Attic Shaped Studio

#226

Hopefully this will be the final revision of the exhaust silencer. In theory I could make the knee wall become part of the silencer above the grille; in practice I'll be doing Gareth's suggestion of adding the knee wall OSB after the silencer is in place. This means I won't be able to see that part of the silencer to check the seal is ok from the knee wall to silencer. So best to make the silencer a self contained unit.

#227

Lunch break & pondering how these silencers work. Stuart has mentioned the change of air impedance between the room, various chambers & outdoors has an effect, but that is a bit beyond my ken.

I've tried to visualise some aspects of them in simple terms using Amray. Of course sound doesn't travel in straight lines and is not completely stopped by the boundaries, and this approximation gets worse at lower frequencies. But some things will be always be true to some extent:

- the boundaries will attenuate the sound going through them
- the boundaries will reflect some sound back
- duct liner on the boundaries will absorb some sound

The sound source is the little brown circle. Interesting to see some of the sound being reflected right back, and much of it bouncing between the baffles. I can see how these will hopefully be quite effective!

As the sound bounces between these boundaries in the silencer, standing waves will appear and presumably various cancellations / additions will happen. Is it possible the silencer might be "tuned" to some extent?

#228

That's a great visualisation with Amray. It would be great if that could be added to a sticky at some point demonstrating the concept of these silencers. I tried to make sure that each baffle went slightly past the half way point (well once it had duct liner on anyway) to ensure there was no straight line path down the middle. Not sure air can flow like that, but just in case!.

Gareth

#229

You were correct to do so. Rod Gervais does this in the example in his book, and so does Gregwor's design;

#230

Supply silencer assembled and test fit of the room side grille. The grille is a "Lorient LVH44 - 200 x 200mm Intumescent Grille". In the event of fire, the fins expand and seal off the duct, preventing the spread of fire to other rooms. This is a high performance model giving a high free area percentage, low static pressure and quiet operation.

The fins are spaced well apart so on the exhaust silencer (where air is flowing in to the silencer) I'll put a low grade filter in front of this to stop flies and other debris being suck into the silencer.

The inner and outer layers for the "lid" have already been cut, ready to close it up once the duct lining is in.

#231

Loads of work done over the last two days, main things;

- Tightened up the design and determined the location of the soffit vents. Did a trial fit of the flex duct; it'll be tight on the supply side due to intrusion of a roof structure (marked "ladder" on the house plans) but hopefully ok. From inside the eaves, I drilled pilot holes in the roof soffits to guide the chap who'll be fitting them from the outside.

- Figured out location of the hatch in the room below ceiling to allow filter changes. Coving in the ceiling / wall corner in the room below means I can't have the hatch as big as would be ideal. A mock setup demonstrated that it will still be possible to open the filter box lid and change the filter though.

- Tidied up the cabling in the eaves bays for both silencers. The exhaust side bay was a mess - mains cables intruding into the space the silencer will go in to. Fortunately there was enough slack to clip them out the way. The black cables are mostly computer signal ones. They'll be routed along the bottom of the wall. The grey mains cables clipped a foot or so above floor level are for mains sockets on the knee wall. A foot or so higher than that will run the audio snakes to room 2.

Terrible photo, but there are four grey mains cables now clipped out the way.

- Cut the main parts of the exhaust silencer, and did a test fit. It works.

Here's one of the things that required change of the original silencer design. There's a dwang (?) running along above the ceiling below that reduces the available height for the silencer. There's a wall below there; will seal some of the edges to be sure of maximum sound isolation to the room below.

And finally, the test fit revealed that tolerances for the width of the silencer are going to be a little too small. The end of the bolts protruding about 20mm in to the bay are for holding a tripled set of roof rafters together. Unfortunately It's not possible to turn the bolts around so the less protruding hex end is showing. On the other side there are cable clips. The silencer width is 302mm. The smallest side to side measure in the bay is 320mm at a couple of places, but they don't align.

I'm going to reduce the silencer width by 20mm to allow easier fitting. It will be tricky enough without having to work to such fine tolerances. It means the effective inner width of the silencer from duct lining to duct lining will be reduced from 180mm to 160mm. Will increase the static pressure to the fan a little. No big deal, theory says at max speed the fan could do 15 air changes per hour, so there is plenty spare capacity.

#232

This is more a feat of engineering than simply making a silencer box, and impressive, too!

#233

Had to be me to ask didn’t it... what’s a ‘dwang’?!

#234

garethmetcalf wrote:Source of the post Had to be me to ask didn’t it... what’s a ‘dwang’?!

Scottish synonym for nogging

#235

Did a more thorough evaluation of the wiring on the exhaust silencer side and re-routed cables accordingly. These photos roughly show the plan;

- audio snakes to room 2 at the top. I'm going to run another 8 way as a spare. There are 3 XLR cables running to the landing; sometimes I record guide vocals from there for recordings when the band are playing "live" and rooms 1 and 2 are used by musicians doing "keeper" recordings.
- socket mains (grey) at mid height. I decided to put both sockets on the side closest to the wall.
- other mains along the bottom out the way of the silencer.
- computer signal cable going to their patch bays in the pattress boxes. Computer is on the landing between the rooms. 3 x USB, 2 x VGA, 2 x HDMI. These will run along the floor as some of them e.g. USB are limited to 5 metres in length. I want as much cable as possible in the room, so the runs were limited to 2 metres, which is just enough to make it from computer on the landing to patch bay.
- 7 x ADAT optical, 1 x wordclock cable, 1 x midi that can be a bit longer also routed along the floor.

I tried to keep all crucial cables as far away from mains as possible. The two black cables at the bottom close to the grey mains are optical ADATs so no problem there.

The audio snakes will go through 50mm wide ducts above the silencer for easier access and replacement. The ducts will allow additional cables to be added when required.

The is no space for a duct to run the cables going along the floor. If any of them fail, there is redundancy as all they are all duplicated. I've run an unused USB in case any of the others fail.

The outer layer for the "stepped" exhaust silencer is complete.

This is more a feat of engineering than simply making a silencer box, and impressive, too!

Thank you. I find these engineering projects are often like a spiral, you go round and round the iterations trying to bring order out of chaos. It was unsettling at first with so many variables, so many unknowns going from design to reality, but I am finally starting to feel like I have a bit of a grasp on this one.

#236

Maybe a little late, but have you considered audio over cat6 cable? 1 cable will run 4 balanced audio signals ... way fewer wires for spares, or even main.

https://www.amazon.ca/Lyxpro-Channel-Co ... 491&sr=8-5

#237

Cheers Dante. It's not too late, but running the XLR snakes through the eaves from room 1 to room 2 is one of the easier parts of this project . Most of the cable spaghetti is from computer signal, ADAT (mic preamps) & SPDIF (monitor controller) cables going from the control room to the computer on the landing between the rooms. Having the computer there means fan noise isn't a concern. The computer has an RME Hammerfall HDSP 9652 audio interface which has no analogue inputs, just digital.

Saying that, most rooms in the house have an ethernet socket, so I considered using Dante, but can't afford it just now!

#238

Finished all the woodwork for the exhaust silencer today. I've also cut the "lids" to close it off when ready. These inner layer pieces took ages to cut!

#239

endorka wrote:Source of the postThese inner layer pieces took ages to cut!

Well, you are not making a simple rectangle so it is no surprise that such an intricate shaope requires more time. It looks great so I hope you feel the time was well worth it.

#240

Starlight wrote:Source of the post
endorka wrote:Source of the postThese inner layer pieces took ages to cut!
Well, you are not making a simple rectangle so it is no surprise that such an intricate shaope requires more time. It looks great so I hope you feel the time was well worth it.

I do indeed feel it was worthwhile spending time doing it correctly. I would cut one piece to size, fit it, seal "complete" corners, then cut the next piece, and so on. Finally seal the remaining corners. Doing this allowed a fairly precise fit, and far less sealant to fill the gaps. Very satisfying.

Working out the "bank vault" corners in advance in the design to avoid overlapping joins was fascinating too. At first confusing how to do this consistently, but after the light bulb went on in my head, the way it is done has an elegant simplicity.

Cheers!
Jennifer

Soundman2020 - Studio Design Forum