Home Theater Forum and Systems banner

Completed Soffits

5K views 34 replies 6 participants last post by  Bob_99 
#1 · (Edited)
I’m not sure if anyone is interested but here are the results of the soffits that I just built right behind my listening area. I decided to build them so they would be more acceptable to my wife since she was kind enough to let me put up Real Traps in my listening area.

The soffits are 13' long and 12” wide by 10” high with ¾” wood frame. I know that there may be better ways to do this but since this is my first (and only) time, I went for something that would be light. I initially tried to screw the top piece to the joists above the ceiling but even after I crawled up there to measure the distances, it turned out that the joists curved enough from the center where I measured to where they connect to the wall, that I missed them, so I ended up using molly bolts (old house). Molly bolts are a story in themselves which I won’t get into.

For the inside of the soffits, I went with four inches of 703 12”wide on the bottom and 6” high on the side. I had some R11 that I wanted to use up, so I filled the inside with that.



The whole thing is wrapped in poly batting and I put some plastic corner molding on the bottom for a smoother curve.



I have some cloth that needs to be put over the whole thing and then some quarter round molding and end caps will finish it up.

The big question, was it worth it? Here are some measurements using RplusD (sorry REW folks, I’m just use to this software).

Before


And after:



There appears to be some improvement but obviously not as much as I would have liked. It really does take a lot of surface area to reduce decay time at low frequencies. After I finish up the soffits, I’ll take a look at other wall/ceiling corners and see what I can sneak in for treatment. After that I may redo the listening area to cover more of the corners.

Special thanks to Bryan and others for answering many of my questions before I started on this project.

Bob
 
See less See more
4
#2 ·
Looks like you got most of it above 70Hz or so - what is left is travelling waves - not standing waves. That's a function of other things. The issue with soffits is that it deals with a TON of height dimension and little of the length dimension - where most of the deeper modes usually present themselves.

That said, the decay time in the room is absoltely going to be more in line and the overall experience is definitely going to be much better.

Bryan
 
#3 ·
Hi Bryan

the graphs don't look that different to me....is it possibly one of those cases where the changes won't show up on the graphs???

Bob, what is your subjective response to the changes, even if the graphs don't seem to be all that different is the case that it sounds a lot/moderately/slightly different/better???
 
#4 ·
I see changes from 70-200 in the slope of decay. The tails that remain, as I said, are travelling waves and harmonics of the length dimension.

Bryan
 
#5 ·
Terry,

It's like one of those puzzles where you get two pictures and have to find the differences. At first I thought that I had wasted my time until I started to look closely. In addition to what Bryan has mentioned about the decay slope, there is also a slight increase in the dB level. To my ears, I can hear the loudness and it does sound a bit clearer but to be honest, anyone who treats a room wants to hear a difference so badly that I can't be sure what is psychological and what is real which is why I used the graphs.

Bryan,

Thank you for your analysis. While I understand the graphs a bit, I had no clue about the tails and couldn't figure why they weren't going away. It looks like I still have quite a bit to learn.

Bob
 
#6 ·
Don't read too much into them. They weren't taken from the same place most likely. I see things in the 2nd one that aren't in the first one at different places.

Also, the low tail at 60Hz is most like a hum from a flourescent fixture or a power supply somewhere. Everything from about 70 up is pretty well gone by .25 seconds except at about 140 and 160 - not sure what those are. Have you tried anything thick on the back wall behind your head?

Bryan
 
#7 ·
I have not yet treated the back wall behind me. It is about 24' away and is the kitchen area with cupboards on the wall. I am planning to put something above them but once again will have to work it into my wife's decorating so I'm not going to get anything super efficient. Although she has been on a minimalist mood of late so maybe I could persuade her to get rid of everything in which case I could install something better :devil:

Bob
 
#8 ·
thanks guys, am interested cause treatment is next. Out of curiosity, would the 'chasing down' of the items at 140 and 160 (in relation to boundaries at whatever distance) be helped by looking at the impulse graphs?? ie seeing at what time events occur and then relating back to distances?-or am I barking at the wrong tree.
 
#9 ·
Well, it's always best to look at all of the information available - though I don't think the impulse response will identify why those are there.

Bryan
 
#10 ·
I will play around with the room furnishings this weekend to see if it's an object in the room or if it's the room itself. Perhaps it's a spot that's yet to be treated. I'll post any results when I'm done.

Bob
 
#11 · (Edited)
As I continue to analyze the results of the soffits, I am hoping that someone can answer a few more questions. A lot of attention is paid to the low end of the frequency spectrum but what about the mids? Bryan had suggested putting a foil (frk) on the bottom of the soffit depending on how the walls were treated. Since one side under the soffit is an 6'x8' sliding glass door covered with a medium weight curtain and the other side is almost all glass, I didn't put anything on the 703 assuming that there was a lot of reflection under each soffit. Here are some pictures of one of the walls and my measurement of the mids:

(Incompleted soffit)


Before the soffits:



After the soffits:



My questions are:

1) Do I look at graphs of the mids or should I just go by ear?

2) I thought the purpose of treatment was to get an even decay time and that may be true for the lows but the highs just decay too quickly and I don't believe you can do it across the entire range. What is the correct way to look at the relationship of decay time between the highs and the lows?

3) The after graph seems to show more decay time in some area? I had to check several times to make sure I was looking at the correct data. Is this possible? Should I place a foil on one side of the soffit?

4) Should I be looking at placing diffusers somewhere?

Again, thanks to anyone who can provide answers.

Bob
 
#13 ·
Up high, don't count on the waterfalls too much. The room should have somewhat even decay times across the spectrum - being slightly longer on the bottom and slightly shorter on the top. The plots you showed have a period of 2ms - that's impossible to see anything on realistically.

Bryan
 
#14 ·
Bryan,

I'm not sure if you're saying that waterfalls aren't accurate for mids or if the way I did it was incorrect. As far as I know, in RplusD the gate time has to be really low to get the mids. Is there any way to measure mids because rellying on my ears could be problematic. With my ears, I would probably settle for OK when there's something better. A good example was thinking my system was OK until after treating the room and hearing what was really possible. Unfortunately, now I have the bug to wean every bit of good sound out of the whole environment.

Thanks again.

Bob
 
#15 ·
If you want to try it, that's fine. Make sure you have a decent vertical range. Your LF measurements you had up to 44db (and even that's kind of skimpy). On the mid/highs, you have less than half that. Yes, you need shorter gate times for mids/highs but 2ms is kind of pushing it. I mean realistically, if your TARGET decay time in the room is 200ms, what are you going to see in 2?

Bryan
 
#17 ·
No probs. It's just that realistically, once you get above a few hundred Hz, you're more interested in frequency response and impulse response issues rather than decay times. It's pretty easy to get enough mid/high control in a room.

Bryan
 
#18 · (Edited)
After completing my soffits (minus some trim), I decided to redo my listening area. As shown in the pictures below, I initially had two Real Trap low frequency minis on the front ceiling/wall corner and one each on the left and right wall.



I decided to build a super chunk type of trap for the front wall/ceiling.



Move the original front minis to double up on the sides:

Right side:



Left side:



and add a couple of small panels behind the front speakers:

Finished front:



Measured results confuse me and hopefully Brian or Ethan can offer some advice.

Frequency response looks good (I think):



but the waterfall graph still befunkles me as I still don't see a decline in the low end decay rate.
:scratch:



I didn't expect to eliminate the low end decay but with everything I have on the walls, I sure would have expected some decrease. I also still have some noise between 120 and 170. What am I missing to help knock down the low end decay? :wits-end:

As always, thanks for any help.
Bob
 
#19 ·
It's not really all that bad. Your FR is looking really good except for that one null but even it is only down 6db from a REALLY nice flat 0 response for most of the rest.

The FR shows peaks about where you have the 'noise' so there are still modal issues at work. You can kill every corner in the room and if the ringing is coming off the back wall behind your head for instance, you'll still get the long decay times.

However, much of what I see is not a nice smooth decline. It goes down nicely and then levels off. Tells me there is some other random noise in the room or the measurement system somewhere that is giving some false readings - not sure what though.

You're actually looking pretty good down to about 50ish. Below that it's just really tough. Decay times are not supposed to be a flat line. The bottom end will be longer and the top end will be shorter. Try rerunning things but use a full 60db scale on the vertical axis and let's see if some of those aren't travelling waves or if we can figure out exactly what the story is. 140-160 is a second harmonic of an 8' ceiling by the way

Bryan
 
#20 · (Edited)
Bryan,

Thanks for the input. I'll run more measurements but when I took these, I was alone and able to turn off anything that might be a source of noise (including lights) which is why I'm baffled. Do you think it could be a cabling issue?

140-160 is a second harmonic of an 8' ceiling by the way
I was thinking of adding a larger ceiling cloud (6'x8'x4") but was about two panels short. I also wasn't sure how much of an improvement it would make. I have a hunch that I'm rapidly approaching the diminishing return principle.

Bob
 
#21 ·
Bob,

the waterfall graph still befunkles me as I still don't see a decline in the low end decay rate.
As Bryan said, it's not as bad as it looks. This is one of the things that bugs me about the R+D software - there's no way to control the vertical dB range on waterfall graphs. If you were able to set the total range from top to bottom at 30 dB, like most of the ETF graphs I post, the decays would appear much shorter.

--Ethan
 
#22 ·
Down to about 70Hz or a bit lower, you're down 30db (and likely more) by 200ms. At 50Hz to have a 400ms decay time is not at all bad. You've smoothed the response nicely too - especially from 60Hz down which is tough to do.

The only overhead I'd be concerned with is the one right over the seating position. Did you end up putting anything on the rear wall behind the seating position?

Bryan
 
#23 ·
Ethan,

Thank you for your thoughts.

This is one of the things that bugs me about the R+D software - there's no way to control the vertical dB range on waterfall graphs.
Yes, I'm hoping that Doug will change that in a future revision.

Bryan,

Did you end up putting anything on the rear wall behind the seating position?
The back wall (37' away) is the kitchen area which means double trouble. One, it's a wife area and two, the cabinets are there. I did manage to get a 24"x24"x34" chunk into the corner above the cabinet when my wife away :whistling:. (She took it very well). I'll see what I can get up against the back wall that will not be too obtrusive. She really has been outstanding about letting put up all the panels in exchange for having a free rein with a paint brush. Thus, all the pastel colors.




Thanks again to both of you for your input. It is very much appreciated.

Bob
 
#25 ·
Thanks. I tried to make it fit in to the decor so it would be acceptable to my wife. Physically, it was a pain because not only is the corner not 90 degrees, the left wall bows out a bit. Old houses can really be annoying when trying to do a project.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top