Using Soundeasy files in PCD and Mic calibration help

[JMB]

I took measurements in Soundeasy and went through the post processing of combining woofer responses and tweeter but have not yet had the chance to work the crossovers in Soundeasy and I will be travelling. I am nervous about bringing my Rockey so I would like to play around with the measurements in PCD. Does anyone know how to export the combined woofer responses (the tweeter response is less of an issue as it is unchanged) so that I can use it in PCD? It gets saved in the driver file but it seems to be saved as separate measurements that the program comes back and combines the way I told it to. Perhaps I am just too new with Soundeasy but everything else so far has gone smoothly.



I would also like to calibrate an uncalibrated mic with my calibrated one. I have measurements for both and after finding inconsistency in the nearfield but consistency in the farfield, I will likely build the chamber recommended at the ARTA site. Can anyone describe what software to use and how to find the difference between the two measurements and then combine this with the Calibration file from my calibrated mic?



Thanks,

Jay

[cuibono]

Hi Jay,

I'm not sure what PCD is. I've used Soundeasy for a long time, and find its data handling to be very frustrating. I'd say you're likely to find the same. But look at the 'export' features. There are only a couple, so it will be easy to try.

Don't bother trying to calibrate your mike yourself - I don't think it is possible. Send your mike to Cross·Spectrum - Microphone Measurement Services - they do a great job, as does Kim Girardin. It is worth the $55 if you're serious about knowing what is going on acoustically, and is totally a deal when you consider how much time you'll save not fussing with it.

[JMB]

I appreciate the feedback and I have tried the various export options without success in Soundeasy. Regarding mic calibration, it is really an exercise for me as I have a calibrated mic. I am practicing different methods and comparing results of my two ECM8000's, one of which Herb calibrated for me (at Cross Spectrum). Just looking for some additional direction, however.

Thanks,
Jay

[john k...]

There are several ways to export a driver's response with SE. Here is one way/

1) Open the driver file in question.

2) Go to the CAD screen and set up a simple circuit with an input node and the driver. It should look like this:



3) Open the Plotting screen and plot the response (in this example at node 0) INCLUDING DRIVER.

4) Go tot he Import/Export Data menu and select Export SPL /Phase at SPL dB (text File).

5) When the dialog screen opens save the file with a unique name.

The file will be saved with a TXT extension. So locate the file and change the extension to FRD and you should be all set.

[speaker dave]

If you have a calibrated mic and an uncalibrated mic, I would take a response curve of a good speaker out doors at 1 to 2 meters. Compare the two curves and see how much you would need to shift the uncalibrated measurement to get the best overlay between the two curves. (A lot of measuring software will let you level shift. How many dBs of shift give best overlay?) Some software lets you subtract one curve from another. This would ideally give a straight line difference (both mics same curve) with the shift of the straight line (from zero) equalling the calibration difference.

If you have discrepencies between near field and far field then rely on the far field. Take care when you switch mics to repeat exactly with the mics in the same exact position. If the curves vary in shape then lean towards the mid frequency level difference over the differences at the frequency extremes.

I have calibrated an omni condensor and True RTA to an Ivie IE35 that way and get good tracking between the two.

David S.

[JMB]

Thanks, John. I will do that but I would love to see Bohdan make it easier to export a combined curve without going through the extra steps.

Dave, I appreciate your response but I think that you really answered how to determine a difference in mic sensitivity. I have a mic calibration curve for one mic and measurements (nearfield, farfield, and even ground plane) for the calibrated mic and the uncalibrated mic with care being taken to measure at the same positions. I guess I am wondering what software to use to subtract the difference and then add to the calibrated curve. Unfortuneately, the number of samples is substantially larger for my calibration curve so just importing into excel won't work unless there is an excel work around.

Jay

[speaker dave]

Originally Posted by JMB

I guess I am wondering what software to use to subtract the difference and then add to the calibrated curve. Unfortuneately, the number of samples is substantially larger for my calibration curve so just importing into excel won't work unless there is an excel work around.

Jay

I see. Try this SPL Tools They have a program called SPL View that lets you reintrpolate a curve to a new point spacing. I have used their software to both trace a curve and also to change point spacing. It works with a little effort. Once you have your curves with the same frequency spacing you can subtract in most measurement packages or, as you say, in excel.

Most good mics have a fiarly smooth curve and I would be tempted to plot the two curves and subtract point by point graphically (by eye or with a homemade paper dB ruler) to derive a calibration curve of adequate resolution. I've done that before and it really isn't very hard.

David S.

[JMB]

Thanks, Dave, I had forgotten about that program. I used it years ago before I had measuring equipment. Jay

[JMB]

Well, that won't work as it has 17737 data points and the program can only do up to 10000. I can't break down the frequency range as it does not appear to be divided by decades or by octaves. I am trying to figure this out.

[john k...]

Originally Posted by JMB

Well, that won't work as it has 17737 data points and the program can only do up to 10000. I can't break down the frequency range as it does not appear to be divided by decades or by octaves. I am trying to figure this out.

As I said, that is one way. Another is to open the driver file, go to the HBT screen, click the buttons for amplitude and phase under Reference SPL and then export the SPL as before. There should only be 750 pooint in the data file. I don't know how you are getting 17737 point.