Hello
Anyone that could explain how to measure IMD and what levels are "OK"
As far as I know you feed the amp with two different freq´s and compare the fundamental with the IMD products. But how, do you sum the products on both sides, do you sum all harmonics?
What frequencies do you use, at what ratio?
Please help
Anyone that could explain how to measure IMD and what levels are "OK"
As far as I know you feed the amp with two different freq´s and compare the fundamental with the IMD products. But how, do you sum the products on both sides, do you sum all harmonics?
What frequencies do you use, at what ratio?
Please help
I never measured percentages of distortions, but I know that there are many standard tests available. Try to google for example CCIF DIN IMD.
I have no direct experience of that measurement, but there is some information in this file that may help you:
http://rane.com/pdf/ranenotes/Audio Specifications.pdf
http://rane.com/pdf/ranenotes/Audio Specifications.pdf
Tubelab has posted several different IMD measurements on his website and probably in some of his posts too. George loves his software package, he says that it makes things like fft analysis and imd measurement really easy.
isaacc7 said:Tubelab has posted several different IMD measurements on his website and probably in some of his posts too. George loves his software package, he says that it makes things like fft analysis and imd measurement really easy.
Not to threadjack, but it would be great if someone who has an easy FFT/distortion analysis setup could describe the basics in semi-baby steps..........
What soundcard(s)
What software
buffers req'd, etc.
Connect here, look at this, etc...
That would be really cool.
Anyone that could explain how to measure IMD and what levels are "OK"
IMD is measured using 2 frequencies, they can be equal level typically 19kHz and 20kHz or unequal level with one high say 60Hz and one 7KHz with higher level.
The non-linearity of the amplifier produce different products that are the sum or difference between the harmonics of the 2 signals.
2nd order non-linearity produce sum and difference of the 2 original signals, for 19kHz and 20KHz you get 1kHz and 39kHz. 3rd order non-linearities give products that are very close to the original signals, example 2 times 19KHz - 20KHz is 18kHz and 2 times 20kHz - 19 kHz is 21 kHz meaning that these products will be found just outside of the 2 original signals.
The total level of IMD is the RMS value of all products but I think it is of more interest to look at the level of the 1kHz difference tone and the level of the 3rd order IM products, which value is good enough it is for yourself to judge, have a look at Stereophile tests of some amplifiers then you can get some idea of what you should expect.
The highest level that is used for IMD tests are normally where each signal would give an output power of 6dB below clipping, then the amplifier will just clip when the 2 signal voltages add together so the peak power will be the same as the clipping level with one signal.
I measure IMD using a program called audiotester and a PC with external soundcard it gives result like this
Attachments
Thanks guys, I'll do some measurements and I might have some more questions later.
The thing is that I have a breadboard up and running and the sound doesn't really meet up to the reasonable harmonic distortion I measure. There is a harshness in the upper regions that I am chasing. Can be many things but I thought IMD could be one.
/Olof
The thing is that I have a breadboard up and running and the sound doesn't really meet up to the reasonable harmonic distortion I measure. There is a harshness in the upper regions that I am chasing. Can be many things but I thought IMD could be one.
/Olof
- Status
- Not open for further replies.