Initial room measurements

dmjung · Post by **dmjung** » Fri Feb 03, 2006 2:38 pm

Ok, I've finally got around to measuring my new room. Attached should be pdf's of the data and a chart along with the actual excel spreadsheet. Also some pix of the room since that will probably make more sense than what I write.

Trapezoid shaped room; short parallel wall is 16' 4"; long wall is 19' 4" and these two parallel walls are 19' 8" apart. The ceiling is 8' 9". Tile floor; sheet rock walls/ceiling. There are two 6'x9' rugs angled at the back of the room (may not be visible in the pix.)

Measurements were done using a Radio Shack digital spl meter at various positions using 1/3 octave pink noise. The first 3 series were done using some old Polk Audio speakers and the last series was done with some new Axiom M80's (which I would expect to be more neutral and have a wider range than the Polk's.) Speakers were located at the short parallel wall with their driver side a little over 3' from the wall and roughly 8' apart. They are toed in to form a rough triangle with the chair.

Series 1 was measured roughly in the center of the room between the chair and the ottoman; pointed between the speakers.

Series 2 was measured from the left, rear quadrant; about where you'd be sitting behind the blue drums.

Series 3 was measured from between the speakers and facing the far wall.

Series 4 (new speakers) was measured from the same position as Series 1.

The chart is drawn using the adjustment factors for the RS meter to supposedly make it more linear. If you don't think that's necessary, the original measurements are in the excel sheet. I noticed the meter was jumping around quite a bit over a range of about 4 or more db at 40 Hz thru 100 Hz. Is this just the meter or the room being funky?

The chart shows a drop off at the ends of the spectrum. My ears tell me the room is "middy", perhaps slightly "boomie", but the low bass disappears. I don't hear a lot of real highs either, which may be partly the room and partly my ears since my daughter says the room sounds "trebly". I guess what I'm hearing corresponds to the measurements...although I wonder how much of this is coming from the room, the speakers' performance, or the meter's capability...obviously an interaction between the three is going on. Thoughts?

So, does anything jump out at anybody? What's my next step?

Thanks!

--David

knightfly · Post by **knightfly** » Sat Feb 11, 2006 8:35 am

Doesn't look that bad, but 1/3 octave isn't enough resolution to judge; your comments make me think you might be helped by this -

http://www.johnlsayers.com/phpBB2/viewtopic.php?p=18059

HTH... Steve

dmjung · Post by **dmjung** » Wed Feb 15, 2006 2:10 am

knightfly wrote:Doesn't look that bad, but 1/3 octave isn't enough resolution to judge; your comments make me think you might be helped ...

Thanks, I'm digesting the room node stuff. Since I don't have a rectangular room, I'm using the average of the two parallel walls. When I get to graphing I'll use the two actual lengths and connect them. I've piddled with moving the speakers and listening position back and forth based on eyeballing the generated graphs...no particularly earthshattering effect. I think the biggest effect might have come from stacking a bunch of large throw pillows in the corners to approximate a corner trap.

I've also downloaded the RealTraps LF test CD and will do some measurements with that. Since I made the initial tests, my subwoofer came in...it added a minor encrease in the bass, much less than expected though. However, it does add to what can be "felt" in the room and heard outside the room so it's at least moving air and the house.

I have the equipment/software to do some full spectrum measurements...just need to find the time to get it set up.

My plan at this point is to build around 15 of the panel absorbers to permit at least 3 per wall and two on the ceiling. Would it be beneficial to build them longer than the ceiling height to allow placing them from the ceiling to the floor at an angle or just go with a more manageable size?

Rigid fiberglass in the panel absorbers...I've got some calling to do to locate this stuff. Can this rigid fiberglass be "built" by just compressing the normal pink fiberglass between a heavy wire assembly? Kind of like a tube bass trap unwound and laid out flat?

--David

knightfly · Post by **knightfly** » Wed Feb 15, 2006 5:00 am

Just tracked this down in the Building supplies forum and "stickied" it, hopefully it will help you -

http://www.johnlsayers.com/phpBB2/viewtopic.php?t=1643

Trying to compress standard fiberglass isn't too practical, you'd need to compress (and keep it compressed) by about 4 TIMES, and even then I doubt the results would be equal since the process is done on huge roller equipment and I've no idea what they do to keep it from "UN-compressing" again - this very well could affect absorption coefficients, etc...

IF your local SPI store doesn't have 703 (or even if they do) you'll probably find that mineral wool/rockwool in same density/thickness is cheaper, and with the possible exception of use in Helmholtz traps should be nearly identical in results... Steve

BTW, sounds like your sub may be in a null of sorts - don't remember which "white paper" covered this, but one way to position is to put the sub right where your head would normally go, then crawl around the floor (kneepads help) til it sounds best, and that's where the sub can go.

dmjung · Post by **dmjung** » Fri Feb 17, 2006 11:43 am

Did some more extensive measurements using 1Hz measurements from 10 to 300...didn't get a reading until around 18Hz.

Speakers are still located against the long parallel wall. Driver side is 26.5" from the wall behind and 55" from the angled side wall. The subwoofer is located 80" from the left angled wall and it's driver side is 29" from the wall behind it.

The RS meter was located 108" from the long wall and 107" from the side (roughly centered in the room.)

Attached is a zip file with the measurements in an excel sheet and pdf of the measurements graphed.

Comments: the dips around 146, 186 and 242 -roughly- correspond with where I think the ModeCalc program indicated problems...not exactly presumably due to not being a rectangular room and being non-empty.(?) The huge dip around 114 also did not show up at all using ModeCalc...if I understand what I'm looking at when I run it.

I also noticed that quite often what I was hearing as increasing and decreasing volume at each frequency did not seem match the meter on a pretty regular basis. I assume this is partly due to my head being in a different location than the meter and typical rock musician deafness.

I turned my contractor loose on locating OC703/705 or similar...we'll see what he turns up.

Suggestions and comments are welcome...

--David