I'm starting this thread as a branch from http://www.diyaudio.com/forums/showthread.php?t=150831 which was starting to devolve into a thread about measuring harmonic distortion. To try to keep things on topic and in the right place, I'm starting a discussion here. This way, if other people are looking for good information, it'll be easier to find.
Perhaps if some good information is gathered here we can put it up in the wiki!
Anyway, the backstory is that I've recently delved into the black art of loudspeaker measurement. I grabbed my friend's B&W 602 (S2) and took some measurements of it, and I plan to do more of that with other speakers I can get my hands on, including ones I'm building on my own. I'm fairly confident that my acoustic measurements are fairly solid at the moment, after having done a plethora of cross-referenced tests. Though, if I'm going to start piling up information on various speakers, I would really like to have a much bigger picture of what's going on, which includes Harmonic Distortion, CSD, etc.
Unfortunately, I don't know enough about these higher-level measurements to know what steps need to be taken in order to ensure accuracy.
Of course, if this information is very clear cut and already available on this forum, feel free to call me on it and blow this thread away. I find, however, that while this forum does have a great deal of information in it, it's often hidden deep within huge threads that have blown way out of context (like a discussion on HD measurements in a thread about B&W speakers), so I'm simply attempting at getting the right information in the right place
I will start off the discussion with my three main questions:
1) Does a microphone produce discernible levels of harmonic distortion that will interfere with speaker measurements?
2) How much do room reflections affect harmonic distortion measurements?
3) How much does ambient noise affect harmonic distortion measurements?
Perhaps if some good information is gathered here we can put it up in the wiki!
Anyway, the backstory is that I've recently delved into the black art of loudspeaker measurement. I grabbed my friend's B&W 602 (S2) and took some measurements of it, and I plan to do more of that with other speakers I can get my hands on, including ones I'm building on my own. I'm fairly confident that my acoustic measurements are fairly solid at the moment, after having done a plethora of cross-referenced tests. Though, if I'm going to start piling up information on various speakers, I would really like to have a much bigger picture of what's going on, which includes Harmonic Distortion, CSD, etc.
Unfortunately, I don't know enough about these higher-level measurements to know what steps need to be taken in order to ensure accuracy.
Of course, if this information is very clear cut and already available on this forum, feel free to call me on it and blow this thread away. I find, however, that while this forum does have a great deal of information in it, it's often hidden deep within huge threads that have blown way out of context (like a discussion on HD measurements in a thread about B&W speakers), so I'm simply attempting at getting the right information in the right place
I will start off the discussion with my three main questions:
1) Does a microphone produce discernible levels of harmonic distortion that will interfere with speaker measurements?
2) How much do room reflections affect harmonic distortion measurements?
3) How much does ambient noise affect harmonic distortion measurements?
They can do. But even if your microphone distorts badly at high sound levels, you can still make lower level distortion measurements. These will still show you most of the drivers distortion profile and most likely show you what a drivers suitable operating range could be - in other words areas you want to avoid if possible.
You'd have to have a really rubbish mic if you couldn't make distortion measurements of at least some usefulness.
They affect them in the same way they affect most other measurements, a reflection is a distortion in itself and will cloud the results in some way. Of course if you are familiar with the way your room interacts with the test signals you can learn that an anomaly at a certain frequency is the room and not the driver.
Noise is just like a reflection, something that distorts the data. However noises that would really affect the measurements are usually transient and you can simply take the measurement again. Try taking some measurements whilst whistling a constant tone, or do them with the hoover on vs the hoover off.
If you have lots of noises happening at random times, such as birds tweeting or something. You can always run a measurement ten times or so and average them, to help reduce their effect. ARTA's swept sine mode lets you do this easily.
Maybe the first thing to do is to take a measurement of your sound card to determine its capabilities. All you should have to do feed its line out into the line in and take a measurement.
Zaph also has useful information regarding HD measurement.
www.zaphaudio.com
- Status
- Not open for further replies.