ahah
i've been searching this forum in vain,or with just too many possible threads to go through ...
So i didn't get any good information on speaker distortion!
What are the different type of distortions created/introduced by drivers/loudspeakers ?
I've read numbers like 2-4% in this forum,is it correct ?
Do we really settle for 2+% distortion from our speakers but get all freaked out by the 0.005% distortion from amplifiers?
omg!!!
So if it is correct to assume near 2% dist, the speakers would be tne weakest link of an audio system ?? thus where the greatest enhancement is possible?
Also, what are the means we do have to try and reduce distortion to more acceptable levels ? is there some type of drivers that offer less distortion with equal power? ..again it should be a compromise between all parameters ..
please share with me(us) your knowledge about this problem ..
thanks all for your time
i've been searching this forum in vain,or with just too many possible threads to go through ...
So i didn't get any good information on speaker distortion!
What are the different type of distortions created/introduced by drivers/loudspeakers ?
I've read numbers like 2-4% in this forum,is it correct ?
Do we really settle for 2+% distortion from our speakers but get all freaked out by the 0.005% distortion from amplifiers?
omg!!!
So if it is correct to assume near 2% dist, the speakers would be tne weakest link of an audio system ?? thus where the greatest enhancement is possible?
Also, what are the means we do have to try and reduce distortion to more acceptable levels ? is there some type of drivers that offer less distortion with equal power? ..again it should be a compromise between all parameters ..
please share with me(us) your knowledge about this problem ..
thanks all for your time
2% distortion from a driver is pretty good, but of course that depends on level. I've seen way over 20% THD (mostly 2nd and 3rd harmonic) from some drivers when driven at any kind of significant level. Basically once you start to make a speaker produce sound, it starts to distort. The harder you push it, the more it distorts (usually).
Why don't you go read Sigried Linkwitz's website? He talks quite a bit about various kinds of distortion. But basically, yes, the loudspeaker is one of the weakest links in the reproduction chain and the reason you can't find any good data is because it all looks so bad compared to THD numbers for amps, for example (which are fairly meaningless in any case).
Why don't you go read Sigried Linkwitz's website? He talks quite a bit about various kinds of distortion. But basically, yes, the loudspeaker is one of the weakest links in the reproduction chain and the reason you can't find any good data is because it all looks so bad compared to THD numbers for amps, for example (which are fairly meaningless in any case).
Yes, loudspeakers introduce orders of magnitude more distortion than preceding stages. There are ways to reduce distortion, but it's a compromise. Drivers made of extremely rigid materials, horns and electrostatic speakers offer lower distortion. Bringing the speaker inside the feedback loop can also lower distortion significantly.
Yes, JinMTVT, speakers are the weakest link. A speaker with 2% distortion (at a good listening level) could actually sound quite good. Many speakers produce over 10% distortion as you turn up the volume!
Distortion in a speaker is proportional to the electrical power, and to the cone movement. Reducing either of these tends to reduce distortion. Schemes to do this include horn loading, line arrays, and splitting the frequency spectrum across a larger number of drivers.
Horns reduce the required cone movement (and power) by providing something stiffer for the cone to push against- you'll hear horns reffered to as acoustic transformers. I'm still not up to speed on acoustic impedance- but the general idea is that you put a horn on a driver, and the driver doesn't have to move as much to create the same level of sound.
Line arrays and multi-driver setups; such as a 3 way, 4 way, or even 5 way systems allows each driver to reproduce a section of the audible range that it is most comfortable with, reducing the power sent to it and reducing cone excursion when you move a highpass point up in frequency. The tradeoff here is that they can be tough to design, and more large numbers of drivers are expensive.
Don't forget that the driver is not the only part of the speaker that can cause distortion, also be aware that there are different types of distortion! Cabinet wall resonances are bad, so are standing waves and reflected sound inside of cabinets.
Overall, I would have to say to look at both http://linkwitzlab.com and http://sound.westhost.com/articles.htm for more information and some articles.
I am currently trying to learn more about speaker distortion, and how I can build a super-cheap copy of the Orion, so I will be keeping an eye on this thread to see if more links show up.
Joe
Distortion in a speaker is proportional to the electrical power, and to the cone movement. Reducing either of these tends to reduce distortion. Schemes to do this include horn loading, line arrays, and splitting the frequency spectrum across a larger number of drivers.
Horns reduce the required cone movement (and power) by providing something stiffer for the cone to push against- you'll hear horns reffered to as acoustic transformers. I'm still not up to speed on acoustic impedance- but the general idea is that you put a horn on a driver, and the driver doesn't have to move as much to create the same level of sound.
Line arrays and multi-driver setups; such as a 3 way, 4 way, or even 5 way systems allows each driver to reproduce a section of the audible range that it is most comfortable with, reducing the power sent to it and reducing cone excursion when you move a highpass point up in frequency. The tradeoff here is that they can be tough to design, and more large numbers of drivers are expensive.
Don't forget that the driver is not the only part of the speaker that can cause distortion, also be aware that there are different types of distortion! Cabinet wall resonances are bad, so are standing waves and reflected sound inside of cabinets.
Overall, I would have to say to look at both http://linkwitzlab.com and http://sound.westhost.com/articles.htm for more information and some articles.
I am currently trying to learn more about speaker distortion, and how I can build a super-cheap copy of the Orion, so I will be keeping an eye on this thread to see if more links show up.
Joe
ok thanks all for the nice info
i'll re-re and re read linkwitz' site and look for distortion info!!!
line arrays ...
about using multiple drivers to do the same job as a single driver would
in what magnitude does using 2 drivers instead of 1 to get the same SPL at the same distance, reduces the distortion ?
and wich type of distortion does it affect ?
then 4? 8 ?
how does a 200$ Seas unit compare in distortion to a similar 20$ TB unit ? is their a significant difference in distortion ?
what about using 2 drivers in push/pull with inverted phase side by side a la "W" sub frame ? i read that it cancels mechanical distortion ? can we use this for higher freq. drivers?
does it work for more than 2 units?
i'll re-re and re read linkwitz' site and look for distortion info!!!
line arrays ...
about using multiple drivers to do the same job as a single driver would
in what magnitude does using 2 drivers instead of 1 to get the same SPL at the same distance, reduces the distortion ?
and wich type of distortion does it affect ?
then 4? 8 ?
how does a 200$ Seas unit compare in distortion to a similar 20$ TB unit ? is their a significant difference in distortion ?
what about using 2 drivers in push/pull with inverted phase side by side a la "W" sub frame ? i read that it cancels mechanical distortion ? can we use this for higher freq. drivers?
does it work for more than 2 units?
Remember that harmonic distortion < 1% is inaudible (ref Martin Colloms - High Performance Loudspeakers), so numbers like .001% distortion in amps is academic. Cone area, horn loading and a good magnetic system are the most effective way to keep distortion below the threshold of audibility.
www.gattiweb.com
www.gattiweb.com
Super-cheap? Wow, I want to hear about that.
Re speaker distortion, afaik you need to have cones always operating in pistonic region, avoiding cone breakup like plague. The reality is compromises.
Very nice site: http://www.lenardaudio.com/education/05_speakers.html
best of a bad job i reckon and we've just got to live with it to some degree
the way i see it you've got a compound effect of distortion from the microphone in recording to the speaker in playback
any distortion in one bit of kit is just going to get multiplied by the next thing in the chain and so yes the lower your amps THD the better
if all the kit has less than 0.01 THD before it meets the 2% of speakers you're still only going to be around 2% THD in total
stop worrying about what comes before and it all starts going horribly wrong
dave
the way i see it you've got a compound effect of distortion from the microphone in recording to the speaker in playback
any distortion in one bit of kit is just going to get multiplied by the next thing in the chain and so yes the lower your amps THD the better
if all the kit has less than 0.01 THD before it meets the 2% of speakers you're still only going to be around 2% THD in total
stop worrying about what comes before and it all starts going horribly wrong
dave
Try these linksfor in depth decsription of distortion:
http://www.klippel.de/pubs/default.asp
http://www.klippel.de/pubs/alma2001/website_alma-Dateien/frame.htm
Seas provides distortion measurements on their drivers:
http://www.seas.no/
http://www.klippel.de/pubs/default.asp
http://www.klippel.de/pubs/alma2001/website_alma-Dateien/frame.htm
Seas provides distortion measurements on their drivers:
http://www.seas.no/
There are considerable differences in distortion between a Seas Excel driver and a Tang Band, easisly a factor of 20.
The thing that is completely lacking from this thread is that loudspeaker distortion is predominantly lower order, whereas amplifier distortion can have very high order distortions products which can sound a lot nastier. THD is nearly meaningless as a figure of merit.
The thing that is completely lacking from this thread is that loudspeaker distortion is predominantly lower order, whereas amplifier distortion can have very high order distortions products which can sound a lot nastier. THD is nearly meaningless as a figure of merit.
I want to precise some things and make sure I also understand correctly.
Distorsion expressed as a single number is meaningless.
First, distorsion is frequency dependant and our perception of distorsion is also frequency dependant: What I mean is that 10% harmonic distorsion is more listenable at 50 Hz than at 500 Hz. Distorsion is not a number but a spectrum.
Also, distorsion is dependant of sound level. The more the level is increased, the more there is distorsion. So it should be expressed with a level attached. It is really easy to have less than 1% disto from any driver (except low frequencies) at really low level. Problem is when volume get's higher.
So if you have a speaker that produce 5% disto at 50Hz at 110 dB, I say Fine!
If you have 5% disto from 100 Hz to 10KHz at 80 dB, I say Ouch!
F
Distorsion expressed as a single number is meaningless.
First, distorsion is frequency dependant and our perception of distorsion is also frequency dependant: What I mean is that 10% harmonic distorsion is more listenable at 50 Hz than at 500 Hz. Distorsion is not a number but a spectrum.
Also, distorsion is dependant of sound level. The more the level is increased, the more there is distorsion. So it should be expressed with a level attached. It is really easy to have less than 1% disto from any driver (except low frequencies) at really low level. Problem is when volume get's higher.
So if you have a speaker that produce 5% disto at 50Hz at 110 dB, I say Fine!
If you have 5% disto from 100 Hz to 10KHz at 80 dB, I say Ouch!
F
Yes this is the sad truth...
If you want low distortion I recommend you to leave the so called "hifi" drivers and move on to pro sound drivers. JBL's and Beymas have pretty much the lowest distortion in the industry (generally below 1% @ quite high SPL's). The only down side is that the use of pro sound drivers is addicting - don't say I did not warn you!
Well engineered and powerful magnetic structures and shorting rings are some features to accomplish this. Horn-loading further reduces distortion as said before.
True, but while harmonic distortion in itself is pretty useless it nevertheless reveals the nonlinearity that give rise to IMD and other malign distortions. If you have a very linear device (be it amp, speaker or whatever) THD will be very low. If it isn't the device is nonlinear. It is as simple as that.
/Magnus
If you want low distortion I recommend you to leave the so called "hifi" drivers and move on to pro sound drivers. JBL's and Beymas have pretty much the lowest distortion in the industry (generally below 1% @ quite high SPL's). The only down side is that the use of pro sound drivers is addicting - don't say I did not warn you!
Well engineered and powerful magnetic structures and shorting rings are some features to accomplish this. Horn-loading further reduces distortion as said before.
True, but while harmonic distortion in itself is pretty useless it nevertheless reveals the nonlinearity that give rise to IMD and other malign distortions. If you have a very linear device (be it amp, speaker or whatever) THD will be very low. If it isn't the device is nonlinear. It is as simple as that.
/Magnus
Pro-drivers are an interesting area, in that they can sound really good (amplifier levels of THD and IMD) and then suddenly they're much worse than normal consumer audio drivers. This is especially true in horn loaded systems. The trick is to align the correct drivers so that you never go past this point of no return in the home environment (reasonably easy with everything except subs built around pro-drivers). Or so I've been told.
By comparison hi-fi drivers have a much smoother rise in distortion as you increase the SPL. This is compounded by the use of too small drivers for the intended SPL, not a mistake often made in the professional arena.
Carrying that thought on: Open baffle!
No air spring (so relying on the inherent linearity of the motor structure/cone suspension)
Higher excursion demands
My guess is that people only like them because of the way they interact with poor acoustic environments
By comparison hi-fi drivers have a much smoother rise in distortion as you increase the SPL. This is compounded by the use of too small drivers for the intended SPL, not a mistake often made in the professional arena.
Carrying that thought on: Open baffle!
No air spring (so relying on the inherent linearity of the motor structure/cone suspension)
Higher excursion demands
My guess is that people only like them because of the way they interact with poor acoustic environments
Before worrying about distortion in your speakers, are you going to ask the Tuba, Clarinet, Bassoon player about the distortion levels in their instruments? Or is it the distortions inherent in each which give them their own unique signatures. Do you get your ears checked on a regular basis, do they conform to the DIN standard ear'ole shape, thus giving perfect conditions for the reception of distortion free music, or are yours as different from mine as mine are from my neighbours, do we not all hear the same thing differently?
Don't take all this too seriously, I don't; listening to music should be fun not an excercise in minute deconstruction, which surely gets between your ears and the very thing you are supposed to be listening to.
Don't take all this too seriously, I don't; listening to music should be fun not an excercise in minute deconstruction, which surely gets between your ears and the very thing you are supposed to be listening to.
jo-briggs,
Yes off course, otherwise all instruments would sound the same (simple sinewave).
Has absolutely NOTHING to do with the distortion or fq response of the speakers. If we are going to reproduce the original sound (which we might like or not, that's personal) we need speakers with low distortion and flat response.
Not in a million years. If we are going to enjoy the music the way it is supposed to be we need speakers that are technically perfect. Distortion and funky response is what's get between your ears and the music.
Yes off course, otherwise all instruments would sound the same (simple sinewave).
Has absolutely NOTHING to do with the distortion or fq response of the speakers. If we are going to reproduce the original sound (which we might like or not, that's personal) we need speakers with low distortion and flat response.
Not in a million years. If we are going to enjoy the music the way it is supposed to be we need speakers that are technically perfect. Distortion and funky response is what's get between your ears and the music.
But if your ears have an uneven frequency response it matters not how perfect the reproduction, you won't hear that perfection. My point is that would you walk out of a Stones/YoYo Ma concert because the sound of the venue wasn't perfect? One can become obsessed by the minutiae and lose sight of the whole. Would you throw a wobbly if your partner moved an occasional table and upset the sound balance of your living room by introducing a 2Db dip in the room response at 11khz; do you go into the street to stop the children riding their bikes, prevail upon air traffic control to redirect flights to Heathrow Airport while you listen to the Messiah? If not, then perfection is unattainable.
How many times have I heard, in the HiFi Press over the last 40 years that "These speakers opened up a whole new vista/window/world .... " By now the view must be so wide that it encompasses the whole universe!
How many times have I heard, in the HiFi Press over the last 40 years that "These speakers opened up a whole new vista/window/world .... " By now the view must be so wide that it encompasses the whole universe!
It reminds me of the anecdote where an audiophile is walking and finds someone listening to music on a cassette tape boom box. The audiophile says (in disgust) "What are you listening to?" and is met with the reply, "Beethoven"
I agree that some people with too much money go much too far, and I laugh at the audio reviews that verge on pornography... but if you're into DIY audio, it's worthwhile to ask yourself "How can I improve this?"
Yes, it really is all about the music, and enjoying things- and no, nothing will ever be perfect. We should all take a moment to remember that. However- if you want to build something that's really a joy to listen to (whether you're watching a movie, playing cards with your friends, or sitting exactly in the sweet spot in your favorite chair)- there are more things to pay attention to than "Watts" and "frequency response"
Personally, I'm a nerd and an engineer. I enjoy thinking about the physics of sound reproduction, and the idea of doing it as right as possible appeals to me on a non-musical level. Also, minimizing distortion in your speakers allows them to sound better loud
I agree that some people with too much money go much too far, and I laugh at the audio reviews that verge on pornography... but if you're into DIY audio, it's worthwhile to ask yourself "How can I improve this?"
Yes, it really is all about the music, and enjoying things- and no, nothing will ever be perfect. We should all take a moment to remember that. However- if you want to build something that's really a joy to listen to (whether you're watching a movie, playing cards with your friends, or sitting exactly in the sweet spot in your favorite chair)- there are more things to pay attention to than "Watts" and "frequency response"
Personally, I'm a nerd and an engineer. I enjoy thinking about the physics of sound reproduction, and the idea of doing it as right as possible appeals to me on a non-musical level. Also, minimizing distortion in your speakers allows them to sound better loud
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.