Announcement

Collapse

Columbus Day Sale

Extended One More Day! Columbus Day Sale at Parts Express! 10/15/19 Click the Parts Express Logo at the top left of this page.
See more
See less

VituixCAD v2 released

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • kimmosto
    replied
    Here is one example. 3-way with 12" PA coaxial + 12" woofer. All far field measurements of higher frequencies are processed with minimum phase calculation for the right side simulation. Then far field is merged to near field measurements of lower frequencies which are processed with diffraction simulation to convert quasi far field. Difference between acoustic centers of coax tweeter and coax cone is defined with excess group delay method. That is simple and fast if separate sum measurement is not available (or shape is too bad for three measurement method). Difference result was 138 mm which is entered as negative value to Z coordinate of coax cone and woofer. Z of coax tweeter is zero.

    As we can see, axial/listening window to 0-20 deg is equal. Both methods are valid for that part. Minimum phase method gives the same magnitude and phase responses as with actual measured phase so difference detection with excess group delay method was okay. But problems start to rise step by step to off-axis, and 40 deg is already ruined at higher XO freq. The reason is that with simple geometry simulation tweeter has rotated to the same distance from mic/listener than acoustic center of cone when off-axis angle is 90 degrees. This error will even increase to rear sector with closed boxes because sign of distance differences swap. This is much better controlled with actual/measured phase.

    Bad result with minimum phase method is fully expected. I suppose experienced designer is able to ignore power response and directivity index simulation with this kind of driver geometry to avoid wrong choices in crossover design. But it would be unfortunate (though not my bad) if someone willingly throw useful information away due to limited measurement gear or method.

    Leave a comment:


  • dlr
    replied
    Originally posted by kimmosto View Post
    Here we have that kind of software. Have you considered testing or discussion about VituixCAD, or are you just defending your measurement gear, software and methods?

    We could also say that it is possible to design very good speakers without measured phase or without off-axis measurements or maybe without measurements at all. It is just designers choice how much accurate simulation information he/she requires or would like for see to get better forecast in each different case, or how much rely on luck or excellent acoustic basic design or experience or cloning from some other designs - or ears. Corrupting measured phase by default is one step towards blindness and ignorance.
    We have a basic disagreement with your last comment the focal point. There is nothing in using direct measurements that cannot be done equally as well with generated minimum-phase responses in the design process. Both have their uses and it is not blindness nor ignorance. My main system incorporates the Ultimate Equalizer that results in a linear phase system. I would not call the required alteration of the individual driver phase responses to be corruption.

    I will exit this discussion on that note. I can see it will not be constructive.

    dlr
    Last edited by dlr; 11-04-2018, 02:53 PM.

    Leave a comment:


  • kimmosto
    replied
    Originally posted by dlr View Post
    IMO the best option, if one wants accurate off-axis to use in optimizing a system, is to take multiple off-axis measurements and feed them into a system that can optimize the crossover taking these into account.
    Here we have that kind of software. Have you considered testing or discussion about VituixCAD, or are you just defending your measurement gear, software and methods?

    Originally posted by dlr View Post
    This package is what taught me that measured phase isn't a requirement for excellent crossover design.
    We could also say that it is possible to design very good speakers without measured phase or without off-axis measurements or maybe without measurements at all. It is just designer's choice how much accurate simulation information he/she requires or would like for see to get better forecast in each different case, or how much rely on luck or excellent acoustic basic design or experience or cloning from some other designs - or ears. Corrupting measured phase by default is one step towards blindness and ignorance.
    Last edited by kimmosto; 11-04-2018, 01:34 PM.

    Leave a comment:


  • dlr
    replied
    IMO the best option, if one wants accurate off-axis to use in optimizing a system, is to take multiple off-axis measurements and feed them into a system that can optimize the crossover taking these into account. In fact, while limited in number of measurements that can be imported, CALSOD, the DOS design software I started using in 1995 that had its origins, I believe, in the late 80's, had the ability to import multiple measurements, apply some basic diffraction influence (e.g. baffle step, etc.), set weighting schemes to each measurement, and optimize for that and other optional influences. This ran on IBM-PC 286 systems. CALSOD can still run in Windows 7 (I haven't tried Win10, it probably will). It's clunky, but I used it for final crossover work for a long time, even after having newer software.

    This package is what taught me that measured phase isn't a requirement for excellent crossover design.

    dlr

    Leave a comment:


  • dlr
    replied
    Originally posted by kimmosto View Post
    Single channel with MP extraction has chances when simulating axial only, but off-axis - especially with combination of surface mounted flat and deep/non-flat drivers such as horns is something between uncertain and mission impossible if geometry simulation is based on simple x,y,z coordinates without smarter (TL/horn/wave guide) simulation which could compensate variable length route of sound wave. Therefore I prefer and recommend dual channel gear and mode only. Makes my life much easier and keeps other options still available.
    I suppose I should have said that typical tweeters are 12db highpass rolloff. I have yet to work with a tweeter that was not, but I work only with ordinary type tweeters, not horn a large wave guided ones and of course with tweeters the measurements often go into the noise floor, so assuming as 12db rolloff for typical tweeters is a valid assumption for most cases.

    Off-axis it's best to measure to have the most accurate curves as well, of course, but the variation in response due to atypical drivers is, I suspect, not significant with respsect to the changes in excess-phase in the off-axis. For examining the off-axis due to the crossover, I don't see it as being a big factor. Much bigger is the variation due to diffraction differences, so the model would need to include that and would be dependent on how accurate that influence is to the off-axis.

    If the measurements are taken on-axis in situ of the baffle, then to be accurate, the on-axis would have to have the measurement normalized to remove the on-axis diffraction, essentially transforming them into quasi-anechoic responses, then adding the diffraction back in for the off-axis influence. If off-axis diffraction simulation is added to direct, on-axis measurements, than any and all off-axis simulations will be inaccurate. This sort of thing gets more convoluted if all factors are to taken into account in a single software package. Ultimately, this devolves to taking quasi-infinite baffle measurements, then modeling all aspects with these minimum-phase responses.

    dlr

    Leave a comment:


  • kimmosto
    replied
    Single channel with MP extraction has chances when simulating axial only, but off-axis - especially with combination of surface mounted flat and deep/non-flat drivers such as horns is something between uncertain and mission impossible if geometry simulation is based on simple x,y,z coordinates without smarter (TL/horn/wave guide) simulation which could compensate variable length route of sound wave. Therefore I prefer and recommend dual channel gear and mode only. Makes my life much easier and keeps other options still available.

    Leave a comment:


  • ernperkins
    replied
    Well then, I think we have agreed that measurements can be done with either single or dual channel setups. I have used both methods with equal success. But as always, the devil is in the details.

    Leave a comment:


  • kimmosto
    replied
    Originally posted by dlr View Post
    The tweeter tails are straightforward and easy. The highpass is a closed box, so below the measurement cutoff, it will still be a pure 12db...
    Not always. Also steeper than closed 12 dB/oct exist, though extending of e.g. 18 dB/oct slope is as easy. But extending is not mandatory with measured phase with tweeters because low frequencies are high passed by the crossover and possible problems due to S/N while measurement will not be visible in simulated total.

    VituixCAD is able to crop and extend responses at both ends with minimum phase calculation. User can adjust cropping frequencies and slopes, or leave that job to program at high frequencies, because constant manual slope is not always the best choice for all responses including off-axis. This feature is in Calculator tool so processing can be done to all off-axis responses at once. VituixCAD is also capable to help with defining difference in acoustic centers with three measurement method or directly by difference in excess group delay if measured phase is still available and stable (measured with time reference ie dual channel).
    Other than VituixCAD "total" users are also welcome to use those features in case measurement gear is not dual channel or stable/good enough for designing with measured phase, or cropping + extending with MP manipulation is needed otherwise.
    Last edited by kimmosto; 11-03-2018, 05:04 AM.

    Leave a comment:


  • jhollander
    replied
    Agreed. I've added a few super tweeters where the early phase wrap of the mid-tweeter gets close to affecting with the super tweeter phase. Certainly the slope of the mid tweeter phase changed by adjusting the HF mid tweeter tail. Probably not meaningful but it does make an argument for extending the high frequency measurement window imo.

    Leave a comment:


  • dlr
    replied
    Originally posted by jhollander View Post
    Unfortunately I need FR response below 400 Hz without room interactions for decent level matching to the mid or tweeter so I'm splicing and extracting min phase. I also have OCD where I feel obligated to trim the tweeter tails. civit if you need help determining acoustic offsets I'd be happy to walk you through it.
    The tweeter tails are straightforward and easy. The highpass is a closed box, so below the measurement cutoff, it will still be a pure 12db, so extending it won't introduce any consequential SPL difference from what might be measured. The top end lowpass is always a bit problematic with regard to knowing the actual SPL due to measurement system limitations and driver breakup, but it's is also inconsequential to design in that it is, in all cases I've ever seen, nowhere close to any crossover Fc. Since you generate phase from the final model, as long as you do use the three measurement system and then don't modify the driver SPL tails afterwards, you'll be able to get a nearly perfect match between measured and modeled summed response. The relative offset you need to determine is a function of the driver models. Once set,on a specific axis, you're good to go.

    I guess then that I'm also OCD if doing that is an indication of it!

    dlr

    Leave a comment:


  • jhollander
    replied
    Unfortunately I need FR response below 400 Hz without room interactions for decent level matching to the mid or tweeter so I'm splicing and extracting min phase. I also have OCD where I feel obligated to trim the tweeter tails. civit if you need help determining acoustic offsets I'd be happy to walk you through it.

    Leave a comment:


  • kimmosto
    replied
    Originally posted by Audiotorium View Post
    I am now using a closed loop system (a pair of Cross-Spectrum calibrated mics + FuzzMeasure + RME Babyface Pro interface with loop back)...
    I use measured phase (time of flight built-in) and for off-axis, I just measure.
    This is simple, fast and robust measurement method. No need to play with three measurement system, minimum phase extraction with manual slope settings and assume that all radiators (in the same speaker) are minimum phase and delay/phase behavior to all directions is equal.

    Most USB mics with separate sound card for output probably have that much timing stability that acoustic centers can be defined with three measurements. But some point timing might suddenly jump 1-2 ms without any change in settings or processor load. Some products (mics, sound cards, PCs, OSs) could be more stable than the others, but dual channel gear and measurement mode is easier option to ensure that random jumps or forced normalization of IR to e.g. 0 ms or 300 samples will not happen, and measurements are instantly valid for off-axis summing assuming that geometry is known. Time window could cause some non-minimum phase behavior at low frequencies but that is different story.

    Leave a comment:


  • civit
    replied
    Originally posted by Audiotorium View Post
    I am now using a closed loop system (a pair of Cross-Spectrum calibrated mics + FuzzMeasure + RME Babyface Pro interface with loop back) and for me, it works much better than Omnimic.

    I use measured phase (time of flight built-in) and for off-axis, I just measure.

    Again, this is what works for me, YMMV, but I would never go back to using Omnimic.
    Same, but with ARTA. I could never get reliable results extracting minimum phase. I compensate for most of the time of flight delay in VituixCAD to resolve the phase wrapping problems at HF.'

    I must have tried a hundred times, but I never got measurements from ARTA to work in the acoustic offset calculators out there. Doing a semi-2 channel measurement is much more intuitive for me.

    Leave a comment:


  • Audiotorium
    replied
    I am now using a closed loop system (a pair of Cross-Spectrum calibrated mics + FuzzMeasure + RME Babyface Pro interface with loop back) and for me, it works much better than Omnimic.

    I use measured phase (time of flight built-in) and for off-axis, I just measure.

    Again, this is what works for me, YMMV, but I would never go back to using Omnimic.

    Leave a comment:


  • dlr
    replied
    Originally posted by kimmosto View Post
    This is usually just useless theory,
    No, it's not. Context, that is what measurement system is in use and how the results will be used, in fact, makes it quite relevant.

    ...because measurement programs decide (without asking us) how they behave if reference channel is not available or not in use.
    Not necessarily. Again, it is dependent on the measurement system. I have used a number of measurement systems and all but one provided manual placement of the start time marker.

    So in practice semi-dual or dual channel connection and dual channel (timing reference) mode is usually mandatory, and therefore recommended in this thread.
    Again, I beg to differ, quite the contrary in my experience. I almost never use measured phase, only doing so when making simple, single-point measurements for initial modelling prior to a full-fledged, minimum-phase, full front hemisphere investigation as part of the design process.

    dlr

    Leave a comment:

Working...
X