Building a soundproof box...

Longtallronnie

I've got a piece of equipment that I'm building a test rig for that I need to make as quiet as possible.

Here's the rub, it's a hazard warning signal so it's fecking loud by design - it produces a short 650Hz tone at 120.5db at 1m.

Initial thought is to build a box from 18mm plywood with all joints sealed and caulked then line it with acoustic material, fit a gasket to the lid/door and site it in our plant room which is as far from any occupied offices as possible.
There will need to be provision for a cable to enter the box but this could be sealed with acoustic putty or similar. Heat dissipation shouldn't be an issue as it's fairly low power and will only be run for very short periods at at time (less than 5min).

I've contacted a couple of suppliers but they all seem unwilling to offer advice as it's "a bit too technical" for them!

FWIW, our office also has a large workshop attached which can be noisy. We also have a pier which often has lightvessels alongside that have these warning signals fitted which are occasionally tested so we're not talking about a particularly quiet environment to start with. 

That being said, I don't want to add any more noise than necessary!

Does this sound remotely possible or am I on a hiding to nothing?




 

crosstownvamp

A friend of mine built a box so he could play sax in it. He was in a flat so .. yeah. 

It worked very well, size of a phone box, wood frame filled with 4 inches or so of rockwool and plasterboard inside and out. A felt sealed door but nothing elaborate there. I don't think acoustic plasterboard was around then, it's nearly 40 years ago now.

He'd get really hot in the closed space with a 30 watt bulb and I was surprised he didn't pass out, but if I stood outside and he blasted away it was about as loud as a very quiet radio.

Sax is loud but it's not 120db!

mrkb

Build a box within a box with an air gap (or lots of rockwool) between is better for sound absorption. And lots of mass - especially for lower frequencies that have a lot of energy.

menamestom

Longtallronnie said:

 it produces a short 650Hz tone at 120.5db at 1m.





 

It's not my 6 year old nephew is it?  If so I'd add 50mm lead lining as well.......

crosstownvamp

Don't make it a cube - shouldn't each dimension be different to stop resonance? Google says wavelength of 650 hz is 20.79 inches so avoid that or multiples of it!

ToneControl

I already built one.
Very expensive build.
Oak outer then padding them MDF inner.
With a guitar amp speaker running full whack form a 50w amp, you could just about hear something, but the guitar strings or loose coils on the output transformer made more noise.

Longtallronnie

ToneControl said:

I already built one.
Very expensive build.
Oak outer then padding them MDF inner.
With a guitar amp speaker running full whack form a 50w amp, you could just about hear something, but the guitar strings or loose coils on the output transformer made more noise.

It's for work so budget is less of an issue. I've come up with a brief plan and my boss is ok for me to just give it a go - if it doesn't work, we can put it inside another box was his answer!
Picked up the plywood today and got some 50mm foam sheet, sealant etc coming tomorrow. 
My boss is happy for me to treat it as an experiment so at least there's no great pressure if it doesn't work but I'm feeling fairly confident.

dano

How big and heavy is the thing ? How often will you need to test it ?  What stimulus and response are you testing ? 

If size permits, I'd built a small box containing the DUT, add some rock wool, then place that into a larger airtight container, with 4 pillars to support it inside.

 Then use a small pump to suck some air out, you don't have to lower the pressure that much achieve a good level of sound isolation. 

Also, given this is a work thing, consider your risk assessment before you start, and what level of sound is acceptable during  a test for the the *total* amount the user will be exposed to.

Longtallronnie

It will be for a training/test rig so we can test the control/telemetry systems associated with it. 

It’s relatively small, the box will be approximately 900mm high by 500mm. 

There’s definitely not the space or budget to go as far as you’re suggesting and it probably won’t be used often enough to justify it either. 

We already have RAMS in place for noise etc due to the proximity of workshops etc. 

dano

Longtallronnie said:

There’s definitely not the space or budget to go as far as you’re suggesting and it probably won’t be used often enough to justify it either.

That may be the case, but at 120dB I'd start at the risk assessment, see what that tells you is needed, rather than being cost driven . I've recently been on an IOSH course and have a new perseptive on testing dangerous stuff.

Longtallronnie

Whilst I get what you’re saying dude it’s really not that big of a deal atm. It’s more of a little project to justify some weekend overtime. 

We already have a noisy environment with control measures in place so this isn’t anything out of the ordinary for our workplace - I just haven’t bothered going into that level of detail here as I didn’t feel it was necessary tbh. 

I hope this doesn’t come across as rude, I honestly don’t mean it that way - just don’t wanna get derailed into the world of RAMS etc as that side will be taken care of.

Longtallronnie

So I threw some bits of plywood together, sealed the joints with acoustic sealant and lined with 50mm Class 0 foam. Still need to fit some toggle catches to the lid so that it seals tight but I decided to test it out. 

I don’t have a warning signal etc at home so I used a Bluetooth speaker playing a tone at 650Hz. 

Measured at 1m outside the box was 82db. 

When placed in the box and measured at 1m I was getting 55db. 
Not the most scientific test but a reduction of approximately 33% and I’m hoping that will improve slightly when the lid is properly clamped shut. 

Not sure how that these results will transfer to the actual kit but it’s a start. 

Danny1969

Mass stops sound. 18mm MDF or OSB covered with 12mm plasterboard is quite a lot of mass. Lot cheaper than ply or oak too

Another cheap option which I've used to record amps before is large dog crate. Put device in crate then cover with 2nd hand rugs. 

It's only frequencies below 400Hz ish that are harder to stop 

sev112

Can you 3d print 2 semi hemispheres which bounce all the noise directly back inside ?  Only 1 joint to seal (and undo rather than loads (plus the cable)

fastonebaz

Can you bury it underground too?

Kittyfrisk

fastonebaz said:

Can you bury it underground too?

Eureka! 

crosstownvamp

Longtallronnie said:

We also have a pier which often has lightvessels alongside that have these warning signals fitted which are occasionally tested so we're not talking about a particularly quiet environment to start with.

We've missed the hidden clue. Put it in oil drum and lower it off the pier.

maltingsaudio

Purely as an idea, put two of them facing each other and flip the phase on one , it should at least reduce the volume considerably 

Sporky

maltingsaudio said:

Purely as an idea, put two of them facing each other and flip the phase on one , it should at least reduce the volume considerably 

If you put a load of them in a square matrix with individual phase control you can point the sound whichever way you like. 

Really it's all been covered. Mass, airtightness, physical isolation are the things that reduce sound transmission. I'd probably wrap it in rock wool, box around that, wrap that in rock wool, another box - sealing all the joints. Fairly simple to do I'd think. 

ESBlonde

The double skinned box with stuffing in between is the best solution. But your ply box wrapped in an old sleeping bag is a cheap and cheerful substitute.

PolarityMan

Just tell him to turn down...fucking bass players