It's not for this task, it's just a new machine with sufficient speed for BEM.
So it should have more than ample speed for any plausible calculations on this task.
Best wishes
David
Sounds good to me, and that's certainly understandable - I'm fairly certain we all have more things we'd like to do than we are able, no matter our age.
I ask about the data format because I have built a tool that automates polar measurements at work, and I'd love to spit out the correct format for use in this tool. Is there an example of the proper data somewhere in this thread? If not, could one be posted so I can follow it?
I added the commonly available formats and mostly added some comments to explain the format, for the most part this means expecting a directory full of files which are named according to the angle they represent and that contain data in some known format. The supported formats being
- wav
- dbl
- ARTA PIR
- REW txt format
There is then the HolmImpulse export format which is described in pypolarmap/load.py at master * 3ll3d00d/pypolarmap * GitHub
This is a semi colon delimited format where the angles have been embedded in the header (aka the measurement names) as a convention
I have concluded that I am the wrong person to be key in moving this software along. So consistent with what I said at the beginning, I will make all the code that I have available so that perhaps someone else can move the numerical code along.
This code is in FORTRAN and compiles on the INtel compiler under Visual Studio.
I am not sure how to post it, any ideas? I'd prefer for someone to take this on and then I would work directly with them rather than just posting the coed in the public domain.
This code is in FORTRAN and compiles on the INtel compiler under Visual Studio.
I am not sure how to post it, any ideas? I'd prefer for someone to take this on and then I would work directly with them rather than just posting the coed in the public domain.
So, before you pass the torch, what did time tell?
I assume you would have mentioned if you had been successful, have you reached any conclusions?
Best wishes
David
I meant to comment on that.
The discussions were not very fruitful owing to the fact that he is well over 90 and I'm almost 70. We just did not seem to be able to communicate very well via E-mail.
I have concluded however that all two layer measurements are basically equivalent. What one is doing is measuring the radial velocity rather than the pressure, which will be highly robust to external reflections. The radiation is then expanded in terms of the radial velocity modes instead of the pressure. This can be done with a rotating source just as well as a stationary one. It just requires an intensity type of measurement using the pressure gradient rather than the pressure. A small modification would be required in the math, but nothing complicated.
The discussions were not very fruitful owing to the fact that he is well over 90 and I'm almost 70. We just did not seem to be able to communicate very well via E-mail.
I have concluded however that all two layer measurements are basically equivalent. What one is doing is measuring the radial velocity rather than the pressure, which will be highly robust to external reflections. The radiation is then expanded in terms of the radial velocity modes instead of the pressure. This can be done with a rotating source just as well as a stationary one. It just requires an intensity type of measurement using the pressure gradient rather than the pressure. A small modification would be required in the math, but nothing complicated.
Hi all,
For everyone wanting to understand the topic better, I just found this thesis which is very thorough and well written and a great source of information I think. ( though it might take quite some time t really understand it)
Cheers, Kees
Near-field acoustic holography with sound pressure and particle velocity measurements - Microflown
For everyone wanting to understand the topic better, I just found this thesis which is very thorough and well written and a great source of information I think. ( though it might take quite some time t really understand it)
Cheers, Kees
Near-field acoustic holography with sound pressure and particle velocity measurements - Microflown
Thanks Kessito
A highly theoretical coverage of the subject. Jay Maynard, Earl Williams and Eugen Scudryk were all contemporaries of mine at Penn State. This is how I learned these fundamental techniques. But for our purposes - simplicity - I don't think this kind of intense theory is really an advantage. Makes for a good thesis, but not good background for a simple and effective method.
A highly theoretical coverage of the subject. Jay Maynard, Earl Williams and Eugen Scudryk were all contemporaries of mine at Penn State. This is how I learned these fundamental techniques. But for our purposes - simplicity - I don't think this kind of intense theory is really an advantage. Makes for a good thesis, but not good background for a simple and effective method.
Still watching. I can do polar measurements and computation with a few programs. My SpeakerTesterPro does it well, and ARTA can do a good job of it. This PyPolar program offers options not available in either. If I can help in my limited time in between either projects or keeping the bills paid I'm willing to aid and abet!
Hello Scott.
I have known about this for quite a while. Being able to do the measurements would be the real ticket. I'll take a read of the two pages you put up links to.
I have known about this for quite a while. Being able to do the measurements would be the real ticket. I'll take a read of the two pages you put up links to.
I thought I'd just "throw that in" because of the recent posts on data format.
I actually emailed Ivo about including this (export file-type) in Arta, once he updated Arta for automated polars, ..about a year ago.
Don't know if he ever received the email or not, or if he had - what he thought about including that export capability.
I actually emailed Ivo about including this (export file-type) in Arta, once he updated Arta for automated polars, ..about a year ago.
Don't know if he ever received the email or not, or if he had - what he thought about including that export capability.
I am interested to see this, can you show pictures or provide more information?
Where do you work?
Unless the project is commercial-in-confidence of course.
Best wishes
David
I wrote a detailed rebuttal of this...and convinced myself you may have a point.
Still the same number of unknowns and variables in the equations, just a different set.
Seems plausible but not yet entirely clear to me.
Best wishes
David
The key word in my comment was "basically".
It can be proven that measurements on concentric spheres will not include any sources external to the spheres. It cannot be proven that such is the case with two microphones and a rotating source. However since all of the desired velocities are radial and it is highly unlikely that any room effects will also be so, there is a very strong rejection of anything external to these measurements. How "strong" would need TBD. But I would certainly first try the simple approach before rejecting it.
It can be proven that measurements on concentric spheres will not include any sources external to the spheres. It cannot be proven that such is the case with two microphones and a rotating source. However since all of the desired velocities are radial and it is highly unlikely that any room effects will also be so, there is a very strong rejection of anything external to these measurements. How "strong" would need TBD. But I would certainly first try the simple approach before rejecting it.
Yes, I started to write a rebuttal to prove the approaches were not the same, but then realised they looked quite similar in practise.
That's why I phrased my response as "you may have a point", better not to quibble.
Best wishes
David
I think I have not fully understood what Klippel NFS does. I use Clio FW12 and it does window time-gated measurement to subtract the room reflection from the direct sound. NFS does the same thing but by using the computer-controlled robot to do the same measurement in different angles to make CLF balloon 3D directivity index?
I hope it's on life support! It's still kind of alive over at Audio Science Review.
I've continued thinking about it and poke around into what I may need to learn to go it alone... but that's some heavy math and programing for someone starting out.
I'm curious if a simpler goal is possible, that of only doing "sound field separation" with one reflective surface, and only measuring polar response rather than the whole 3D sound field.
I've continued thinking about it and poke around into what I may need to learn to go it alone... but that's some heavy math and programing for someone starting out.
I'm curious if a simpler goal is possible, that of only doing "sound field separation" with one reflective surface, and only measuring polar response rather than the whole 3D sound field.