Hi ! thanks a lot !
I tried to do what you say but it did not work. 😕
I am always getting just one sine signal.
However I will try again this evening.
Can you confirm please that the two sine option work also with the demo version ? 🙄
I mean, are you using the demo version as well ?
usually these demo versions have limited functionality
Thanks a lot again, gino
Last edited:
Hi ! I do not think so. This evening I will check better.
I cannot do now. I will revert with results.
Thanks a lot again, gino
Hi ! I tried. Did not work.
I will try again in the weekend.
However I am destroyed ...high distortion all over the place 🙁 (with the single tone test).
I do not understand.
Very very sad. I have to try again. It is just not possible.
Very good noise floor however.
maybe I am doing something wrong.
Thanks a lot again, gino
Most likely you are clipping the input. Managing levels to get good meaningful results takes experience so you can identify the source of problems.
Hi and thanks for the valuable advice.
But even with the 1kHz at 0dB the 2kHz harmonic was at about -60dB 🙁
I will post some graphs soon. I am very disappointed.
Thanks again, gino
Hello,
I spent some time researching this topic of noise and the “noise baseline” in frequency domain plots prior to purchasing an audio analyzer.
Be aware; for the FFT Plots or frequency domain plots there is much noise averaging and filtering to reduce the “noise floor”. The “noise baseline” in the FFT plots is not an accurate reflection of actual noise.
Take a look at the link to the technical note at audio precision about noise.
FFT Scaling for Noise - Audio PrecisionAudio Precision
These books were instructional to me.
Spectrum and Network Measurements
Robert A. Witte
Electronic Test Instruments: Analog and Digital Measurements
Robert A. Witte
DT
I spent some time researching this topic of noise and the “noise baseline” in frequency domain plots prior to purchasing an audio analyzer.
Be aware; for the FFT Plots or frequency domain plots there is much noise averaging and filtering to reduce the “noise floor”. The “noise baseline” in the FFT plots is not an accurate reflection of actual noise.
Take a look at the link to the technical note at audio precision about noise.
FFT Scaling for Noise - Audio PrecisionAudio Precision
These books were instructional to me.
Spectrum and Network Measurements
Robert A. Witte
Electronic Test Instruments: Analog and Digital Measurements
Robert A. Witte
DT
This.
Your spectrum with 0 input volume looks good, so if you haven't yet, try reducing to -10 or -20 or so dB and see if the graphs don't improve.
If you have ASIO options in setup -> Audio Devices make sure to use them and set them up properly in the Control Panel
Last edited:
Hi ! yes I tried. And the graphs improve actually.
But I understand that 0dB for the test signal is the standard.
Thanks a lot for the important advice. I will do.
Just to end my ramblings, can I go on thinking that thd+noise test is really significant ? and that the lower the thd+noise the better the interface quality ?
Thanks a lot again, gino
Hi and thanks a lot for the very helpful advice.
I guess you mean this is an extreme situation ? I understand
But it is in the extreme situations that the really good pieces show up. 😉
they say
Like this one here ... too expensive for me by the way.
http://prosound.ixbt.com/interfaces/motu/1248/1248-2444/thd.png
However I am really disappointed with some control panels. They are not user friendly at all.
When I have to select input and output not all the ins/outs appear.
Very weird control panels. For the experts for sure.
Thanks again, gino
Last edited:
Hi again.
Just another quick question.
Is the noise floor test (i.e. without loop between ins and outs) enough to assess the psu influence on a unit ?
I mean, if i get a very good noise floor can i be confident that the possible THD seen in the loop-back test comes all from the unit (i.e. dac and output stages) ?
This would help me a lot because if i get a low noise floor and a high thd this would mean that the problem is actually the audio interface and not the psu.
I am trying to understand some issues i have found measuring thd+noise.
Very low noise floor in the open loop test and not so low thd in the loop back test.
In my view Noise should come mainly from the psu and thd mostly from the audio circuit.
So if i get very low noise floor the psu is just fine.
Thanks a lot again, gino
Just another quick question.
Is the noise floor test (i.e. without loop between ins and outs) enough to assess the psu influence on a unit ?
I mean, if i get a very good noise floor can i be confident that the possible THD seen in the loop-back test comes all from the unit (i.e. dac and output stages) ?
This would help me a lot because if i get a low noise floor and a high thd this would mean that the problem is actually the audio interface and not the psu.
I am trying to understand some issues i have found measuring thd+noise.
Very low noise floor in the open loop test and not so low thd in the loop back test.
In my view Noise should come mainly from the psu and thd mostly from the audio circuit.
So if i get very low noise floor the psu is just fine.
Thanks a lot again, gino
Last edited:
There are many contributors to the noise floor of the system. Power supply is just one. Measuring the shorted input noise will get the voltage noise of the input circuitry. You can determine the effective input noise by shunting with a low value pot (1K) and raising the resistance until the noise floor increases by 3 dB approx. The resistance across the input can be converted to noise using this Thermal Noise Calculator The internal noise will be approx 1/2 of the effectiove noise.
The supply should have little influence unless the circuit is not well suited to the application. Open circuit input will get noise pickup from many possible sources and won't be helpful.
The supply should have little influence unless the circuit is not well suited to the application. Open circuit input will get noise pickup from many possible sources and won't be helpful.
Hi and thanks a lot for the very kind and valuable advice.
Let me rephrase my question.
As a general principle i believe that the lowest the thd+noise the better the audio equipment.
I do not believe in nice distortion i mean.
Quite often these usb audio interfaces comes with an external AC adapter, usually a wallwart smps. And very cheap ones too.
Focusing on these external psu my question is: if i get a very flat and very low noise floor can i be sure that the psu is just fine ?
or i need anyway to check the thd graph as well ?
My guess is that the psu can only contribute to the noise floor.
So if i get a low noise floor and a not so low thd the culprit is the interface and the only way to solve this is change interface 🙄
This is not a problem because i get very low noise floor even with inputs open, a more sever condition from what you say.
This is not a problem.
I think i got it. If the psu is decent quality and without obvious flaws one is like another.
This is very important to me. Because it makes me focusing what really counts. The interface.
Thanks a lot again, gino
Last edited:
- Status
- Not open for further replies.