Hello,
I have strange result when I use the enclosure including the diffraction 0deg file, the directivity result I get after the export is is shooting the SPL to +-175db instead of the +-90db using diffraction only
I can't understand what's wrong, I have redone the process multiple time and end up with the same result.
The Scanspeak 13M4535 isn't in the list I have added it myself :
Thanks
I have strange result when I use the enclosure including the diffraction 0deg file, the directivity result I get after the export is is shooting the SPL to +-175db instead of the +-90db using diffraction only
I can't understand what's wrong, I have redone the process multiple time and end up with the same result.
The Scanspeak 13M4535 isn't in the list I have added it myself :
Thanks
^I guess you have USPL of driver two times. Half space response text box should be empty or at least Full space checkbox unchecked while exporting baffle effect response from Diffraction tool, if exported response is used in Merger tool to convert near field to far field, or in Enclosure tool to convert simulated half space response to full space with baffle loss.
Generally, I'm not sure what you're trying to do and why, but it does not look recommended design procedure. If you're interested in how all this should be done with boxed speakers at home (without anechoic or other decent acoustic lab), please read:
1) VituixCAD Measurement Preparations.pdf
2) Chapter Checklist for designing a loudspeaker, beginning on page 4 in user manual.
3) Whole user manual.
4) Return to items 1 and 2 until this is clear
Generally, I'm not sure what you're trying to do and why, but it does not look recommended design procedure. If you're interested in how all this should be done with boxed speakers at home (without anechoic or other decent acoustic lab), please read:
1) VituixCAD Measurement Preparations.pdf
2) Chapter Checklist for designing a loudspeaker, beginning on page 4 in user manual.
3) Whole user manual.
4) Return to items 1 and 2 until this is clear
^I guess you have USPL of driver two times. Half space response text box should be empty or at least Full space checkbox unchecked while exporting baffle effect response from Diffraction tool, if exported response is used in Merger tool to convert near field to far field, or in Enclosure tool to convert simulated half space response to full space with baffle loss.
Generally, I'm not sure what you're trying to do and why, but it does not look recommended design procedure. If you're interested in how all this should be done with boxed speakers at home (without anechoic or other decent acoustic lab), please read:
1) VituixCAD Measurement Preparations.pdf
2) Chapter Checklist for designing a loudspeaker, beginning on page 4 in user manual.
3) Whole user manual.
4) Return to items 1 and 2 until this is clear
Thanks for the quick reply,
Basically I want to create a pair of baffles with 3 drivers, two I have already, I have not measured anything yet I am trying to simulate driver placement and crossover network and see if that get somewhere acceptable using traced factory Data.
If you want to know more about it there is the link :
Advices on First Crossover Design (VituixCAD2)
I have read 1 & 2 but I probably lack the background to understand everything talked about in these documents, I try to fill the knowledge holes by browsing Internet articles but that excellent software goes very deep, probably more than I can chew.
I will make sure that checkbox is empty and try again
Just reminding that for example these two steps are not required and therefore not listed in "Checklist for designing a loudspeaker". Also VituixCAD has many features which have been added by request or programming challenge, but they are still useless in practice no matter how many might use them and recommend to others.
Of course some knowledge about drivers, speaker constructions and acoustics is needed to select components without extra steps. Basically half space simulation with Enclosure tool for woofer selection and diffraction simulation for layout design (if enclosure will be simple box without decent bevels or rounding) are needed for simple 2...4-way before building, measurements and crossover simulation.
It is mainly the enclosure tool that is the most confusing for me, especially the directivity part, there is that "open cabinet impact response" but it is unclear for me what should I use.
I get around a BassBox6 software easily enough but then I can't use any of the result outside of it.
I encounter a difficulty in VituixCAD to understand if I should use the diffraction result to design the crossover or just the infinite baffle measurement from factory, I don't even speak of the directivity result that make no sense when I look at it.
But then you are right I may simply lack basic knowledge to succeed
I get around a BassBox6 software easily enough but then I can't use any of the result outside of it.
I encounter a difficulty in VituixCAD to understand if I should use the diffraction result to design the crossover or just the infinite baffle measurement from factory, I don't even speak of the directivity result that make no sense when I look at it.
But then you are right I may simply lack basic knowledge to succeed
Now i don't know that software but if it happen visual present any curves it works like a charm use "Tools" then "SPL Trace".
Damn! I didn't think to do a trace on that graph, I feel bad about it.
Thanks BYRTT I can use that
Directivity graph in Enclosure tool is added for very simplified theoretical studies only. It is not meant and needed for typical projects producing standard boxed speakers, and therefore it's not mentioned in "Checklist for designing a loudspeaker". For example I keep it continuously disabled (blank). No need to use any other program for that and then trace curves which is already stated as pointless.
Directivity of box speakers should be simulated with Diffraction tool only. That might be needed in standard project if directivity information in time-windowed far field measurements is not reliable and logical enough to produce off-axis responses to low frequencies below ~150 Hz. That is undocumented "guru" operation which I use occasionally to get more realistic power & DI responses at low frequencies.
I don't know. At least VCAD does not have direct mathematical response interpolation. Interpolation in Polarmap is just graphical with color gradient. Calculator allows some workarounds e.g. weighted average but that's not very intelligent approach.
One answer is that measure woofer's off-axis with 5 deg steps
Rev. 2.0.13.0 (2019-01-31)
* Directivity features removed from Enclosure tool
This makes easier to decide (without reading few pages from user manual and other support material) what is correct tool to simulate directivity of boxed speaker and open baffle. Feature stays available in rev 1.1.33 in case ideal/textbook directivity is needed for some experimental use.
* Directivity features removed from Enclosure tool
This makes easier to decide (without reading few pages from user manual and other support material) what is correct tool to simulate directivity of boxed speaker and open baffle. Feature stays available in rev 1.1.33 in case ideal/textbook directivity is needed for some experimental use.
I hope you mean to say that the directivity graph is gone but that we can still export the piston directivity for use in the crossover tool before we have measurements - like when we are laying out the baffle and need to look at vertical lobing or, as in my current design, evaluating CBT weighting coefficients based on the piston directivity.
^Everything related to directivity is gone from Enclosure tool. That has never been the most featured tool to simulate directivity of real life radiators due to missing diffraction effects to off-axis.
Directivity of circular and rectangular radiators can be calculated with Diffraction or Calculator tool. Proper tool to create off-axis responses for line arrays is Diffraction. Single driver in the middle area of array is simulated, off-axis responses exported and loaded to main program. Main program is able to create and reshape linear or curved line arrays (with Driver layout feature) - with or without crossover, max. 64 drivers at once from response data of single driver.
Directivity of plain radiator (without off-axis diffraction effects) can be created with Calculator tool for any source response, thought that result is not very useful because ideal piston directivity is no common.
Directivity of circular and rectangular radiators can be calculated with Diffraction or Calculator tool. Proper tool to create off-axis responses for line arrays is Diffraction. Single driver in the middle area of array is simulated, off-axis responses exported and loaded to main program. Main program is able to create and reshape linear or curved line arrays (with Driver layout feature) - with or without crossover, max. 64 drivers at once from response data of single driver.
Directivity of plain radiator (without off-axis diffraction effects) can be created with Calculator tool for any source response, thought that result is not very useful because ideal piston directivity is no common.
Tried that, no problems! More questions though
Have woofer column next to full range driver. Looks like I need to do separate diffraction sim for each column of drivers... (starting to get it...)
Need to offset driver on baffle. Moving either circle representing driver or symbol marking center seems equivalent. Is that right?
Should I include floor reflection? But I am simulating CBT ground plane array so I think its more accurate to sim double number of drivers in full space, vertically symmetrical around virtual floor level.
I have coarse traced directivity for the full range driver from its data sheet. Should I use merger tool to add diffraction to those curves?
Have woofer column next to full range driver. Looks like I need to do separate diffraction sim for each column of drivers... (starting to get it...)
Need to offset driver on baffle. Moving either circle representing driver or symbol marking center seems equivalent. Is that right?
Should I include floor reflection? But I am simulating CBT ground plane array so I think its more accurate to sim double number of drivers in full space, vertically symmetrical around virtual floor level.
I have coarse traced directivity for the full range driver from its data sheet. Should I use merger tool to add diffraction to those curves?
Last edited:
Hi Kimmosto
How do I do this?
I want to estimate total output from the array at its excursion limit, with weighting applied to the array. Once I go to the XO simulation, I can't see the excursion of the drivers and that makes sense because supposedly measurements are being used. The closest I can come is to use the enclosure tool with "crossover of driver" checked. But the excitation level of the driver is still set by the "Source V" box, not the XO simulation.
Except for the weighting, one could say well if (per box sim) I get x Db from one driver, then I get 6 db more from two and 12 db more from 4 etc.. That is what I have been doing and I know the weighting costs me roughly 3 db. Is there a better way?
It would be nice to dial up the excitation in the XO sim through an active gain buffer until I reached the target total SPL level and see the corresponding increase in power and excursion of the selected driver in the enclosure tool.
Thanks,
Jack
How do I do this?
I want to estimate total output from the array at its excursion limit, with weighting applied to the array. Once I go to the XO simulation, I can't see the excursion of the drivers and that makes sense because supposedly measurements are being used. The closest I can come is to use the enclosure tool with "crossover of driver" checked. But the excitation level of the driver is still set by the "Source V" box, not the XO simulation.
Except for the weighting, one could say well if (per box sim) I get x Db from one driver, then I get 6 db more from two and 12 db more from 4 etc.. That is what I have been doing and I know the weighting costs me roughly 3 db. Is there a better way?
It would be nice to dial up the excitation in the XO sim through an active gain buffer until I reached the target total SPL level and see the corresponding increase in power and excursion of the selected driver in the enclosure tool.
Thanks,
Jack
Y-middle and Y-center are not the same, but I suppose it is possible to create the same geometry with both parameters is some cases.
That's your decision. I don't add floor reflection to simulations. Other option is to design quite flat power response below estimated Schroeder frequency.
Traced or simulated half space response is combined with directivity with Diffraction tool. Response is loaded to Half space response text box and then Full space checked. That creates full space off-axis data which can be exported with 5...45 deg angle steps.
Last edited:
Unfortunately this is not yet possible for full speaker because crossover simulation in the main program does not handle driver's Sd or absolute source voltage (except for power dissipation).
Workaround is to simulate one driver group (with highest weight) in Enclosure tool with it's crossover.
I may have a bit bad attitude to excursion simulation because that's something we can't increase. Just try to make some workarounds if "adequate" volume excursion capacity is not an option. For example cut sub frequencies or compress with small sealed. In addition, useful excursion could be much different than mathematical Xmax so value of simulation only could be low.
