I read lot of things about FRD files, but lot of these are old, so, in 2021, what is the best or simple software to make FRD files? I'm planning to buy a miniDSP UMIK-1 microphone and a DATS V3 box.
Thank you! 🙂
Thank you! 🙂
FRD is just a file format. I guess you're asking what is the best testing utility program to make measurements?
There are numerous options. I like ARTA, but REW is very popular.
Dave.
There are numerous options. I like ARTA, but REW is very popular.
Dave.
You would possibly prefer a timing reference for some tasks, i.e. determining acoustic centers, and this is why you would like the capability for a loopback on a second channel utilizing a measurement microphone and an interface with two channels.
More specifically not needing to determine acoustic centres but just getting phase right from the start, for a crossover.
Yes, but Umik 1 and 2 are fine for the task with REW. Use the accoustic loopback.
REW made that setting for Umik 1.
This is a myth. Old news that is no longer news. At least with the systems I use to measure, which are all USB based.
I think that's an oversimplification Charlie. I would agree with you when using a single channel measurement system. In that case the Hilbert Bode transform is used to create minimum phase files. But, on the other hand, if you're using a dual channel system (and using the phase from the original measurement) the USB mic can cause problems. From the REW help file:
"Long sweeps may be problematic when the input and output are on different devices, such as when using a USB mic, as their sample clock rates will differ. Over a long sweep a significant difference in clock rates could cause severe distortions in the shape of the impulse response and affect the phase response. This can be corrected by using the acoustic timing reference and the Analysis option to Adjust clock with acoustic ref or, if a loopback connection is being used as a timing reference, the Analysis option to Adjust clock with loopback."
I don't think John Mulcahy would add these features to REW if USB mics didn't have problems in dual channel systems.
The way I read this, a USB mic doesn't impose any limitations for measuring absolute timing (phase). You don't have the loopback option you'd have when using a 2-channel audio interface + XLR mic, but acoustic reference can still be used.
I recently got a Behringer UMC22 and Daytom EMM-6, because having the loopback option makes things a bit more straightforward to me, and there isn't a big difference in price.
Btw, I use REW for making FRD and ZMA files. For measuring impedance I made a cable as described in this article.
Last edited:
I read it to say that John added these functions (acoustic timing reference and analysis options) to REW to overcome the problem that USB mics have when it comes to measuring absolute phase.
If you want to get timing and phase right, don't use a microphone with a built-in USB interface. Using a proper sound card that has analog inputs is better for this, as it allows you to do a proper dual channel measurement with a loopback in the reference channel. Without the loopback, the software will have to guess the latency of the USB interface inside the microphone, which can be different from one measurement to the next. Guessing the latency therefore cannot provide reliable time/phase results.
Oh, if you really just want to make FRD and ZMA files, use a text editor and type some data into an ASCII document. Then save the ASCII data with a FRD or ZMA file ending. 😀
Oh, if you really just want to make FRD and ZMA files, use a text editor and type some data into an ASCII document. Then save the ASCII data with a FRD or ZMA file ending. 😀
Last edited:
When using the acoustic timing reference in REW, can a person make the whole series of off axis measurements (0, 10, 20, 30.... 150, 160, 170 degrees) and maintain the correct phase relationship between each measurement? i.e. will those 17 FRD files "work" in VituixCad2 ?
This is something I have been wondering about. It would be cool if it were true.
j.
This is something I have been wondering about. It would be cool if it were true.
j.
Well, if you don't move the microphone, and you keep the nominal acoustic centre of the test speaker in a fixed location, and you don't move the acoustic timing source, and you keep the acoustic timing source far enough away from the speaker under test so it doesn't cause a reflection, it should be possible.
Should be possible. But if you still need to buy the mic, I’d go for an XLR one with a 2 channel audio interface. Loopback makes proper timing reference a lot easier.
I am not deeply familiar with REW, but I don't see why that would not be possible.
No matter what software you're using, make sure to use the dual channel technique with a loopback channel for timing reference (see explanation in my earlier post).
I use the basic audio interferogram method that Charlie Laub posted on his site years ago.
I also use USB microphone (UMIK-1 calibrated CrossSpectrum Labs). I don’t mess with any timing or synchronizing stuff.
Take 3 sweeps with software of choice (I use REW) without touching speaker or mic: tweeter only, woofer only, tweeter and woofer in parallel. Use calibrated mic and measure the SPL and set Vrms of sine wave to correct reference value. A nice setting is 0.5m mic distance and 2.0Vrms sine wave at 1kHz as calibration. This scales nicely to 2.83vrms at 1.0m easily and is less susceptible to floor bounce dip. Convert sweeps to FRD files. Load FRD files into software of choice. I use Bwaslo’s Xsim but have also used Jeff Bagby’s Crossover Designer spreadsheet program which works great too. Adjust acoustic offset until simulated woofer and tweeter match measured response. Use that offset going forward with XO development.
Here is example of raw data. Your goal is to adjust offset in simulator until you reproduce all the little dips and peaks present in the blue trace:
I also use USB microphone (UMIK-1 calibrated CrossSpectrum Labs). I don’t mess with any timing or synchronizing stuff.
Take 3 sweeps with software of choice (I use REW) without touching speaker or mic: tweeter only, woofer only, tweeter and woofer in parallel. Use calibrated mic and measure the SPL and set Vrms of sine wave to correct reference value. A nice setting is 0.5m mic distance and 2.0Vrms sine wave at 1kHz as calibration. This scales nicely to 2.83vrms at 1.0m easily and is less susceptible to floor bounce dip. Convert sweeps to FRD files. Load FRD files into software of choice. I use Bwaslo’s Xsim but have also used Jeff Bagby’s Crossover Designer spreadsheet program which works great too. Adjust acoustic offset until simulated woofer and tweeter match measured response. Use that offset going forward with XO development.
Here is example of raw data. Your goal is to adjust offset in simulator until you reproduce all the little dips and peaks present in the blue trace:
Attachments
Last edited:
repeat that 70 times for full spinorama and you have got 210 measurements and the time for manual align 😀 two channel measurement setup is cheap compared to the amount of work.
Edit, computed it wrong, it is only the combination measurement extra, plus the alignment procedure.
Edit, computed it wrong, it is only the combination measurement extra, plus the alignment procedure.
Last edited: