Did a ceiling cloud just make my room worse?

[brett]

Greetings all. This is my first acoustics post and I'd love your collective wisdom on the below...

I've just built and installed a bunch of broadband traps and have seen a significant improvement in my room response. However, I finally installed my cloud and can't help but think that in some respects it's made things worse. Here's the background information:

My room dimensions are 4260mm (L) x 3665mm (W) x 3020mm (H). My speakers are on the 3665mm (W) wall, firing down the length. The house is old with a timber frame and weatherboards with no in-wall insulation. There is a large window at the front of the room. In other words, sound can readily enter (and escape) my room.

I have so far installed:

- Large floor-to-ceiling superchunks in each of the front corners (~83cm across the face)
- Two 600mm x 1200mm x 300mm thick RFZ panels on both the L and R side walls at the first reflection points (ie a total of 1200mm x 1200mm x 300mm on both sides)
- 200mm thick corner straddling panels (600mm width) spanning both the front-wall / ceiling corner, rear-wall / ceiling corner


The insulation I've used is polyester and has a published GFR of 7994 Mks rayl/m.

The room is sounding good and shows marked improvements on REW (however I am likely to cover some of the corner traps with plastic to bring a little life back into the room).

So far so good.

However, I've just installed the cloud which is 1800mm x 1200mm x 300mm with a 200mm gap from the ceiling. I used a local equivalent of pink fluffy (fibreglass - published GFR of 5840).

And here's where things get confusing. I'm no expert with REW but any improvements after adding the cloud appear to be marginal at best. In fact, the right speaker now seems to have a null at around 134Hz. I'm not sure if the cloud is making things better or worse. I even swapped out the pink fluffy for the 7994 Mks rayl/m polyester to see if that made a difference but I'm struggling to see any even though at that depth the lower GFR fluffy is theoretically marginally better at absorbing at lower frequencies. I've even tried moving the speakers about a little, including right up against the front wall but if anything this exacerbates the frequency swings.

Can anyone analyse my mdat file to see if I'm going crazy? Have I fixed one problem only to reveal another?

My primary question is: Would you stick with the fibreglass or the polyester or ditch the cloud entirely?

Any suggestions would be most welcome as I'm feeling a little deflated after all the effort of building and mounting the cloud. Perhaps there are improvements overall in some of the other graphs I'm less familiar with.

Here is the Dropbox link to the Mdat file which displays L, R and L+R with no cloud, fibreglass cloud, and polyester cloud.

The attached images are from the right speaker only. The brown frequency response curve is with no cloud, and the green curve with the fibreglass cloud.

The low end tails on the waterfalls seem to change between measurements and across the day (same treatment and same mic and speaker positions) so I suspect some of these tails are noise pollution - traffic rumble etc - from outside but can't be sure.


I'm aware there are other problems in the room, especially the lack of L and R symmetry. Any advice across the board would be gratefully received!

ps - I've also posted this call for assistance on another well known forum. I hope that's OK.

[brett]

Can anyone help? What else can I post from REW to stimulate discussion?

My expectations are realistic. Given this is an existing room in a house I'm not seeking ruler-flat response, just practical incremental improvements. For example, I'm not in a position to cover the entire back wall with > 500mm fluffy as much as I'd like to!

Any suggestions would be most welcome

[brett]

I just realised I hadn't added my location to my profile. Fixed

B

[Gregwor]

Can we see some 1200px pictures of the cloud and how you built it? It does do a lot of what a cloud should do but there are anomalies in your left and right pre-cloud measurements to begin with.

Greg

[brett]

Hi Greg. Thanks for your reply

You are correct. There are significant and frustrating differences in the room response between speakers which I can't pin down.

While the room is symmetrical there are some differences in furniture - I'm considering removing the bookshelves and table on the rear right and the filing cabinet on the rear left as an experiment but I'm not sure that will make much of a difference down low, between 20hz and 300hz.

It was suggested in another forum that construction of the room could be a factor. I'm in a front corner room of an old house. The inner leaf of the room is the old style horsehair plaster. The outer leaf on the Front and RHS is weatherboards (external walls), and the outer leaf is horsehair plaster at the Back and standard (newer) drywall plaster on the LHS (these are internal walls). There is a door at the rear of the LHS and there is also a large window in the front wall (roughly 1.8m x 1.2m) but this is symmetrical. There is no insulation in either the internal or external walls. There is no real difference in the FR or Waterfall plots with the rear LHS door open or closed either.

I have tried playing with the position of the speakers. This does make a difference of course (I can save this conversation for a separate post) but does not resolve symmetry issues. I'm not sure if it's OK to post links to other forums but can link to more information and graphs if allowed, and can certainly post data and graphs directly here that may help the community help me to improve my room!

Regarding the cloud, it's an open frame construction and is three 1200mm x 600mm panels for a total surface area of 1200mm x 1800mm. The insulation is officially 300mm thick, although it's actually closer to 250mm thick as it's compressing slightly under it's own weight, and has a published GFR of 5840. I have wrapped the insulation in dacron (wadding) to prevent the fibres from escaping. The panels are hanging on chains so I can adjust the height. In the pics there's roughly a 200mm air gap.

Obviously the panels are 'naked' at the moment and are yet to be covered

Many thanks

[brett]

Any thoughts Greg?

B

[Gregwor]

Okay, so the first thing you need to do is sort out the left and right response differences. Either one of your speakers is screwed up, your furniture is throwing things off, an equipment/wiring issue or the wall sheathing is causing the issue.

In terms of the cloud causing unfavorable differences in measurements, the only thing I think could cause the issues would be the cloud framing material causing reflections and ultimately canceling out certain frequencies.

Greg

[brett]

Okay, so the first thing you need to do is sort out the left and right response differences.

Agreed

Either one of your speakers is screwed up,

I’ve swapped left and right speakers and there is no difference

your furniture is throwing things off,

Possible and I will experiment with this further

an equipment/wiring issue

While I have swapped the speakers around I haven’t swapped at the amp terminals so will try this. Good idea

or the wall sheathing is causing the issue.

Yeah… Nothing I can do about that right now…

In terms of the cloud causing unfavorable differences in measurements, the only thing I think could cause the issues would be the cloud framing material causing reflections and ultimately canceling out certain frequencies.

Surely a few sticks of 42 mm x 19 mm pine is not going to be causing dips below 300 Hz??

[Gregwor]

I’ve swapped left and right speakers and there is no difference

Did you try swapping everything including cabling as well? It could even go back as far as your D to A converters.

Surely a few sticks of 42 mm x 19 mm pine is not going to be causing dips below 300 Hz??

A 300 cycle wavelength would be 3.75 feet long and it's quarter wavelength then being 11 1/4". So maybe it's possible that the issue is caused by your sistered framing or the long length down the edge of your cloud? Just spit balling here.

Greg

[brett]

Hi Greg

I just tried swapping cabling, from amp to speakers and RME multiface to amp and there's no difference.

I reckon it's a combination of room construction and, probably most importantly, speaker position. Though why the cloud has such a negative effect on FR I still don't know

I've attached 4 FR graphs (plus overlay) with the Right speaker as close to the desk as I can get them, then sliding back along the axis of the 'triangle' 10cm at a time. Which is best?

I imagine the speaker the closest to the listener is better as it reduces the ~75Hz mode and improves the dips at ~130Hz and around 200-220Hz. I can get it looking almost as good as without the cloud but not quite

Thoughts anyone?

Cheers - Brett

[Gregwor]

Do you have any treatment behind your speakers (between the speakers and the wall)?

Greg