Dear Guys,
I wonder how to measure SNR without audio precision equippment or similar expensive measuring instruments?
All what I have is a precise RMS multimeter ( Fluke 83), Tektronix 220, HAMEG 203 and PeakTech DDS generator. I can do something like short inputs, set maximum outputlevel and measure RMS AC level on output. This is my noise floor but reduced in bandwith by multimeter/oscilloscope RMS.
So far so god. Now I would like measure the signal level ( on one single frequency, say 1KHz ) but I do absolutly not know which input level I should use.
Example:
My Phono stage has +40dB@1Khz. Should I increase the input level until "clipping" or get 0dBV ( 1Veff ) or get 0dBu ( 0.775Veff ) or... on output ???
I know that I have to measure the power of the whole audio spectrum and run a quantifier curve A or similar. But unfortunately I have no hight end audioanalyser at a hand. There may be a possibility to use a PC-Soundcard? But I do not know any software which does a SNR analysis.
Thanks for any tipp or hint!
best regards
Karsten
I wonder how to measure SNR without audio precision equippment or similar expensive measuring instruments?
All what I have is a precise RMS multimeter ( Fluke 83), Tektronix 220, HAMEG 203 and PeakTech DDS generator. I can do something like short inputs, set maximum outputlevel and measure RMS AC level on output. This is my noise floor but reduced in bandwith by multimeter/oscilloscope RMS.
So far so god. Now I would like measure the signal level ( on one single frequency, say 1KHz ) but I do absolutly not know which input level I should use.
Example:
My Phono stage has +40dB@1Khz. Should I increase the input level until "clipping" or get 0dBV ( 1Veff ) or get 0dBu ( 0.775Veff ) or... on output ???
I know that I have to measure the power of the whole audio spectrum and run a quantifier curve A or similar. But unfortunately I have no hight end audioanalyser at a hand. There may be a possibility to use a PC-Soundcard? But I do not know any software which does a SNR analysis.
Thanks for any tipp or hint!
best regards
Karsten
The missing puzzle piece in your case is a LNA front-end: it can be as basic as a 60dB low noise gain block, like a mike preamp for example. You could actually use a good preamp in fact.
You need to know the exact gain, and with this information and your existing test instruments, you can compute the noise floor and make the ratio between this floor and the nominal output level.
The weighting curves will require additional filtering, but I am pretty certain you can find applications for sound cards doing just that.
If you prefer hardware, there are examples of filters on the web using a quad opamp and a few passives.
Nothing insurmountable.
For more details, you should ask moderators to move your topic to the Equipment & Tools section: in the tube section, people tend to be more interested in audible qualities rather than in hard measurements, which is probably why nobody responded to your pretty easy query
You need to know the exact gain, and with this information and your existing test instruments, you can compute the noise floor and make the ratio between this floor and the nominal output level.
The weighting curves will require additional filtering, but I am pretty certain you can find applications for sound cards doing just that.
If you prefer hardware, there are examples of filters on the web using a quad opamp and a few passives.
Nothing insurmountable.
For more details, you should ask moderators to move your topic to the Equipment & Tools section: in the tube section, people tend to be more interested in audible qualities rather than in hard measurements, which is probably why nobody responded to your pretty easy query
SNR values for phono stages for moving-magnet cartridges are usually specified for an input signal of 5 mV RMS at 1 kHz, which would be equivalent to 500 mV RMS at the output in your case. SNR values are meaningless without a well-defined measuring bandwidth and phono stage SNR values can be misleading when measured with a shorted input, especially for solid-state phono stages with a bipolar input transistor.
I have used a simple 100 times op-amp amplifier and a simple A-weighting filter from Elektor for home audio noise measurements for many years. I usually don't measure SNRs, but rather compare noise levels to known noise sources, such as resistors. Using a soundcard and an audio editing programme is then enough to measure true RMS noise values, as the unknown gain of the soundcard drops out of the equation.
I have used a simple 100 times op-amp amplifier and a simple A-weighting filter from Elektor for home audio noise measurements for many years. I usually don't measure SNRs, but rather compare noise levels to known noise sources, such as resistors. Using a soundcard and an audio editing programme is then enough to measure true RMS noise values, as the unknown gain of the soundcard drops out of the equation.
A PC and sound card can be used as quite a good audio analyser.
Most sound cards exhibit a dynamic range approaching 90dB.
Frank
- Status
- Not open for further replies.