amir always posts the sine wave graph first. i cannot figure out what it is intended to convey. they all look the same to me except for the vertical rise, but that doesnt seem to be the same as the volts in the following graphs.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
what does the sine wave show?
- Thread starter suttondesign
- Start date
When you see the sine wave you see it in either peak voltage or peak-to-peak voltage.
The RMS peak value is 0.707 or 70.7% of the peak voltage and so it is a lesser value than the peak value.
Last edited:
Simplified:
To calculate the power of a stereo amplifier you would use the power formula of Power=Vpeak^2/Rload
To calculate the power RMS you use Power RMS=[Vpeak(.0707)]^2/Rload where Rload is the impedance of the speaker.
Use order of operation = BEDMAS for calculations. (brackets, exponents, divide, multiply, add and then subtract.)
To calculate the power of a stereo amplifier you would use the power formula of Power=Vpeak^2/Rload
To calculate the power RMS you use Power RMS=[Vpeak(.0707)]^2/Rload where Rload is the impedance of the speaker.
Use order of operation = BEDMAS for calculations. (brackets, exponents, divide, multiply, add and then subtract.)
- Joined
- Jun 19, 2018
- Messages
- 6,652
- Likes
- 9,406
Severe distortion would be visible as an imperfect sine wave: flattened at the top and bottom for odd harmonics, asymmetrical for even harmonics.
Most of the gear tested here doesn't distort severely enough for this to be visible.
You do get a couple of exceptions to this rule though:
Here is the first view.
You already have the explanation for the RMS value differering from the sine wave view.
The FFT view shows you a spectrum analysis ranging from 20Hz to 20kHz and so you should only see a spike at 1kHZ as that is what the sine wave is. The extra peaks you see are harmonics or sometimes are rubbish energy that should not be in the circuitry.
You already have the explanation for the RMS value differering from the sine wave view.
The FFT view shows you a spectrum analysis ranging from 20Hz to 20kHz and so you should only see a spike at 1kHZ as that is what the sine wave is. The extra peaks you see are harmonics or sometimes are rubbish energy that should not be in the circuitry.
- Joined
- May 21, 2019
- Messages
- 4,036
- Likes
- 6,827
Two more examples when it matters:
I used to wonder how a swept sine wave (frequency of the tone increases during the test) could create all the information that it does.
So, I experimented.
Use the swept sine with mathematical analysis - using REW because I'm a math flunky and haven't written any code for 15 years.
And compare that to what a microphone in the room thinks goes on.
For an impulse and for a step - two things that don't look like a sine wave to me at all...
https://www.audiosciencereview.com/forum/index.php?threads/impulse-response.1765/
I'm convinced that "it works".
So, I experimented.
Use the swept sine with mathematical analysis - using REW because I'm a math flunky and haven't written any code for 15 years.
And compare that to what a microphone in the room thinks goes on.
For an impulse and for a step - two things that don't look like a sine wave to me at all...
https://www.audiosciencereview.com/forum/index.php?threads/impulse-response.1765/
I'm convinced that "it works".
- Joined
- May 21, 2019
- Messages
- 4,036
- Likes
- 6,827
@suttondesign One more interesting example: https://www.audiosciencereview.com/...asurements-of-monoprice-hybrid-tube-amp.7513/
MRC01
Major Contributor
One way to think about it is that a sine wave has a specific shape. If the input is a pure sine wave and the output takes on any other shape, the only way to change its shape is to add other sine waves of different frequencies and amplitudes. Those other waves were added by the device, which is distortion & noise. Put differently: every waveform of any shape can be created (or closely approximated) by adding together a bunch of sine waves having different frequencies & amplitudes.
The spectrum analysis is the Fourier transform of the output, which shows all the sine waves (frequencies & amplitudes) that if you add them all together it creates the output signal. If the input is a pure sine wave, for a perfect device (no noise or distortion) the Fourier transform will have a single spike at that frequency, and nothing else. Everything else is junk: distortion or noise.
Now the amount of junk in the output usually depends on the complexity of the input. The distortions can cascade additively: first you get harmonics or multiples of the input frequency, then you can get the differences between those frequencies as new frequencies (intermodulation), then each of those in turn gets harmonics, etc. The net effect is that the device does OK with a single sine wave, but it falls apart when given a more complex input. This is why Amir also does the multi-tone test.
The spectrum analysis is the Fourier transform of the output, which shows all the sine waves (frequencies & amplitudes) that if you add them all together it creates the output signal. If the input is a pure sine wave, for a perfect device (no noise or distortion) the Fourier transform will have a single spike at that frequency, and nothing else. Everything else is junk: distortion or noise.
Now the amount of junk in the output usually depends on the complexity of the input. The distortions can cascade additively: first you get harmonics or multiples of the input frequency, then you can get the differences between those frequencies as new frequencies (intermodulation), then each of those in turn gets harmonics, etc. The net effect is that the device does OK with a single sine wave, but it falls apart when given a more complex input. This is why Amir also does the multi-tone test.
- Joined
- Oct 25, 2019
- Messages
- 127
- Likes
- 133
This is a fantastic, simplified explanation. Thank you.
One more question: I don't see in the sine wave graph an indication of what frequency is being generated. I guess if you eyeball a full cycle, you can see the time interval is 1 millisecond, but that is not marked on the chart. is that right? Does Amir always use a 1 kHz signal?
MRC01
Major Contributor
That's about right. 1 kHz is commonly used, though the amount of distortion often does vary with frequency, and our hearing is most sensitive to distortion & noise in the midrange to treble, roughly the 3 octave range from 600 Hz to 4800 Hz.
For that, a frequency response curve is more appropriate. One can also use a square wave or impulse to get the same information.
My apologies, I'm having trouble understanding what you're trying to say.
Excuse me. I am also having trouble to tell what i want to learn
Do those different sound waves contain anything more than a sine wave? Like decay or attack you wrote in your previous message.
Which "different sound waves"?
A frequency response contains all the information from a time domain response. For minimum phase devices (99.99% of audio components), all of the time domain information is available from the frequency response magnitude alone. I *think* this is what you're asking?
Similar threads
