Hi guys!
I am having an issue, I assume it might have to do with LspCAD 6, but I am not quite sure.
Basically, I do not get the same relationship between the resonant low end and the passband in LspCAD 6 as I do with other simulation tools. As especially LspCAD 6 is a fair bit more advanced than anything else, I think it is safe to assume this has to do with some settings I have not yet found. And as there are tons of settings that affects the damping, I am not paying that much attention to the low end sensitivity, so my focus is on the passband. I believe I have compensated correctly for measurement distance and half/free space (assuming point source). Here are some simulation results at 2.828VRMS:
LspCAD 6: 95,1dB
LspCAD 5.25: 96,8dB
VirtuixCAD: 97,5dB
SpeakerBuilder pro: 97,3dB
So, apparently there seems to be a bit of consensus around this being a 97dB sensitivity driver, but LspCAD 6 gives 89,1 in free space, which should equal 95,1dB, or around 2dB below the others, and that is probably not just an accuracy issue.
Someone might already hear some bells ringing, but here are some of the factors I suspect:
1: The T/S parameters are off.
2: These specific T/S parameters affects the way these simulators work.
3: There are settings I have overlooked.
4: There are errors in some of these simulators.
5: The ways these simulators work will always give results pointing in different directions.
6: The numbers are correct, and I have just forgotten or overlooked something else obvious.
The T/S parameters of the driver model are as follows:
Re: 7.0 ohms
Fs: 25 Hz
Qms: 5.22
Qes: 0.21
Qts: 0.20
Rms: 3.01 Ns/m
Mms: 100 g
Cms: 0.403 mm/N
Vas: 410 liters
Sd: 850 cm^2
BL: 22.9 N/A
Le: 0.8 mH
X-max: 12.5 mm
So can anyone solve this?
I am having an issue, I assume it might have to do with LspCAD 6, but I am not quite sure.
Basically, I do not get the same relationship between the resonant low end and the passband in LspCAD 6 as I do with other simulation tools. As especially LspCAD 6 is a fair bit more advanced than anything else, I think it is safe to assume this has to do with some settings I have not yet found. And as there are tons of settings that affects the damping, I am not paying that much attention to the low end sensitivity, so my focus is on the passband. I believe I have compensated correctly for measurement distance and half/free space (assuming point source). Here are some simulation results at 2.828VRMS:
LspCAD 6: 95,1dB
LspCAD 5.25: 96,8dB
VirtuixCAD: 97,5dB
SpeakerBuilder pro: 97,3dB
So, apparently there seems to be a bit of consensus around this being a 97dB sensitivity driver, but LspCAD 6 gives 89,1 in free space, which should equal 95,1dB, or around 2dB below the others, and that is probably not just an accuracy issue.
Someone might already hear some bells ringing, but here are some of the factors I suspect:
1: The T/S parameters are off.
2: These specific T/S parameters affects the way these simulators work.
3: There are settings I have overlooked.
4: There are errors in some of these simulators.
5: The ways these simulators work will always give results pointing in different directions.
6: The numbers are correct, and I have just forgotten or overlooked something else obvious.
The T/S parameters of the driver model are as follows:
Re: 7.0 ohms
Fs: 25 Hz
Qms: 5.22
Qes: 0.21
Qts: 0.20
Rms: 3.01 Ns/m
Mms: 100 g
Cms: 0.403 mm/N
Vas: 410 liters
Sd: 850 cm^2
BL: 22.9 N/A
Le: 0.8 mH
X-max: 12.5 mm
So can anyone solve this?
^I guess you have to contact Mr. IJ if you wish correct explanation. USPL to half space with those T/S-parameters is 97,3-97,5 dB/2.83V/1m depending on environment constants (speed of sound and air density). Basic simulators give that result almost without a single exception, but those do not have any statistical sleeve constants and formulas for reduced Sd or simulation of non-linearities. Common diy-person survives with basic simulation because this part of speaker simulator is used for driver selection and enclosure dimensioning (volume and possible ports) which is not any rocket science.
Thanks Kimmosto!
I e-mailed Ingemar a couple of days ago.
The environment constants should not only affect the passband, but also the low end.
What is "Statistical sleeve constants" and what do you mean by "formula for reduced Sd"? If you, by formula for reduced Sd, mean reduction in Sd due to parts of the suspension becoming stationary during high excursions, it should be SPL dependent (which it is not). The same goes for non linear behavior for other parts of the system. They are also all turned off (unless I have overlooked something).
BTW, you are the man behind VituixCAD? That is a really nice piece of work! Let me know if you want some ideas for expanding it!
I e-mailed Ingemar a couple of days ago.
The environment constants should not only affect the passband, but also the low end.
What is "Statistical sleeve constants" and what do you mean by "formula for reduced Sd"? If you, by formula for reduced Sd, mean reduction in Sd due to parts of the suspension becoming stationary during high excursions, it should be SPL dependent (which it is not). The same goes for non linear behavior for other parts of the system. They are also all turned off (unless I have overlooked something).
BTW, you are the man behind VituixCAD? That is a really nice piece of work! Let me know if you want some ideas for expanding it!
For example Fill % is combination of undocumented experimental constants and formulas, for some damping material in some location inside the box. Nothing exact or instructive for Bob the box builder.
Sd is typically measured with ruler or tape. Effective area is something else because cone is not ideal piston and could be function of frequency and linear (independent of excursion).
Basic simulators do not calculate for example internal radiation impedance which is function of internal dimensions. World is full of possibilities to make simulators in different ways but significance of more accurate science and availability of needed simulation parameters are not so clear. Personally I don't care too much about enclosure simulation. Most important feature is driver database browser.
Yes I am. I will read feature requests, and many of those are finally implemented due to programming challenge, though I resists almost everything in the beginning
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.