Attic Shaped Studio

#421

Progress has been slow because of using the room for recording, and also Christmas and new year. Here's the latest update.

I've attached the ventilation plenum to the wall vent, which has a 200mm diameter duct flange for the room ventilation. I have several options for routing this out the lower part of the soffit and to the back of the room, but which I choose will depend on the thickness of absorption on the side wall, to be determined later. I'll fit an 80mm duct flange to the top of the plenum for speaker box ventilation.

Test fit of speaker box. The main parts of the speaker shelf have been put in place, which is only attached to the knee wall at the front of the room. None of the soffit frame is attached to this wall, and the shelf and soffit frame are not connected directly at all, so there should be some degree of decoupling. The soffit structure is not finished, and more reinforcement will be possible once I know more about the cable and duct routing.

A quick question: I've made the shelf top extend almost all the way back to the rear wall. I thought the longer it was, the more reinforcement it would offer. Will this interfere with the acoustic properties of the soffit though? If so it's easy to make it shorter.

Cheers!
Jennifer

#422

no, the length is fine. except for the duct space and some access to the speaker box, this will be full of insulation so there shouldn't be any issue and in fact may be helpful in support said insulation so it's not compacted much.

#423

Thanks Glenn, that's a good benefit it has in preventing the insulation from getting compacted, I hadn't thought of that.

Cheers!
Jennifer

#424

depending on the assembly, i'll add layers of chicken wire to support about 2ft per layer as a reasonable method to prevent packing.

#425

I've built the shell of the ventilation supply plenum from 18mm OSB and done a test fit.

It fits entirely within the soffit structure and does the following;

1) Routes the airflow from the grille in the knee wall to the side of the soffit.
2) Hopefully reduce fan rumble.
3) The lid can be unscrewed to allow the whole box to be easily removed for (hopefully never required) access to cables running over, under, and alongside plenum.

It's really just an extension of the silencer behind the wall. It adds 4 more 90° turns and unlike previously, avoids having the inlet and outlet holes in the same plane. According to Rod Gervais book this should help the sound attenuation.

Although the fan rumble on the existing setup met noise requirements for a control room, it still bothered me a bit, so more reduction would be welcome. By ear alone I'd say it has already offered a significant reduction, even though I still have a seal the joins, fit gaskets, and put the duct liner in.

I sized the baffle (marked green) so the cross section over it is about 200cm2, smaller than the silencer cross section, but I'm happy to make that tradeoff in reduced airflow if it reduces the noise. I can reduce the baffle height if airflow is too restricted. The outlet to the room will be about 380cm2.

Do you think adding enough mass to double the surface density of the plenum walls would offer a worthwhile further reduction than single walls? I have some MLV offcuts left over from the door project I could use for this.

What a mess of cables! I can arrange them so the mains supply is well away from the audio cables. Should signal cables such as HDMI, USB, midi, coax SPDIF also be kept away from audio cables? All audio cables are balanced.

Cheers!
Jennifer

#426

the MLV could be useful for the mass as well as some damping of the cabinets forming the duct.

in general, parallel runs of high freq protocols with electrical an add noise to the electrical. so spacing at least a 750mm apart from the electrical - although running those low power cables together should be ok. the low signal audio cables should be kept separated as well both electrical and high freq. metal conduit for low signal audio cables can help but shouldn't be needed for electrical or high freq.

of course you can tape a bunch together for testing and see if you actually experience noise from their proximity. a lot of properly shielded and grounded cables can co-exist.

#427

Thanks Glenn, very useful information. Cheers!

#428

I'm trying to discover if having the monitors bolted to the wall will put more sound into the house structure than the current setup, which is the monitors on Genelec isopods on speaker stands.

To this end I compared a monitor bolted to its shelf within the soffit, to the monitor sitting on the Genelec isopod on the same shelf.

Monitor on isopod.

Monitor bolted to shelf.

I did loud REW sweeps through the monitor in both setups and measured them with the mic in the room below. Black is the isopod, red is bolted to shelf.

My conclusion:
- Below 150Hz they're pretty much the same.
- Between 150Hz and 400Hz the isopod may be offering some reduction.
- Not shown, but above 400Hz there is no effect.

On the other hand, an experiment like this is not conclusive and will be subject to error and confounders. I still have find it hard to believe that these isopods are offering so much benefit. Perhaps the effect is instead from having the speaker firmly bolted to the shelf making it more efficient, and therefore putting airborne sound out more efficiently at those frequencies?

Any thoughts welcome. If the experiment is valid, then one obvious answer would be to keep the speakers on isopods instead of bolting them in place. Running some REW sweeps of both setups within the room might provide some insight.

On the other hand, this business of floating the speakers on sorbothane to eliminate any doubt is getting tempting. I already have 8 turnbuckles suitable for building Glenn's isolation system, and they fit within the existing speaker box dimensions.

Cheers,
Jennifer

#429

the isopod would certainly (well almost certainly) provide isolation at the higher frequency unless they're specifically matched to the speakers - the whole point in the isolation is to get the natural resonance down to about 1/3rd of the lowest frequency of the speaker. so if the low end is say 39hz @ -12db from roll-off, then you'd want the isolators to be set to around 13hz.

that said, the real test would be a piezo attached to the frame to determine the vibration frequencies vs the listening space. not as accurate as a proper vibration test device but would be close enough.

#430

I should have been clearer about the SPL results from above about 500Hz in the room below, there may or may not have been an effect from the isopod, I just couldn't measure it because the sound had disappeared into the ambient noise above those frequencies.

I could feel the shelf & frame the speaker is attached to vibrating at certain frequencies. The knee wall this is fixed to, I couldn't feel vibrating at all. As mentioned before, this wall is very heavy and well damped. There's no direct coupling between those and the isolated soffit baffle structure, and indeed the vibrations didn't seem to pass through into it. This at least is good news

The side wall of the room is less massive than the knee wall, and I could feel it vibrating at certain frequencies. Almost certainly through airborne excitation.

It's worth noting that these tests were done at far louder volumes than normal monitoring levels.

I'll sleep on this and ponder.

Looking at your turnbuckle isolation pad design Glenn, you have square pads of sorbothane. Would similarly sized rings be a better fit for this application? With the egg shaped Genelec speaker, I'm thinking there would be less variation in compression over the pad with those than the squares.

Cheers,
Jennifer

#431

rings would be fine. glad that there is minimal vibration as-is.

#432

Thanks Glenn.

#433

I sealed the joints on the supply plenum, put in a gasket for where it meets the wall, and added duct liner to it. Having a layer of OSB behind the plasterboard made screwing it to the wall easy. The plenum only contacts the knee wall, it will not touch the soffit baffle structure or faces.

I made some measurements with the UMIK-1 calibrated measurement microphone by taking many averages around the mix position using real time analysis in REW.

- Black plot is ambient sound with the fan off.
- Blue is the fan on low without the plenum.
- Red is the fan on low with the plenum fitted.

In general the plenum seems to offer:

- A few dB reduction at 45Hz.
- About 8dB reduction above 85Hz

The fan noise is audibly quieter than without the plenum. The rumble around 80Hz and above was probably the most noticeable thing before, so the reduction there is very useful. The fan noise also seems to come from the centre of the front wall rather than the left corner, so that is also a bonus.

I measured with and without the front face covered in MLV and could find no significant difference.

The peak around 28Hz was also present in the ambient noise so has nothing to do with the fan. I am not aware of it so presumably it's below the threshold of hearing.

With the fan at high speed the low frequency reduction from the plenum is greater, about 6dB at 38Hz. Fascinating!

#434

quieter is betterer :-)

the thing about quieter though, is when things are quieter, you can hear more noises :-)

you might measure ~1m from the vent openings on the plenum just to get the level at the opening. as you build out the soffit, with absorption etc, then you may find it reduces a bit more on the <100hz

#435

That make sense Glenn thank you. I reckon I could get away with a slower and therefore quieter fan speed yet, so at some point I may get a variac that supports lower speeds. Or even connect it in series with the 3 way switch currently used the gives low, medium and normal speeds.

Cheers!
Jennifer

Soundman2020 - Studio Design Forum