Major CR nulls.

Hi, this is my first post. I've just recently finished building a studio in a wearhouse, and I'm having a lot of issues with my control room's acoustics. The dimentions of the room are 125"x109"x96". The walls were built with 2x4 studs, 16" on center, covered with two layers of sheetrock and a layer of Duroc on each side, making the overall thickness 6.5". The ceiling is sheetrock over 2x6. The room has been treated as follows: The two rear corners have 4" of 703FRK from floor to ceiling, the rear wall is essentially covered, floor to ceiling, with 4" of 703 (unfaced), spaced 2" off the wall. The front wall is treated the same way as the back wall, and the ceiling has 36 sq. ft. of 4" 703 and 703FRK in 2 sq. ft. sections, spaced from 0" to 2" off the ceiling. I also built two 6' long slat resonators across the front corners, starting 18" off the side walls, floor to ceiling. Those were tuned to the modal frequencies of the room. Now, I thought I had treated the room relatively well, but, on testing, found there to be a 15dB hole centered at 119Hz, about 12-15Hz wide. There is also a hole around 280Hz which is around 8dB deep and a bit wider. There are a number of others, all of which seem to be centered at harmonics of those two. I have tested a number of positions and found that the room is nearly flat (within 5dB, and a much smoother response) if I lower my mix position from the original 47.5" (I know, I know), to about 36". But that is too low for me to sit comfortably, and the monitors are below my desk/racks. At all other heights (I tried anywhere from 40" to around 58") the nulls are too deep to work with. I tried shifting the mix position and the monitors back and forth, but with no success. I also used the mode calc to find the best position for my monitors/mix position, but I was already pretty close and the change had nearly no effect. Is there anything I can do? There is no space to add more treatment and I can't change the size of the room. I'm not concerned with a budget at this point, so make whatever suggestions you feel might help. Thanks in advance. Richard

I'd like to check out the data from your room myself. What software are you using? Can it export an impulse response as a wave file? If not, could you record a test tone I would give you and send it back to me? What mic are you using? I'd be happy to help you find a better location for your speakers if you could draw up some kind of sketch of the room and provide some pics of the room. You can get a hold of me and send me files through my email (button below post).

I'm using Fuzz Measure (both pink noise and log sweeps) with a Behringer ECM8000, and I double checked my results with a Radio Shack SPL meter and 1Hz increment sine waves. Yes, I believe I can export an IR as a wave, though I need to double check that because I'm using the free version. I have some things to attend to in the morning, but tomorrow afternoon I can get a sketch of the room up, some pictures, as well as the test data. I appreciate your help.

Well, if you've already discovered that the height is the issue, then have you considered going the other way? Try raising the monitors and your seating position a foot. It may require building a platform but that's not all bad. If you did that, you could also tune that platform as a perforated Helmholz resonator for the frequencies at issue. Other options would include a large poly (big enough to assist at those frequencies) directly overhead (potentially not feasible due to depth requirements - OR - tuned resonator overhead. You may also find that it's not just the height but the buildup of the height and one or more tangential or oblique modes which happen to coencide where you're sitting. In this case, potentially some straddling broadband absorbers on the side wall/ceiling intersection directly beside your mix position. Personally, I like the first solution as it's really dealing with the problem (seating position at 1/2 of the height) as opposed to treating the symptoms. Bryan

Ok, I drew a quick sketch, the dimentions of the room are in my original post. The angled walls at the front of the room are slat resonators. I also took some more measurements and found I could not reproduce the flattest response I mentioned previously. I'll post the graphs of today's measurements, and also some pictures. I also have IRs in the form of AIFF files, but I can convert those to WAV if necessary. Thanks for all your help. Richard

Alright, here are graphs of the tests I did today, sorry about the bad resolution, it's all I could get. Test 1 was with the monitors at the original height (monitor midpoint at 49", center of driver at 45"), spaced 35" apart, 18.5" from the front wall. The mic was placed at 49", 35" from each speaker. Test 2 was the same, with the monitors and mic moved up to 61". The center of the driver was at 57". For test 3 I moved the mic back down to 47.5" (height of my ears while seated) but left the speakers at 61". Test 4 was everything moved down to 36", still in a 35" triangle, still 18.5" from the front wall. EDIT OK, sorry, the images were in CMYK, fixed now. /EDIT

Here are some pictures of the room, they should be self-explanatory. If you have questions, please ask. The side resonators are actually in three sections, so I can use the windows on either side of the room while I'm tracking. I've posted a picture of the wall with and without the middle section.

Could you email the IR's to me along with test detials like stimulus, speakers, mic, locations of those last two, etc.?

Sure. The mic/speaker location was the same for all 4 tests, only the height changed. Do you need the IRs to be in wave format, or can I send them to you as aiff files? Thanks, Richard

I just took a good long look at everything, and I'm still not exactly sure what's causing the low end notch. I don't think it can be modal since your room is modeless in that frequency area. Also, the slice goes away later in time. That leaves boundary interference. The only clue I have is that the notch seems to go up in frequency when when move the speakers down, which suggests that a surface below the speakers ias causing the notch. Down there lies the desk and the floor. I haven't seen a desk do anything like this before, but I'd like to see you put some thick absorber on the desk with a ncie airspace and measure again. How audible is this problem? Play a sweep while listening at the normal position. Is it bad? There's no point treating something we can't hear. I'm thinking this will be audible, though. You really have a pretty nice room there. This is a pretty minor feature in the grand scheme of things. My advice to you at this point would be to experiment. Move things around. Move the desk/speaker arrangement forward. Try a slightly larger listening triangle. I'd be happy to look at any future IRs you make.

Thanks for the quick reply, David. FWIW, the desk is quite large in relation to the room, but I'm not sure there's much I can do about that. The floor is concrete, but it's pretty much covered by the desk between me and the speakers. If putting absorption on the desk helps, is there anything short of doing so that will help fix this? Also, for these tests, I moved the desk out of the way so that I could fully experiment with placement. Moving the desk didn't seem to do much of anything, response-wise (I tested identical positions with and without the desk). Sorry, that probably would have been useful information. The problem is very audible. When I play a series of 1Hz increment sine waves, it starts getting quiet after 100Hz, and is almost silent at mix position when 119 rolls around. It returns to normal level somewhere around 130Hz. When I first moved in, I didn't think it was going to be a big deal (it's pretty narrow, right?) but I've found it to be nearly impossible to mix the way things are. Mix translation sucks. Thanks for the kind words, I'm pretty proud of it. I'll go back and try to work on positioning a bit, see if I can some up with any answers. I have an extra box of 703 saved for this type of thing. Richard

Can't tell from the pix; are your woofer cones anywhere near 30" away from your front wall? Also, what is the front wall width between the slats? Steve (Second harmonic of length is around 108 hZ, but this could change somewhat with the absorbent front and rear)

Richard - is that a sub under the desk - if so what is the crossover frequency of the sub?? cheers john

Steve - The satellite woofers are 18.5" away from the front wall (the actual wall, not the absorption). The front wall width between the slats is 73". John - Yes, that is a sub, the crossover is centered at 80Hz. But, for these tests I've been using a pair of Event TR6s because I didn't want the positioning of the sub to screw with the results. I did do some tests with the sub, and the response below 80Hz really isn't too bad. I was a bit surprised, considering I just shoved the sub in the middle of the room, under the desk (it was really the only place it would fit). Thanks, Richard

OK, and how far was the mic from the front wall for the tests? (That distance you gave is to the cone itself, not the rear of the box, right? Just making sure)... Steve

Mic was 47.5" from the front wall, which puts it just slightly inside the 35" triangle of the monitors. And yes, the distance was to the cone itself, not to the rear of the box.

Yeah, I did all this math, and none of the numbers worked out. :( Hard to say what's wrong without being there.

OK, well I spent about 4 hours running tests today. Basically, I found that height had a lot less to do with the problem than I thought. The biggest changes happened when I moved the monitors either closer together or further apart. Basically, closer together was flatter, but I can't really have them in any closer than 35" or so, and the difference between that and 20" or even narrower was very small (I tried 5 inch increments from center out to 54" apart, the widest I can go). The very best (excluding those I cannot implement, like everything 35" off the ground) position I found was with the microphone at 47.5" from the front wall, and the monitors in a 35" triangle. I did a LOT of tests (changing height of monitors and microphone, moving the mic around the room, etc.). I concluded that height really had little effect on the overall response, compared to the other variables (I tried anywhere from 35" up to 61"). So, I just went with the monitor stands I've been using (putting the midpoint of the monitors at 49"). FWIW, higher was worse. At this point, I decided to see what effect the side walls were having. Seemed like the logical choice, since moving the monitors in relation to them changed the response considerably. So, I took out the middle section of the slat resonators. Bingo. Problems got worse, nulls were deeper, wider, etc. I thought, well maybe the slat resonators aren't really doing their job. I replaced them with 6" of 703 placed at an angle across the corner. The difference between null and peak was reduced about 3dB more than with the resonator, BUT the null was quite a bit wider. And it was still over 10dB deep, so I'm not satisfied with that either. Any ideas?