 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Natural Science Forum / Physics / Acoustics / October 2005MLS / SWEEP signals for measuring RT
Hello everybody, Regarding to measuring RT, what is the difference in using MLS or SWEEP signals as stimulus ?? Which one is better ? If the guess RT at 500Hz to be approx. 1 sec, then making use of 1.5 sec signal length is enough ? Longer signal length is better ? CSL MLS is a coded puls train. So the PC will "see"the coding and can avarage random noise out. But this is vulnarable to time variance, like from airco's, tempearature variations etc. Sweep is a bit more stable, also in the low freq range. Modern MLS usus often pink noise filtering to increase low frequency signal-noise ratio. both systems are very usable Thanks ! Could you advise a suitable signal length ? Is a longer signal length better ? CSL "bert stoltenborg" <bert@chaingang.nl> ???????:1130318093.539253.13200@g47g2000cwa.googlegroups.com... > MLS is a coded puls train. So the PC will "see"the coding and can > avarage random noise out. But this is vulnarable to time variance, like [quoted text clipped - 3 lines] > signal-noise ratio. > both systems are very usable mls (i have an old DRA mlssa system) can average. So the pulse train is repeated up to 16 times to improve S/N-ratio. It also has fiilters, you can get 65000 samples over 0 - 40.000 Hz, but also 65000 samples over 0 - 1000 Hz, to increase low freq resolution. With measuring RT you should however be careful with averaging because you can get anomalies. a 65000 sample range over say 20.000 Hz should be sufficient the used stimulus, mls or sweep, should be at least as long as the RT-time under investigation. By the way, there is a plug in for Adobe Audition (Cool edit pro) to do both measurements. you can find info on http://www.ramsete.com/aurora/ > the used stimulus, mls or sweep, should be at least as long as the > RT-time under investigation. > By the way, there is a plug in for Adobe Audition (Cool edit pro) to do > both measurements. > you can find info on http://www.ramsete.com/aurora/ Also, very important for RT measurement using MLS, do not overload the power amplifier or loudspeaker. Even a small amount of non-linearity will spoil the signal to noise ratio. If you find that the decay does not have enough range, try reducing the drive level as well as increasing it. Bert, thanks for the Aurora link - anyone used it and compared with "real" MLSSA?  SignatureTony Woolf My e-mail address has no hyphen - but please don't use it, reply to the group. Hey Tony, I bought a copy some time ago but never really used it. (I have the mlssa system and a laud system, and I'm a bit lazy :-)) I heared from a friend who used it that is is a bit cumbersome, but of course a lot cheaper than most dedicated programs, and better than something like ETF. Draw back is it cannot make impedance measurements, for a speaker guy a bit of a bummer :-). It can import mlssa files and do sweeps; mlssa can't. Maybe it's important to emphasize the importance of the soundcard quality. Some standard codex in PC's are not able to do good measurements. > I bought a copy some time ago but never really used it. (I have the > mlssa system and a laud system, and I'm a bit lazy :-)) [quoted text clipped - 3 lines] > Draw back is it cannot make impedance measurements, for a speaker guy a > bit of a bummer :-). Thanks. I need to know for when my MLSSA system gives up. I wonder when that will be? I do sometimes think it isn't good for credibility to turn up with a battered looking 15 year old PC running MSDOS 5.0, even if it is a lot more reliable than most modern portable PCs. SignatureTony Woolf Tony Woolf Acoustics 17 Canfield Gardens, London NW6 3JP Tel. 020 7624 2512 Fax 020 7372 3054 Mlssa's known failure is a blown up IC with impedance measurements. I replaced the chip by a socket and ordered a couple of these IC's :-) For credibility I wear an expensive watch (a fake from Thailand) and rely on my shiny smile :-)I keep a couple of spare ISA motherboards, with a desktop PC case and a tft-monitor it's pretty compact. And I like the software, maybe because I'm used to it, but I don't see many improvements in other software exept from the sweep option, FWTW. And it seems the new ISO standard point towards correlated methods for measurements on sound isolation, so it's quiet usable. > > I bought a copy some time ago but never really used it. (I have the > > mlssa system and a laud system, and I'm a bit lazy :-)) [quoted text clipped - 14 lines] > 17 Canfield Gardens, London NW6 3JP > Tel. 020 7624 2512 Fax 020 7372 3054 > mls (i have an old DRA mlssa system) can average. So the pulse train > is repeated up to 16 times to improve S/N-ratio. but i think either system can make synchronous stimulus/measurement and average. > It also has fiilters, you can get 65000 samples over 0 - 40.000 Hz, but > also 65000 samples over 0 - 1000 Hz, to increase low freq resolution. > With measuring RT you should however be careful with averaging because > you can get anomalies. > a 65000 sample range over say 20.000 Hz should be sufficient in article 1130589883.651282.207660@g44g2000cwa.googlegroups.com, bert stoltenborg at bert@chaingang.nl wrote on 10/29/2005 08:44: > the used stimulus, mls or sweep, should be at least as long as the > RT-time under investigation. > By the way, there is a plug in for Adobe Audition (Cool edit pro) to do > both measurements. > you can find info on http://www.ramsete.com/aurora/ one more thing is that MLS has a known problem that is pretty well understood that linear-swept frequency measurements do not have. a stationary non-linearity in the system being measured will result in little spikes that appear at constant locations in the IR and no amount of average it out. this was discussed a little in the JAES in the early 90s and i did a short analytic tutorial: http://www.dspguru.com/info/tutor/mls.htm . Signaturer b-j rbj@audioimagination.com "Imagination is more important than knowledge." Thanks, Robert! > Thanks, Robert! you're welcome FWIW. Signaturer b-j rbj@audioimagination.com "Imagination is more important than knowledge." |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2009 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.