Hi,
I have combed the internet without an answer for the below and hope someone can give me an answer.
I am making a chip amplifier, LM6886 68W per channel, powered by a toroidal 28V 250W. The speakers I have are rated 8 Ohm, 250W, 8 Inch full range (+tweeter). My intension is to listen to (high) quality well balanced music at low sound levels, without distortion.
I have read - if the speakers have a much higher wattage that the amplifier then there will be clipping, sound will be distorted, sound will not be lively and etc. There are lots of answers about sound quality but no explanation to the reason. I'm searching for the reason for this quality drop. Can someone please explain; why does a higher watt speaker create sound quality issues.
Please your answers will quench my thirst for knowledge also consider buying (new) speakers.
I have combed the internet without an answer for the below and hope someone can give me an answer.
I am making a chip amplifier, LM6886 68W per channel, powered by a toroidal 28V 250W. The speakers I have are rated 8 Ohm, 250W, 8 Inch full range (+tweeter). My intension is to listen to (high) quality well balanced music at low sound levels, without distortion.
I have read - if the speakers have a much higher wattage that the amplifier then there will be clipping, sound will be distorted, sound will not be lively and etc. There are lots of answers about sound quality but no explanation to the reason. I'm searching for the reason for this quality drop. Can someone please explain; why does a higher watt speaker create sound quality issues.
Please your answers will quench my thirst for knowledge also consider buying (new) speakers.
It doesn't necessarily. Maybe they were saying that if you try to go to 250W then you'll see the limits of the amp? Of course 250W is a lot, and nothing says you have to try.
Chances are there is no problem. If you want opinions, tell us the sensitivity of the speaker.
Chances are there is no problem. If you want opinions, tell us the sensitivity of the speaker.
It will take some time and experience to understand the realities of the audio business hype.
1. It is possible to make a 68W rms chip amp, but that is a limit you should stay away from. It is only possible if you use a huge heat sink and rarely if ever drive the amp that hard and play music where the average power is a fraction of 68W. It is reasonable for class-D amplifier chips but LM6886 is not a class-D amp.
2. The 250W rating of an 8-inch speaker are most certainly exaggerated, probably by a factor of about 4x. Do an internet search for "peak music power". That kind of power from an 8" speaker would require inches of cone movement at 30Hz, but many such small speakers simply dissipate more power as heat because they are relatively inefficient. The assumption that high power speakers are less efficient and therefore more likely to require you to clip your amplifier is not true. More often, high power speakers survive because less of that power is wasted in heat. High efficiency speakers are typically at least 15 inches.
3. The clipping rational is used to sell large amplifiers but the reality is that high power amplifiers are a hazard that can destroy, especially your tweeter/horns in an instant. A rational person would choose a speaker that can survive whatever power the amp is capable of, but many enthusiasts love to rebuild or replace their equipment constantly. Assuming your speakers have a typical sensitivity of at least 90dB/watt@1 meter, you will rarely need more than 10 Watts unless you are fond of deafening sound levels. Class-A amplifiers and vacuum tube amps are rarely more than 20Watts.
1. It is possible to make a 68W rms chip amp, but that is a limit you should stay away from. It is only possible if you use a huge heat sink and rarely if ever drive the amp that hard and play music where the average power is a fraction of 68W. It is reasonable for class-D amplifier chips but LM6886 is not a class-D amp.
2. The 250W rating of an 8-inch speaker are most certainly exaggerated, probably by a factor of about 4x. Do an internet search for "peak music power". That kind of power from an 8" speaker would require inches of cone movement at 30Hz, but many such small speakers simply dissipate more power as heat because they are relatively inefficient. The assumption that high power speakers are less efficient and therefore more likely to require you to clip your amplifier is not true. More often, high power speakers survive because less of that power is wasted in heat. High efficiency speakers are typically at least 15 inches.
3. The clipping rational is used to sell large amplifiers but the reality is that high power amplifiers are a hazard that can destroy, especially your tweeter/horns in an instant. A rational person would choose a speaker that can survive whatever power the amp is capable of, but many enthusiasts love to rebuild or replace their equipment constantly. Assuming your speakers have a typical sensitivity of at least 90dB/watt@1 meter, you will rarely need more than 10 Watts unless you are fond of deafening sound levels. Class-A amplifiers and vacuum tube amps are rarely more than 20Watts.
Last edited:
Greets!
Example for reference only! Please get familiar with this orchestra acoustic power chart and this acoustic power tutorial with calculators and once you've figured it out you'll find that if your speaker system is peaking @ 92 dBSPL/63 Hz/1.0 W (1% eff), then it needs 1 kW! short term power (dynamic headroom) to 'cleanly' (no distortion/'clipping') hit 122 dB/250-500 Hz.
I use this prosound example to highlight just how important it is to have the right amount of speaker eff/average peak power rating and the necessary (reserve) amp power handling rating (AKA 'music power', 'crest factor') for the intended app.
edit: a more familiar looking frequency response plot of the acoustic power plot.
Last edited:
Lots of stuff to unpack here. The amp power is the absolute maximum if built to the specifications needed to get that much power out of it. Most likely your build will max out with a lower maximum output power level. Driven beyond that max value you are in clipping which means the amp hit the wall power wise and is now putting out DC more or less. It’s this clipping that destroys speakers. You can fry a 500 watt speaker with 100 clipped watts, maybe even less. Speakers convert energy into movement making sound, if the amp clips it sends DC to the speaker which goes to an extreme and sits there and heats up because it is not converting that power into sound. Your amp may have enough power to reach the sound level you are looking for before it clips especially if you have relatively efficient speakers. The other 2ay you can run into trouble is having a powerful amp and soeakers that can take that much power when you turn it up and you beat the snot out of the speaker. Your amp didnt clip but you overdrive the speaker. It’s important to remember that just because you have a 100 watt amp it doesn’t mean that you are using 100 watts all the time, most of the time at normal volume levels you will be using much less than 10 watts, more likely 1. If you turn it up and it sounds like crap turn it down, it’s really that simple.
No, you can't.
What kills speakers is the long-term (relative to the thermal time constants) average power being delivered. With reasonably dynamic program material, a 100w amplifier being driven 6dB into clipping delivers slightly less power than a 400w amplifier power at the onset of clipping.
This is obvious: the 400w amplifier is, for brief periods, putting out more power than the clipping 100w amplifier on account of the 400w amplifier having 2x the voltage swing available. On bass/snare drum hits (the loud bits), the 400w amplifier will put out a large signal, while the 100w amplifier will just clip that bit of waveform. The 100w amplifier, on account of briefly producing lots of harmonic distortion, will also show a small shift in the frequency spectrum, but this is negligible.
Conversely, if we were to play square waves through the amplifier, once it reaches clipping, it won't put out any more average power. No matter how far you push into clipping, it'll always be a square wave.
The "DC" the appears on the output of a clipping amplifier is only for very brief periods (ie, not really DC), so won't cause the heating you've described.
Idevotta, the 250w rating on your speaker is the maximum long-term power it can take while still surviving. It's like the top of the rev meter in a car - it's okay to go there occasionally, if you need to accelerate quickly or whatever, but if you find yourself redlining your engine often, it's time for something more capable. Similarly, with speakers, if you're routinely reaching or exceeding the long-term thermal rating, then it's time for some more capable speakers. Your 68w amplifier is unlikely to cause your speakers any distress, and, so long as it has enough power to reach the listening levels you desire, all is well.
Whoever told you this stuff was probably trying to warn you of the potential problems of an amplifier that's too small for the job. 68w will be fine for sensible domestic levels. Heck, with a reasonably-high-efficiency 8" full-range unit and your mention of "low sound levels", you could probably get away with 6.8w.
Chris
????? Internet has tons of answers,problem is most are contradictory or wrong.
You mean LM3886?
You mean 28+28VAC? So +/-35V rails under load.
Unqualified/Marketing "250W" probably means 70W RMS or so, so in fact quite matched to your a,mplifier.
And if they stand somewhat more power, fine.
In fact LM3886 can put out 50W into 8 ohm; not 68W; which is fine anyway.
It will sound fine, excellent amp.
Just get your chipamp from a reputed seller, not Alibaba or Ebay, LOTS of fakes around.
What Country are you in?
No, bunch of bollocks there.
IF original, your amp will sound fine.
Actually, it is real life experience that selling drivers to the masses that proves my hypothesis. I spent many years at a large speaker store and you are ten times more likely to blow a tweeter with an underpowered clipping amp as you are a much more powerful one. The thermal capacity is reached far below the rated power under clipping due to the overheating of a non moving driver. Much of the energy is transformed into motion if the driver is getting clipped power there is less motion and more heating. If it was all about the rated power you couldn’t blow a 100 watt driver with a twenty watt amp, you can.
speaker wattage is the max watts a speaker is designed to take (but you really dont want to operate any driver near this limit) , amp wattage is the max it can put out before it clips, there is no connection at all.
All else equal more wattage would never be bad thing, but maybe unnecessary.
Efficiency is relevant to amp distortion.
I would guess your 250W 8" full range driver is a pro sound driver, that typically means high efficiency which means less power and therefore less distortion from the amp for the same SPL compared to typical home hifi speakers with lower efficiency.
High efficiency speakers are in fact synonymous with more dynamic or 'lively' sound , so you have nothing to worry about.
All else equal more wattage would never be bad thing, but maybe unnecessary.
Efficiency is relevant to amp distortion.
I would guess your 250W 8" full range driver is a pro sound driver, that typically means high efficiency which means less power and therefore less distortion from the amp for the same SPL compared to typical home hifi speakers with lower efficiency.
High efficiency speakers are in fact synonymous with more dynamic or 'lively' sound , so you have nothing to worry about.
Last edited:
This is internet nonsense and has zero grounding in physics or reality, unless it somehow refers to trying to drive large speakers to their maximum using a tiny amplifier. Amplifiers clip, not speakers, and as has been suggested, excess amplifier power can be an advantage where headroom is concerned - I always strive for this in my systems. All things being equal, a 1000W amp playing at 100W is going to sound better than a 100W amp playing at 100W!
For example the midrange compression drivers in my PA are rated at 150W AES continuous, 1000W peak, and the HF 80W continuous, 320W peak, so absurdly powerful amplifiers are required just to get the voltage swing to properly reproduce transient events. The unclipped level you can attain is determined solely by the efficiency of your speakers (dB/W 1m), and I can virtually guarantee that most PA systems will play louder with a Watt than any hifi speakers. Even 10W will give you an spl of about 100dB with average speakers in a domestic setting - that's a level which would require ear protection in industry. If you like the way your speakers sound and they play loud enough for your tastes, then there is no requirement to change them based on numbers!
First it all depends on the sensivity of your speaker. A 90dB sensitive (90dB at 2.83V at 1m) will need 90w tk reach 100dB at 3m listening distance. But like said you need to count headroom to avoid amp distortion, especially with class D and chipamps. So if you keep 15dB headroom (normal standard) you will need 2846w for undistortioned sound at 100dB.
But in general you won't need 100dB, that is nightclub level, and i doubt you will listen at that volume. Most people listen to 70 to 90dB, and 90dB is much louder than you think. It's already loud enough to get cops at your door for nuisance in a lot of houses, esepcially when neighbours live near. But let's say you need 90dB undistortioned from that setup above, then you need 285w. For the same level with a 85dB sensitive speaker you need 900W.
the formula to calculate (if you're interested) is:
Equations used to calculate the data:
dBW = Lreq - Lsens + 20 * Log (D2/Dref) + HR
W = 10 to the power of (dBW / 10)
Where:
Lreq = required SPL at listener
Lsens = loudspeaker sensitivity (1W/1M)
D2 = loudspeaker-to-listener distance
Dref = reference distance
HR = desired amplifier headroom
dBW = ratio of power referenced to 1 watt
W = power required
Source: https://www.crownaudio.com/en/tools/calculators
Tube amps are different, as distortion of tube amps is more gradual, and harmonic. It's a part of the sound. Some low power class A amps also fall in that category (like most of the designs of Nelson Pass). There headroom is less important as there is always harmonic distortion, that is the warm sound many like...
But in general you won't need 100dB, that is nightclub level, and i doubt you will listen at that volume. Most people listen to 70 to 90dB, and 90dB is much louder than you think. It's already loud enough to get cops at your door for nuisance in a lot of houses, esepcially when neighbours live near. But let's say you need 90dB undistortioned from that setup above, then you need 285w. For the same level with a 85dB sensitive speaker you need 900W.
the formula to calculate (if you're interested) is:
Equations used to calculate the data:
dBW = Lreq - Lsens + 20 * Log (D2/Dref) + HR
W = 10 to the power of (dBW / 10)
Where:
Lreq = required SPL at listener
Lsens = loudspeaker sensitivity (1W/1M)
D2 = loudspeaker-to-listener distance
Dref = reference distance
HR = desired amplifier headroom
dBW = ratio of power referenced to 1 watt
W = power required
Source: https://www.crownaudio.com/en/tools/calculators
Tube amps are different, as distortion of tube amps is more gradual, and harmonic. It's a part of the sound. Some low power class A amps also fall in that category (like most of the designs of Nelson Pass). There headroom is less important as there is always harmonic distortion, that is the warm sound many like...
Oh my gosh I'm overwhelmed by the answers. Thank you for taking the time out to respond. If I were to summarise my understanding. If I try to drive the speakers to its max the amp will clip. When the amp clips it will send DC to my speakers and not really cause movement. This scenario is whats bad for the speakers. Tuning to a low volume from a high power amp will not result in clipping. I want to listen at low volumes, thus no clipping, no issues.
Many thanks once again, you all have been awesome.
Many thanks once again, you all have been awesome.
I spent 35 years repairing Pro Audio equipment. My workshop loudspeakers were a pair of 15" + 2" HF, rated at 100W. They often had 2000W amplifiers on test; BUT NEVER EVER CLIPPING.
All four drivers were still original after 35 years.
All four drivers were still original after 35 years.
Not true.
Clipped Audio is still AC, only square wave topped.
You send speaker, say, perfect squarewave clipped 40Hz (just to analyze a very low frequency) , cone is moving back and forth 80 times a second.
Do you call that "not moving"?
WHO wrote that nonsense for the first time and injected it into some Forum, where it spread since?
People copypaste ideas without analyzing.
Or thinking or anything, yeah that is unfortunately 99% of audio these days. 🙁
To respond to the topic starter.
Speakers don't "have" wattage.
The only thing that exist is a maximum power rating. Or in other words, above this power dissipation it's very likely the speaker will get (permanently) damaged.
Even that maximum power rating depends heavily on the context, although that's a much deeper debate.
That speakers with an higher maximum power rating, could get damaged by normal usage with a lower power (well voltage in most cases) is just the dumbest thing ever said.
Said by people who don't even understand basic high school physics and electronics.
There won't be sound quality issues. In fact, as long as you keep an eye on the maximum voltage coming out of an high power amp on a low power speaker, that also will be totally fine.
Unless you go into weird overly complicated bends and turns to explain things with strange peak powers and such, but that's not the point nor the question of the topic starter, let alone how one would normally use and play music.
Very simple question were some people need way to much jargon and text to explain the basics.
It was actually JBL which said something like that, BUT it applies to destroying tweeters with heavy clipping, the reason being that tweeter voice coils by themselves typically only stand 10% or 5% of total cabinet power.
They are usable because crossover removes most of a typical full range program and to boot "highs are low power" or if louder, come in short bursts (think cymbal crashes, acoustic guitar picking, etc.)
Now clipping produces harmonics so "highs appear which are not in the recording" and increase recording "density", compressing dynamic range so increasing previously "weak" highs contribution; both mechanisms easily surpass "expected" 5 or 10 % program delivered to Tweeters.
I guess in that era JBL also had a rash of blown tweeters repaired under warranty, definitely bad business, so they covered themselves.
Just as a personal example: I make PA, Bass and Keyboard amplifiers, all cabinets have Tweeters (duh!)
I use own made 4" cone tweeters
sporting "high power" 19mm voice coils, they stand some real 15/20W RMS continuous, so each one can be safely used with up to 150/200W worth of woofers .... and yet all power amps have built in distortion detecting limiters and all tweeters use protective car lamps in series, just in case a happy slapper or a Synth guy or a microphone squeal provides enough HF program anyway to blow the coils.
They are usable because crossover removes most of a typical full range program and to boot "highs are low power" or if louder, come in short bursts (think cymbal crashes, acoustic guitar picking, etc.)
Now clipping produces harmonics so "highs appear which are not in the recording" and increase recording "density", compressing dynamic range so increasing previously "weak" highs contribution; both mechanisms easily surpass "expected" 5 or 10 % program delivered to Tweeters.
I guess in that era JBL also had a rash of blown tweeters repaired under warranty, definitely bad business, so they covered themselves.
Just as a personal example: I make PA, Bass and Keyboard amplifiers, all cabinets have Tweeters (duh!)
I use own made 4" cone tweeters
sporting "high power" 19mm voice coils, they stand some real 15/20W RMS continuous, so each one can be safely used with up to 150/200W worth of woofers .... and yet all power amps have built in distortion detecting limiters and all tweeters use protective car lamps in series, just in case a happy slapper or a Synth guy or a microphone squeal provides enough HF program anyway to blow the coils.
TTBOMK +20 dB is/was the recording standard and presumably why it became the (HT) cinema standard, so when did (consumer only?) musical recordings lowered or is this just what's being done for ???? reason(s)?
That is what was used in pro audio when i worked there, where the equipment is calibrated that 0dBVU is -15dBFS (so 15dB headroom) And in studio's where i worked it was 0dBVU = -18dBFS mostly. It may have changed with that modern music is limited a lot, but actually i don't know why.
Let's get this straight:
DC is not the same as clipping. Clipping is where the waveform has a flat top, and contains more HF harmonics than the original signal. If we feed that signal to a driver, it may be stationary for a couple of milliseconds. 50ms, if you're reproducing a 20Hz tone and have pushed it so far into clip that it's a square wave. And then the cone will move again.
A 100w driver won't blow with a 20w amp. A 100w-rated 2-way speaker could have its tweeter damaged by a 20w amp, because tweeters have very little thermal power handling. They have tiny voice coils, little thermal mass in the motor, relatively little heatsinking, and no forced-air cooling. A typical dome tweeter might stand 5-10w, and, as you've noted, tweeters can be blown by amplifiers rated for less than the total power of the speaker cabinet.
You seem to be jumping between "speaker" (ie, complete cabinet, often with multiple drivers) and "driver". It's possible for the former to be damaged by a surprisingly small amplifier, due to assumptions made about the program material when the power characteristics of the cabinet were determined. Individual drivers, however, have set criteria that must be met in order for them to achieve a given power rating. They will not blow when fed from an amplifier that's rated for much less than the driver.
Chris