Need Help With FBQ2496 Filter Settings

Michael Toor

Registered
Thread Starter
Joined
Oct 13, 2017
Messages
7
Hi everyone,

I have a home studio and worked with an acoustic designer to build it. He designed it and I did the construction. Throughout the process, I used REW to coordinate the status of the control room to him (he lives halfway around the world). I'm very grateful for this software and how much it has helped me. In the final stages, my designer had me try various filter settings on the FBQ2496 unit in my chain to get the mix position sounding as flat as possible.

The reason for this post is that I seem to no longer be able to reach the designer who created the filter set (attached) and I want to verify my filter settings. I've changed out the carpet in the control room for one slightly thicker one (the old one's edges were curling all around) and wanted to make sure the mix position is still sounding as good as possible. However, I'm not sure how calculating filter settings works. I took some REW tests before and after changing the carpet out and my plan was to just look at the SPL curves, RT60 graphs, and Waterfall Graphs to determine if much changed. From what I can see, those seem to look consistent at a glance, however I'm not experienced in acoustical analysis. The extent of my use of REW has just been calibrating the system and generating test results for the designer I worked with.

I would greatly appreciate it if someone with experience could help me out here by taking a look at my results and the filter settings (which were applied during all of the tests since the FBQ2496 unit was active in the chain) and verify if I need to change anything. I noticed when I sat down after switching out the carpet to test some music, it felt like the music sounded much tinnier and the mids/body of the sound was lacking. I could just be paranoid but it would help to have someone who understands REW tests better to verify. I've attached the REW tests.

A few things that might be worth noting:
- I set the soundcard and mic calibration files prior to running any tests
- I'm not sure if I was supposed to add the filter set somewhere before running the tests but I added the filter set to each of the tests AFTER running all of them just so you can see the values used. I'm not sure if that warps any results but if so, just remove the filter settings and you'll see the results as they were when the tests were run --- with the FBQ2496 unit active in the chain but the settings not applied in REW itself
- There are 8 tests in total; the first four with the old carpet and the other 4 with the new carpet
- Test one is with the mixing desk chair placed right behind the mic stand and me crouched in the room under the mixing desk
- Test two is with the mixing desk chair placed right behind the mic stand and me outside of the room entirely
- Test three is with the mixing desk chair placed in the back corner of the room and me crouched in the room under the mixing desk
- Test four is with the mixing desk chair placed in the back corner of the room and me outside of the room entirely
- The next four tests were the same scenarios in the same order but with the new carpet instead of the old one
- The monitors being used are Adam A7X satellites with an Adam Sub8 (the designer chose the settings for the EQs, etc. as well for these)
- I can't quite remember for sure and can't check because the speakers are in an elaborate enclosure but I think the tweeters were set to +4 and the lows were -5 for the satellites
- The sub crossover is around 130 Hz I believe and I think the phase was 0 with the 85 Hz cutoff applied before sending the L and R signals out to the satellites

I would also appreciate your thoughts in general on how the filter set was working prior to changing the carpet. Was the calibration done correctly with my studio designer? It sounded pretty flat but please let me know.

Thank you for taking the time to read all that and again I would greatly appreciate any help and advice.

- Michael

UPDATE: I had to divide the .mdat into two due to file size, so Part 1 contains tests 1-4 and Part 2 contains tests 5-8
 

Attachments

  • Part 1.mdat
    12.5 MB · Views: 8
  • Part 2.mdat
    12.5 MB · Views: 4
Last edited:

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
There don't look to be any significant differences with the new carpet. It looks like you may have made the measurements with both speakers playing, that generates comb filtering at the high end if the mic isn't equidistant (to the mm) to each speaker, better to measure each individually.

The filter set is very extensive, quite excessive really. Broadly speaking EQ is best restricted to the range below a few hundred Hz unless you are compensating for some inherent characteristic of the speakers (rather than of the in-room measurements). You should probably make some measurements with the filters bypassed to see what the responses look like without them, and spend some time listening with the filters bypassed to see whether they are really helping.
 

Michael Toor

Registered
Thread Starter
Joined
Oct 13, 2017
Messages
7
Hi John,

Thanks for your quick reply. Yes, I had both speakers playing.

Here are some new tests. Part 1 is various speaker arrangements (Sub Only, L + Sub, R + Sub, and ALL ON) with the FBQ2496 in bypass mode. Part 2 is the same speaker arrangements with the FBQ2496 filter set active (each measurement has this filter set added in the file after taking the measurement). Since the signal is quieter with the FBQ2496 unit active, I had to adjust the mixer gain to make sure -12 dbFS noise was playing at 80 dBC (slow) (verified via an SPL meter) with a single speaker on in both cases (with the FBQ2496 active the mixer master gain was set about 5 dB louder to achieve the 80 dBC noise loudness).

The filter set was created with help from the John Sayers production forum so I don't really know the reasoning for the values used; it's just what the designer suggested. Like I mentioned in the previous post, the designer had me set very extreme EQ values on the back of the Adam A7X speakers prior to locking them away in the elaborate soffit mounting. I can't recall what the high shelf was set to but the tweeters were set to +4 and the low shelf to -5 if I remember correctly so that may be why the EQ filter set is so drastic. The filters could also be used to treat problems created by the room dimensions. The room was specifically built based on the designer's specs so the designer could have been treating the anticipated problems based on the dimensions he spec'd with the filters. I don't really know what I'm talking about so it's just conjecture.

When the FBQ2496 is in bypass mode I definitely hear a massive difference. All the mids get drowned out by intense highs when the filter set is off and elements like vocals are buried. When the filter set is active, the sound is pretty neutral and sounds very flat.

Please let me know your thoughts if you're able to take a look at the .mdat files attached. Should I rethink the filter set being applied?
 

Attachments

  • Part 1.mdat
    10.7 MB · Views: 2
  • Part 2.mdat
    10.7 MB · Views: 3

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
The results with filters look much better than without. Filter 19 (11,642 Hz) acts like a tone control for the high end, so you could alter the cut on that to adjust the high end to taste. Filter 10 (570.2 Hz) could perhaps have a little more cut applied, maybe another 2 dB.
 

Michael Toor

Registered
Thread Starter
Joined
Oct 13, 2017
Messages
7
Thanks for the quick reply again and willingness to help John. It's much appreciated.

Thanks for clarifying. I decided to go with another carpet and I have a stand coming in for my keyboard and mouse so I'll run some tests with those in place and the extra 2 dB cut. Should be a week or so before I get everything in.
 

Michael Toor

Registered
Thread Starter
Joined
Oct 13, 2017
Messages
7
So I decided to go with the original carpet replacement in the Part 2 file in my first post (also in Part 1 and Part 2 of the second post) but I did add the stand on my desk (see attached photo). I'm wondering whether having this stand will cause early reflection issues or some other kind of issues I don't know about. I've attached a new test with the new stand in place, the FBQ2496 active, the same carpet used as the last tests, and the mic as close as I could get it to where it was in the last tests I ran.

The stand is the wooden structure supporting the keyboard and mouse above the mixer. I don't really use the mixer much when working with digital audio so it was practical for me to create a space where I could have a centred keyboard and mouse. If I ever need to do live tracking, it's easy to just remove the stand and put it off to the side.

I'd really appreciate it if you could please let me know how this test set compares to Part 2 in my last post. Also please let me know if you suspect having this stand here will create listening issues or if you see any issues in the test file that suggest that. Thank you.

By the way, I didn't lower the 570.2 Hz filter by 2 dB yet. I didn't want to change too much in one stage of testing but it looks like the bump in the SPL graph disappeared after adding the stand. Let me know if you still want me to try a test with the 2dB lower 570.2 Hz EQ.
 

Attachments

  • PXL_20210320_162628451.jpg
    PXL_20210320_162628451.jpg
    302.3 KB · Views: 13
  • New Tests.mdat
    8 MB · Views: 4

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
Looks much the same. Playing with the EQ is probably something to experiment with if you are unhappy with some aspect of how it sounds.
 

Michael Toor

Registered
Thread Starter
Joined
Oct 13, 2017
Messages
7
It's been a busy few weeks but I ended up making the 2 dB tweak you suggested to Filter 10 (570.2 Hz) in my FBQ2496 where I adjusted the gain by -2.0 dB.

The first three tests iterate through Left satellite + Sub, Right satellite + sub, and All speakers on with the filter unchanged (the tests where the titles say OLD FILTER).

The last three tests iterate through the same speaker arrangements with the filter gain (for 570.2 Hz) dropped 2 dB as you suggested (the tests where the titles say NEW FILTER).
 
Last edited:

Michael Toor

Registered
Thread Starter
Joined
Oct 13, 2017
Messages
7
Looks much the same. Playing with the EQ is probably something to experiment with if you are unhappy with some aspect of how it sounds.

Here are the new REW tests (see descriptions of the tests in the above post).

Is anything useful happening? If so could you please describe what you see? The room sounds good in both cases. If there's no major difference I will probably leave the settings as they were but I would appreciate your input.

Please let me know. I appreciate the help very much. Thank you.
 

Attachments

  • April 20 2021 Control Room FBQ Trial Change.mdat
    17 MB · Views: 3
Last edited:

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
The differences are very small, but surely you can see that yourself just by looking at the graphs? There is no magic to it. If you are not hearing anything significant then why worry about it?
 

Michael Toor

Registered
Thread Starter
Joined
Oct 13, 2017
Messages
7
The differences are very small, but surely you can see that yourself just by looking at the graphs? There is no magic to it. If you are not hearing anything significant then why worry about it?

Hi John, I just wasn't sure if there was some deeper reason you suggested the 2 dB reduction that I was unaware of. My understanding of acoustics is minimal and I know the science behind acoustics can get intense. Thank you so much for taking the time to help me with all of this. I really appreciate your help.
 
Top Bottom