Suggestions for BFD and REW Tips

[Sonnie]

In this thread you can make a suggestion to add a tip for using the BFD or Room EQ Wizard. It will be reviewed by the moderators of this forum and considered for inclusion in the appropriate thread. We reserve the right to edit the tip if it is included. Thanks!

[Malice]

Hi Sonnie, I made it over!!

My $0.02........

My experience with the BFD and REW has shown that I prefer different house curves for different source material. For movies I like the curve to be flat, and for music I like to have the lower frequencies boosted a little. But how much boost?

This is where the BFD's multiple profiles or presets can be used to good effect to create multiple versions of your preferred house curve.

What I would say, however, is that without a MIDI connection to the BFD from the computer I would have found this a complete PITA! So my advice is to get a MIDI set up as soon as you can.

Then you can easily create multiple profiles like the ones below.



The red (top) is the unfiltered resposnse, and you can see the other presets I've dialled (well Midi'd) in. #1 is for movies and #2 for music. #3 and #4 are good, but #2 has the edge.

I would have had even less hair if I had not used a MIDI set up to get all those filters in place.



Bob

[Sonnie]

Hey Bob! It's about time! Welcome to the Shack and we are certainly glad to see ya!


Excellent tip... and I'll copy it over to the TIPS thread right away.

Thanks!

[Sonnie]

As soon as I can figure out how to move just one post.

[JohnM]

Wireless network cards can cause interference on audio inputs resulting in incorrect measurements, they should be turned off or disabled when making measurements.

[Jack Hidley]

I had a need to take some acoustic noise measurements in a room recently, and wanted to try to use REW for it. The measurements needed to be A weighted. Unfortunately, REW doesn't have any weighting options for SPL measurements.

To enable me to take A-weighted SPL measurements, I calculated the SPL offsets for an A-weighting function from 5Hz to 20kHz, and then multiplied all of the amplitudes by -1. I saved this as a cal file. If you load this cal file into REW as a mic cal file, when you use the RTA/Spectrum feature to look at the noise spectrum in the room, the SPL value next to the Spectrum label in the bottom left will be A-weighted. The SPL value on the tool bar is NOT affected by the cal file, so it will remain unweighted.

You will of course need to turn off the cal function when taking impulse response measurements.

I've attached the A-weighting cal file.

[brucek]

There are a couple of problems here.

and then multiplied all of the amplitudes by -1

REW mic calibration files reflect the actual response. In other words, if a mic reads 10dB low at 20Hz, then the cal file entry is: 20 -10. Your file is the opposite of this. Your entry at 20Hz for example shows 50 and it would need to be -50..........

You will of course need to turn off the cal function when taking impulse response measurements

The cal file is reflected in the frequency response and not applied to the impulse response measurement. This how you are able to take an old measurement and apply a new cal file to the frequency plot.

To enable me to take A-weighted SPL measurements

You may be confused. The intent of REW is to produce response measurements as if every mic/meter used was perfectly flat over the frequencies of interest. If we actually had a flat mic, we would not use a cal file and we would not click the C-Weight checkbox. The mic is already flat. That's in a perfect world.

When we use a "flat" mic such as an ECM8000 we still use a cal file to correct its top and bottom end to result in a response that theoretically (and practically) used a flat mic.

The same applies to a Radio Shack meter that only has a switch to filter its inputs to C-Weight and A-Weight. On REW we apply a cal file to remove that weighting influence in the measurements, so as to result in a response that theoretically (and practically) used a flat mic.
(note, if the Radio Shacks meter response tracked a C-Weight curve exactly, then we would not need a cal file, we could simply click the C-Weight checkbox).

So, if you have your meter set to A-Weight and you offset it in REW with a A-Weight cal file, you have essentially taken measurements with a flat mic.........

brucek

[Wayne A. Pflughaupt]

Yeah, seems to me that all you need to do to get an A-weighted REW graph would be to set the meter to A-weighting. A calibration file would only be needed to compensate where the meter might deviate from the A curve, if that kind of accuracy was needed.

Sure would be curious to know why and what kind of REW response plot would need to be A-weighted...

Regards,
Wayne

[JohnM]

I think Jack's file is correct for the stated purpose, which is to impose A weighting on a measured response. The RMS value in the Spectrum/RTA takes account of the cal file which is why it works there, but not in the SPL Meter dispaly which is calculated from the raw captured data.

[brucek]

I think Jack's file is correct for the stated purpose, which is to impose A weighting on a measured response

Ahhh, so you think he has a flat mic and wants to create A-Weight with REW?

brucek

[Jack Hidley]

In this case, I wasn't using REW for taking frequency response measurements. I just needed something portable to take environmental noise measurements. In this case it was to certify the performance of a room for acoustic measurements. I needed to be able to measure down to about 15dB SPL for this. Acoustic noise measurements are frequenctly performed with A-weighting since that it gives a good 1st order approximation of loudness.

The mic I'm using is an Earthworks TC30. It is -3dB at 9Hz. No need to calibrate for the mics response.

I was just trying to show people something else you could use REW for, that wasn't a built in function.