stop when the wave squares off, but it would slow down.
That's a common misconception, based on looking at a set frequency square wave on a scope, then assuming that's the waveform that the driver attempts to reproduce. In practice that never happens. Here's a link to another authoritative source:
http://forum.qscservice.com/viewtopic.php?f=29&t=2736&sid=546d7d2ba49d2b016b8b9bfeaf969008
 
#42 ·
Since we have moved beyond the origial scope of the qurstion, which is fine, from my understanding there are several different styles/types of clipping.



Input clipped, which be signal based or mismatched voltages......amp clipping, where it runs out of power and output clipping, where it can be power based or signal based, bringing full circle back to input clipped, since in order to have input clipping, something has to be outputting.....:D


I guess it would be signal clipping.........voltage clipping and power clipping.......which could be a rca voltage mismatch and a amp/watts/power clipping where the demands of the signal/driver exceed the amps capacity to deliver.


I also imagine these would create different outcomes in driver damage possibilities.......or some would create no dangers.

such as the AVR outputting .9volts and the amp wanting 2 volts input, so in essence that style of clipping would be the same as turning down the gain......it would lessen the load, rather than rise it.........:confused:


I have no idea as everything ties/works together dynamically and can not be analyzed seperate.


As a example was pointed out to me when building my box........winisd could show driver excusion over xmax at 10hz with max amp power.......but the apparent power shows 1200 watts and I only have 600 watts and that is also at max, which no one really listens at and if they do it is for seconds and if it hits that in a movie it is for milliseconds and does not have the amp power to push it into excursion anyway.............making the real world usage render the possibility of over excursion almost impossible due to insufficient power available to do that and no almost one maxes out the volume while watching a movie.......I certainly do not.


I know real world performance can get pretty complex......:D
 
#44 ·
I only have 600 watts and that is also at max, which no one really listens at and if they do it is for seconds and if it hits that in a movie it is for milliseconds and does not have the amp power to push it into excursion anyway.............making the real world usage render the possibility of over excursion almost impossible due to insufficient power available to do that and no almost one maxes out the volume while watching a movie.......I certainly do not.
Most folks here push closer to 1kW to the HT's for dynamic source. You'd be surprised how fast power gets eaten up. Remember, the difference between 600W and 1000W is 2.2dB and that's only if you're not hitting any other kind of mechanical limit from the driver. Not that noticeable of a change. If you're curious about real world performance you can set up a jig and measure your excursion. Play a source with strong ULF like The Incredible Hulk when he punches Abomination into the ground or How to Train Your Dragon's epic dragon crash scene. Turn your system to reference level and look for clip lights.
 
#55 ·
I have a question


















Cause I sure the hell am:p
 
  • Like
Reactions: unretarded
#58 ·
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top