CY referenced this software in another thread. If I could remember the thread I would just send you to it, but since I can't here is the URL:
http://audio.rightmark.org/download.shtml
You can measure distortion using a microphone with less distortion than the device you are testing. You can measure distortion without destroying the device you are testing if you are careful.
The software is the easy part. Setting references, calibrating the test set up, and protecting your ears from the SPL are all the difficult parts.
Best to have a very good reason for measuring distortion before going to all the work to be able to measure distortion accurately.
Good designing and good building,
Mark
http://audio.rightmark.org/download.shtml
You can measure distortion using a microphone with less distortion than the device you are testing. You can measure distortion without destroying the device you are testing if you are careful.
The software is the easy part. Setting references, calibrating the test set up, and protecting your ears from the SPL are all the difficult parts.
Best to have a very good reason for measuring distortion before going to all the work to be able to measure distortion accurately.
Good designing and good building,
Mark
goldyrathore said:
Actually you could probably do alright with something like Cooledit and one of those cheap Panasonic capsules. Those mics are actually pretty good, but if you think about it you can get yourself in the far field in a big space where the mic (and your ears) only see a lower level. A simple frequency sweep could be used to roughly calibrate. There are still several sources of error but at highish levels you are looking at percents of distortion from most drivers.
Thought number two, has anyone ever tried this? Put the driver under test several feet away. Right next to the mic put a small good driver and adjust the amplitude and phase of the same frequency to null the fundamental. Now the mic sees no sound pressure at the fundamental and only the harmonics from the DUT are picked up.
Is there any better way to measure driver distortions then those common harmonics?
It seems to me that harmonic dist plots says very little about how a driver sounds and performs, some of my low harmonic dist drivers can sound quite harsh and ear shredding at some freq and with some music, something some drivers with higher harmonic dist can handle with ease! Listening to a driver with some rather complex music seems to be the most accurate test, but is there a way to measure this?
It seems to me that harmonic dist plots says very little about how a driver sounds and performs, some of my low harmonic dist drivers can sound quite harsh and ear shredding at some freq and with some music, something some drivers with higher harmonic dist can handle with ease! Listening to a driver with some rather complex music seems to be the most accurate test, but is there a way to measure this?
What would be the advantage? The fundamental can be removed with signal processing and the distortion of a good microphone is negligible.
Room EQ Wizard is free with a bunch of features including distortion
ARTA is free with limited features - considered state of the art for measuements
For distortion measurement you must standardize your method - both acoustically and electrically. Nearfield measurements are easy, but the microphone or sound card easily distorts (too high spl). 1-2 meter distace is practical, spl level should be 90-95dB(Z). You must standardize spl or voltage.
Some excellent home pages with home-made measurements
Zaph|Audio
Measurements and compare | HiFiCompass
www.audioexcite.com
http://rutcho.com/speaker_drivers/default.html
ARTA is free with limited features - considered state of the art for measuements
For distortion measurement you must standardize your method - both acoustically and electrically. Nearfield measurements are easy, but the microphone or sound card easily distorts (too high spl). 1-2 meter distace is practical, spl level should be 90-95dB(Z). You must standardize spl or voltage.
Some excellent home pages with home-made measurements
Zaph|Audio
Measurements and compare | HiFiCompass
www.audioexcite.com
http://rutcho.com/speaker_drivers/default.html
Last edited:
Who knows?
Perhaps a shot, and an analysis of its reproduction?
The reproduction of a few particular instruments, if it conforms to their characteristic of emission in the space, can deceive me (reality vs reproduction).
The fact that the vast majority of live music is emitted by equipments drives us away. We're missing something. That's a trap also for professionals, IMO.
Last Sunday I listened to some rehearsals: the sounds made by the drumset were surprisingly far from what you hear from a playback system.
During the concert instead (amplified sounds), this sensation disappeared.
Perhaps a shot, and an analysis of its reproduction?
The reproduction of a few particular instruments, if it conforms to their characteristic of emission in the space, can deceive me (reality vs reproduction).
The fact that the vast majority of live music is emitted by equipments drives us away. We're missing something. That's a trap also for professionals, IMO.
Last Sunday I listened to some rehearsals: the sounds made by the drumset were surprisingly far from what you hear from a playback system.
During the concert instead (amplified sounds), this sensation disappeared.
This is not so easy to quantify with a single % number.
Hello,
Total Harmonic Distortion and Intermodulation distortion are nonlinear distortions. Nonlinear distortion adds harmonics that are readily seen in a FFT plot and contributes to the THD percentage. Nonlinear distortion is readily measured and quantified with an easy single % distortion number. Low order 2nd and 3rd harmonic distortion and to a large extent IM distortion is masked by the program material, even if it measured to relatively high percentage levels it remains inaudible to our ears / brain.
Linear distortion is the primary distortion culprit that detracts from a speaker’s audible performance. Linear distortion adds no harmonics. Linear distortion will show up in the frequency Response as peaks, dips notches or completely missing frequencies. Missing low frequencies, high frequencies sound like a tin can with a string. Impedance peaks, reflections and time / phase alignment also contribute to a speaker’s quality performance.
Much of the distortion measurement you talking about is going on behind proprietary doors at the manufacturer’s labs. Or it is going on in public in the form of SpinOrama and subjective testing with trained listeners. This is not so easy to quantify with a single % number, it does not work that way.
Thanks DT
Hello,
Total Harmonic Distortion and Intermodulation distortion are nonlinear distortions. Nonlinear distortion adds harmonics that are readily seen in a FFT plot and contributes to the THD percentage. Nonlinear distortion is readily measured and quantified with an easy single % distortion number. Low order 2nd and 3rd harmonic distortion and to a large extent IM distortion is masked by the program material, even if it measured to relatively high percentage levels it remains inaudible to our ears / brain.
Linear distortion is the primary distortion culprit that detracts from a speaker’s audible performance. Linear distortion adds no harmonics. Linear distortion will show up in the frequency Response as peaks, dips notches or completely missing frequencies. Missing low frequencies, high frequencies sound like a tin can with a string. Impedance peaks, reflections and time / phase alignment also contribute to a speaker’s quality performance.
Much of the distortion measurement you talking about is going on behind proprietary doors at the manufacturer’s labs. Or it is going on in public in the form of SpinOrama and subjective testing with trained listeners. This is not so easy to quantify with a single % number, it does not work that way.
Thanks DT
!
Thank you.
Only swaps with multiple tones can feed my naive fantasies.
But I read that even intermodulation distortion is a "product" of harmonic distortion. So I will remain with my sense of unfinished reproduction.
It isn't, both are a product of a nonlinear transfer function. If the transfer function has a term corresponding to the square of the input, for example, the output will have 2nd harmonic distortion and 2nd order intermod components (at the sum and difference of the input frequencies if a two tone signal is used).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.