Vs: 70-te generaatori ja harmooniliste teema
Muidugi võiks nüüd korraliku kalibreeritud helikaardiga üle mõõta. Ja mõõtemääramatust võiks ka ligukaudselt hinnata. Generaator + väljundis olev lisafilter peaks siis tagama 100X (40dB) väiksema moonutuse, kui mõõdame võimendil. Lisandub veel mV-ga mõõtmise mõõtemääramatus, arvutatakse vähemalt 3 mõõtmise keskmine jne.
The accuracy of tuning is critical - a 40dB deep notch will show the distortion as 1%, even though it may be much less. A 60dB notch reduces this to 0.1% and so on. For a 100dB notch, you will need all components accurate to within 10ppm (parts per million), or 0.001%. Even a small temperature change can send the meter needle (or oscilloscope trace) straight off the scale. I know this, because it happens every time I try to measure very low distortion levels, and I can't even get to 0.001% on my meter because my oscillator has more distortion than that.
The ideal measuring meter is an oscilloscope, but a millivoltmeter may be used. Without the oscilloscope you will be unable to see the 'quality' of the distortion components, but use of an amplifier will allow you to listen to the residual - make sure that you have a limiter circuit on the monitoring amp, or a slight bump of the oscillator frequency control will blow your head off!
The final measurement will include the distortion from the audio oscillator, and it is likely that this will be greater than that of many amps. It is not really possible to tell you how you can subtract this from the measured distortion, since the distortion waveform has a huge influence over the result. The distortion waveform is very important - a low average level spiky waveform (typical of crossover distortion) will sound much worse than an apparently higher level of 'clean' 3rd harmonic distortion from a well designed push-pull amplifier stage.
Most sine oscillators will have distortion figures that are well within the measurement range of the distortion analyser, so expecting to be able to measure much below around 0.02% THD is generally unrealistic.
Kõige suurem kitsaskoht on väga väikeste moonutustega generaator ja 50Hz mürad.
Muidugi võiks nüüd korraliku kalibreeritud helikaardiga üle mõõta. Ja mõõtemääramatust võiks ka ligukaudselt hinnata. Generaator + väljundis olev lisafilter peaks siis tagama 100X (40dB) väiksema moonutuse, kui mõõdame võimendil. Lisandub veel mV-ga mõõtmise mõõtemääramatus, arvutatakse vähemalt 3 mõõtmise keskmine jne.
The accuracy of tuning is critical - a 40dB deep notch will show the distortion as 1%, even though it may be much less. A 60dB notch reduces this to 0.1% and so on. For a 100dB notch, you will need all components accurate to within 10ppm (parts per million), or 0.001%. Even a small temperature change can send the meter needle (or oscilloscope trace) straight off the scale. I know this, because it happens every time I try to measure very low distortion levels, and I can't even get to 0.001% on my meter because my oscillator has more distortion than that.
The ideal measuring meter is an oscilloscope, but a millivoltmeter may be used. Without the oscilloscope you will be unable to see the 'quality' of the distortion components, but use of an amplifier will allow you to listen to the residual - make sure that you have a limiter circuit on the monitoring amp, or a slight bump of the oscillator frequency control will blow your head off!
The final measurement will include the distortion from the audio oscillator, and it is likely that this will be greater than that of many amps. It is not really possible to tell you how you can subtract this from the measured distortion, since the distortion waveform has a huge influence over the result. The distortion waveform is very important - a low average level spiky waveform (typical of crossover distortion) will sound much worse than an apparently higher level of 'clean' 3rd harmonic distortion from a well designed push-pull amplifier stage.
Most sine oscillators will have distortion figures that are well within the measurement range of the distortion analyser, so expecting to be able to measure much below around 0.02% THD is generally unrealistic.
Kõige suurem kitsaskoht on väga väikeste moonutustega generaator ja 50Hz mürad.
Comment