Since you want to use your measurements with Vituix, you'll have to understand how Vituix uses the data. The Vituix manual(s?) have some instructions, but it's not always easy to understand what's going on behind the scenes.
See links to measurement instructions in my signature below. That is quite universal and standard method to produce measurement data at home (without large space or anechoic). Compatible with any XO simulator which does not damage timing differences in measurement data with minimum phase extraction and has simple geometry calculation with driver's origin and mic position. For example LspCAD and VituixCAD.
Last edited:
^It is not needed to measure at 1 m. It's just decent compromise indoors at home to get at least 4.5 ms time window to far field measurements to get some resolution to midrange without reflections. Some diyers could have huge space, outdoors or some other anechoic to be able to measure at 2-2.5 m or with time window much longer than 4-5 ms.
This has nothing to do with -180/+180 deg wrapping of phase response. Excessive wrapping of phase response is removed by setting Reference time in FFT tool so that one of the drivers in the construction has very close to minimum phase response, and other drivers use the same Reference time to capture timing differences between the drivers. Usually Reference time is adjusted for tweeter. This is mentioned in both measurement instructions.
In my documentation ARTA just measures and saves pir files. Convert IR to FR tool in VituixCAD makes time windowing, FFT and timing offset for pir files, and Merger tool combines near field and far field measurements (or far field and far field).
This has nothing to do with -180/+180 deg wrapping of phase response. Excessive wrapping of phase response is removed by setting Reference time in FFT tool so that one of the drivers in the construction has very close to minimum phase response, and other drivers use the same Reference time to capture timing differences between the drivers. Usually Reference time is adjusted for tweeter. This is mentioned in both measurement instructions.
In my documentation ARTA just measures and saves pir files. Convert IR to FR tool in VituixCAD makes time windowing, FFT and timing offset for pir files, and Merger tool combines near field and far field measurements (or far field and far field).
Where is the FFT tool located? How to set the reference time?
I take a FR2 measurement in ARTA, then through the 'Recorder' I look at the delay time. Then I switch to IR. I set the delay from FR2. I take a measurement. Then, from the first peak of the impulse, I count the delay back and put the cursor. After the end of the impulse with a max peak, I put a marker so that I get the required number of samples.
If ARTA has the correct delay value, it must draw the correct graph. Here you can make the gate larger. So that there is a minimum number wrapping of phase. Are you doing another?
I read the instructions. I understood almost everything, but it takes time to try. If I clicked 'Get Ref. Time', how to undo it? Or how to take a step back? When I do IR in ARTA, is it not necessary for me to get the delay value first?
And again about the distance from the microphone to the speaker. I have 4 speakers + port. To make the correct measurements of the far field, it is need to put the microphone at a height of the middle of the tweeter at a distance of 1 m and from this point to measure all the speakers? And then in Vituix, enter the vertical offset for the three speakers except tweeter? Or measure each speaker at its height at a distance of 1 m and then do not enter their vertical offset from the tweeter?
And again about the distance from the microphone to the speaker. I have 4 speakers + port. To make the correct measurements of the far field, it is need to put the microphone at a height of the middle of the tweeter at a distance of 1 m and from this point to measure all the speakers? And then in Vituix, enter the vertical offset for the three speakers except tweeter? Or measure each speaker at its height at a distance of 1 m and then do not enter their vertical offset from the tweeter?
My signal circuit is computer> computer USB> external sound card> line-in output of the card> analogue input of the amplifier> amplifier> speaker. I checked the speaker and wires from the amplifier output - everything is correct. Maybe you have some files to check the polarity of the output of the amplifier and sound card?
The first peak may be negative when measuring multiband acoustics. This indicates that the tweeter is in antiphase to the woofer.
Also, the amplifier in its path can invert the signal in phase. This is checked simply: supplying a signal from the computer to the amplifier input. From the amplifier, the signal through the voltage divider goes to the input of the computer.
I started the IR measurement from the sound card, the amplifier turned off. The polarity is upside down. So this makes a microphone or sound card. At the same time I found out that the amplifier is not necessary for measuring IR using REW. Sound card amplification is enough.
Now the question is, how to find out who inverts the polarity - a card or a microphone?
Now the question is, how to find out who inverts the polarity - a card or a microphone?
Do two loopback measurements (soundcard output connected directly to soundcard input). One dual-channel measurement, one single channel measurement (DUT channel only, no compensation). The dual channel measurement should have normal polarity. If the single-channel measurement has negative polarity, you know it's not the microphone.
Sorry if this has been asked before, but is there a way in Arta to do "deferred" impulse response measurement ?
By this I mean is there some way to export the test signal generated by the "Impulse response measurement" function (swept sine, periodic noise etc) as a WAV or FLAC file instead of being played back through the sound card ?
And then at a later time manually trigger an impulse record without simultaneously generating a new test signal ?
The use case for this is quite simple - I would like to be able to accurately measure the frequency response of DACs which cannot be connected to a PC running ARTA.
Such as the analogue output of an iPhone, the headphone output of a TV set, other music playing devices with their own inbuilt DACs etc.
These devices can play back WAV or FLAC files so could play back the test signal generated by ARTA if there were some way to export it.
Of course absolute phase information can't be obtained since the impulse gathering is not being performed in full duplex so there is no precise time control between when the test signal is played and capture is started, however a frequency response is all I'm looking for.
Is this possible in ARTA at the moment and if not would it be difficult to add something like this ?
By this I mean is there some way to export the test signal generated by the "Impulse response measurement" function (swept sine, periodic noise etc) as a WAV or FLAC file instead of being played back through the sound card ?
And then at a later time manually trigger an impulse record without simultaneously generating a new test signal ?
The use case for this is quite simple - I would like to be able to accurately measure the frequency response of DACs which cannot be connected to a PC running ARTA.
Such as the analogue output of an iPhone, the headphone output of a TV set, other music playing devices with their own inbuilt DACs etc.
These devices can play back WAV or FLAC files so could play back the test signal generated by ARTA if there were some way to export it.
Of course absolute phase information can't be obtained since the impulse gathering is not being performed in full duplex so there is no precise time control between when the test signal is played and capture is started, however a frequency response is all I'm looking for.
Is this possible in ARTA at the moment and if not would it be difficult to add something like this ?
Last edited:
