ARTA

Hi,

<<<<<<< Chapter 1/4 of ARTA user manual:

Delay (ms) - enters the delay in the measured system, i.e. delay from loudspeaker to microphone (used only in a dual channel mode).

Chapter 4/4:

In the ARTA, the referent position is sample index 300. In single channel measurement mode there is no reference zero delay position, and zero delay position is not fixed - ARTA always puts at referent position the first sample whose amplitude is 20dB below the impulse response maximum. >>>>>>>>>>>



-----------------> OK but while being in dual channel mode I'm unable to get correct times/distance from mic to measured driver! Why? How to proceed?

Thanks
 
Latest manual 1.6.1 Page 60 Chapter 4.5

4.5 System Delay Estimation
The fixed delay in some measured system, for example the delay of sound wave from loudspeaker to microphone, or delay in digital processing systems, can be easily estimated (to accuracy of one sample), from the position of maximum of the cross-correlation function of the system excitation and measured response.
The estimation of delay can be obtained automatically by clicking the menu command 'Recorder->Crosscorreletion/delay estimation' in Fr2 mode, after initial measurement of frequency response. That command opens the dialog box 'Input Channels Cross-correlation' as shown in Fig. 4.14.
 
Hello,

If the impulse reponse has been acquired by the logsweep method and without the "center" option of Arta then you can use the mouse and move the arrow in order to place it at the beginning of the rise of the main pulse Then you can read the corresponding arrival time expressed in milliseconds near the bottom left corner ("cursor").

If the impulse response has been acquired by the logsweep method using the "center" option of Arta which allows to obtain additionally the distortion curves H2, H3 and H4, then:

after opening a .pir response impulse: using the mouse move the arrow in order to place it at the beginning of the rise of the main pulse and you can read the corresponding time expressed in milliseconds near the bottom left corner.

Remember this time t_read. Then the arrival time t_arrival you are looking for is the result of the formula:

t_arrival = t_read - 2721.354

t_arrival is expressed in milliseconds

Best regards from Paris, France

Jean-Michel Le Cléac'h
 
Hello,

I forgot to mention that the formula I gave is for an record done using a "sequence length" of 256 k and a sampling requency of 48kHz.

As I only use that large sequence length I cannot say if the formula is valid for other sequence lengths...

Best regards from Paris, France

Jean-Michel Le Cléac'h


Hello,

If the impulse response has been acquired by the logsweep method using the "center" option of Arta...

... the arrival time t_arrival you are looking for is the result of the formula:

t_arrival = t_read - 2721.354

t_arrival is expressed in milliseconds
 
Hello guys,

As I am able to measure in dual channel mode, jerome69's advise works fine! Finally I got correct results using this method. Thanks Jérome!

Jean-Michel: I am unable to get correct times using your two single-channel methods even with the same sampling (48k)and length(256k)...Are you sure about?
 
Hello,

I never use the 2 channels method (due to the added complexity). But as, in order to estimate the arrival time, a crosscorrelation seems to be performed between the ouput of the amplifier and the signal recorded by the microphone it is probably the best method as it will avoid artifacts due to the operating of the soundcard.

If a single channel is used we have to take in account some lattency between the input and the output of the soundcard and everyone will have to use a different formula...

Best regards from Paris, France

Jean-Michel Le Cléac'h






Hello guys,

As I am able to measure in dual channel mode, jerome69's advise works fine! Finally I got correct results using this method. Thanks Jérome!

Jean-Michel: I am unable to get correct times using your two single-channel methods even with the same sampling (48k)and length(256k)...Are you sure about?
 
ARTA 1.7

As every year just before xmas .... :)

ARTA 1.7 will be published on next sunday (probably).

New features in ARTA (extract):
- Mlssa export now without normalization to 1.
- New Mlssa import types (impulse and transient response)
New menu in Imp window:
- Record -> Signal Time Record opens recording dialog with added generator
- Generator in transient recorder generates sine burst and pulse signals
- Generator has new multitone signal for measuring TIM (DIM)
- Periodic square and triangle signals added to ARTA generators
- Copy bitmap string can be saved between opening of dialog
- FR and spectrum graphs with tick marks and more dense grid
- Downsampling (menu Edit->Downsample in Imp window) analysis of CSD in low frequency range.
- Better ISO3382 compatibility in estimation of reverberation time and noise reduction
- Implementation of ASIO INTERFACE to soundcards.
- more simple callibration procedure
- Added overlay plotting in Step response, Dft and IR Envelope windows
- Impedance curve as overlay in FR plot (Imp and FR in one plot)
- Thick line option enabled in all windows.

New in Steps:
- Muting of transients after sine cut-off (at cost of 50% larger measurement time)
- impedance curve overlay in FR magnitude plot
- ASIO interface.

New in LIMP:
- ASIO interface.

Regards
Heinrich
 
ARTA 1.7

... ARTA 1.7, manuals & compendium are online :)

Regards
Heinrich
 

Attachments

  • Xmas.jpg
    Xmas.jpg
    16.9 KB · Views: 825
ARTA H&T Manual 1.01

The ARTA Hardware & Tools Manual is now available on Ivos homepage (sorry, at the moment in german language only):

http://www.fesb.hr/~mateljan/arta/AppNotes/ARTA-H&T-D1.01.pdf

There are a lot of pictures, schematics and sketches in the manual. So many things should be self-explanatory, even without knowledge of german language :)

Regards
Heinrich
 
Last edited: