Home Theater Forum and Systems banner

1 - 9 of 9 Posts

·
Registered
Joined
·
7 Posts
Discussion Starter #1
Hi John, it would be possible to export the impulse as wav including compensation of calibration mic.

Thanks in advance!
 

·
REW Author
Joined
·
6,711 Posts
Afraid not, sorry. The calibration files are only used when generating frequency response plots - that makes it easy to remove or replace them afterwards if necessary.
 

·
Premium Member
Joined
·
1,562 Posts
John,
It is a nice convenience to be able to convert from one cal file to another in the REW measurement panel. The current setup provides for that. I have done that on occasion when I had the wrong mic orientation loaded when I took the measurement for instance. I like this choice of setup for most common uses of REW. With the current setup the SPL/Phase charts are accurate even at the extremes of the frequency range.

I do find my SR charts and a couple others are questionable however. If my limited understanding is correct (questionable), the SR chart in REW is not accurate as the impact of the cal files is not included. The SR can be misleading for those of who are optimizing FIR phase correction over the entire 20-20k range range.

Possibly there could be a separate utility or documented workaround that would allow the cal files impact to be added into a measurement IR if needed? A workaround to convert the mic spl/phase cal file into an IR and convert the SC SPL/phase cal to an IR may be enough? Those 2 IRs could then be combined back into a measurement IR using REW trace arithmetic (invert/multiply??). Once combined then the resulting IR would provide the accurate SR and all all others (so long as the measurement panel shows no cal files loaded).

I have been meaning to try to understand this better to see if I can figure out a workaround myself, but my attempts so far have just resulted in mass brain confusion/shutdown. There appears to some type of
limitation there. :(

The SC and mic cals should be min phase so a conversion from SPL/phase into IRs should be possible (Hilbert Transform?). [I'm also thinking the REW trace arithmetic does not now include the needed functions even if the 2 cal IRs are available?]

I'm fine working outside REW in a separate utility if necessary for some of the steps if it makes the most sense. I am challenged though to purchase Mathcad or other pricey software and understand it well enough to do it that way.

It's clear this capability also has utility for those exporting measurements for DRC or other EQ uses as well.

Any suggestions from you, or out other members here, for a workaround procedure?
 

·
REW Author
Joined
·
6,711 Posts
SR meaning step response I guess. Cal files are only applied where a frequency response is generated, so they are not included in time domain responses such as envelopes or step responses. While it would not be difficult to include them, I think it would generally not be a good idea as the typical effect is significant boost at the lowest and highest frequencies, producing time domain responses with correspondingly higher levels of noise in those frequency regions and making them harder rather than easier to interpret (in my view).
 

·
REW Author
Joined
·
6,711 Posts
Regarding how to generate an IR that includes cal data, basic steps are :

Take FFT of IR
Convert complex output of FFT to amplitude and phase
Apply the cal files, interpolating their values as required
Convert back to complex
Inverse FFT to get back to the time domain

If interpolating phase need to unwrap it first. Note that many cal files don't include phase, but Hilbert transform should be fine to generate the min phase response.
 

·
Premium Member
Joined
·
1,562 Posts
Thanks John.
I will see if I can do this. I enjoy puzzles.
I am just curious to see how the step response changes with the 2 cals included. I know it will be a trivial change, but I want to see if it follows my expectations. It helps me determine if my understanding is correct.
 

·
REW Author
Joined
·
6,711 Posts
Unlikely to make much difference on a speaker or drive unit measurement, since the bandwidth of the speaker is less than the mic and so the the speaker is the limiting factor in the IR.
 
1 - 9 of 9 Posts
Top