As suggested by Skirge01, I'm starting a discussion thread about the improvements and upgrades that will be financed in part by the funds contributed by the generous community of SPCR visitors.

Several new measurement tools were acquired over the last few weeks:

SpectraPLUS RTA software for the PC -- this $1000 package contains a complete SPL tool in its huge array of acoustic measurement features. Needs a high precision low noise mic.

Another B&K 2203 sound level meter with 1" mic and vibration meter kit. This was a lucky eBay purchase. It's always nice to have a backup, and I nabbed it for just $600. The SLM looks to be in excellent condition, its readings match the existing B&K 2203 SLM and the SpectraPLUS SLM within about 0.5 dB. The vibration meter attachment is really interesting. Have not had a chance to test it yet, but if it works well, it could help in the construction of the anechoic chamber -- and in vibration testing of HDDs.

Kanomax Anemomaster 6803 -- very high precision $700 anemometer kit accurate within 1% down to 40 ft/m; this is very low. The vane sensor is impressively low friction; just walking while holding it turns the vanes and gives a reading. It's the 3rd anemometer in the lab, and by far the most accurate. It was deemed necessary for more consistent fan airflow testing. You'll read about it in a new fan test methodology article that will come in a couple of weeks.

ACO Pacific 1" test mic kit -- This $2000 mic kit is the lowest priced precision mic with self noise at 10 dBA or below that I could find. The mic head has the same fitting as the B&K 1" mic head -- they are interchangeable. This mic head will be the reference to which all the others in the lab will be calibrated -- it is new, with its own individual response chart (flat to 8kHz, then within 0.5 dBA out to about 18 kHz; perfect for SPCR needs). It will be used with SpectraPLUS. No chance to do any serious work with it, but it clearly works and it's extremely quiet.

The anechoic chamber is still in the planning stages. One major change is location. I've decided after much consideration that building it in the garage will be more practical, and just as effective for noise isolation. Building a room in an existing room poses very tough technical challenges I don't want to face if I don't have to.

Photos and more to come later.

ooh, very impressive, particularly the anemometer. The vibration meter sounds like it has a lot of potential too - be interesting to see how quantitative readings correlate with your subjective scores.

I've already suggested to Cyril at the Techreport that he outsource the sound measurement part of their reviews to you. You never know, this chamber may turn into a source of income for SPCR.

Sounds like you are going to have a fairly nice test setup there. Cant wait to see the anechoic chamber, I love those things. It's always a strange feeling while working inside one.

Further thoughts on the anechoic chamber...

When you come right down to it, there's no way it can really be an anechoic chamber as it is strictly defined. #1 reason -- it's just too small, and I cannot treat all six inner boundaries of the room. The floor needs to stay firm.

Does this mean it can be a hemi-anechoic chamber? It's what most the PC big boys are using, with a reflective floor.

No, it can't. Why? Because it's not possible to keep the room anechoic to a low enough frequency. Anechoic means no echo. No acoustic reflections, no standing waves, cancellation points at various frequencies at various points in the room.

What does it take do get no echo down to say 100 Hz? A huge room (say 30x20x12') with probably about 3 feet thickness of damping material (fiberglass, typically) on all the walls.

My room is 12x10x8'. By the time the isolating floating room-in-a-room is built, it will drop to 11x9x7.5' -- add 3' of material on the walls & ceiling and it'll be 5x3x1.5' Just big enough for the cat to stretch out for a quiet snooze.

Thinking practically and going backwards -- what are the minimum dimensions needed for the room to be useful? About 8x6' w/ a 6' ceiling. That leaves space for a standard test table, and the mic, and some room between them and the walls. That means a max of 1.5' thickness of damping material can be applied to each wall and ceiling.

We'll start with that and see what we end up with. I think we can use the RTA system to determine the best positions for the mic and for the objects being measured -- the positions least affected by standing waves, cancellations, etc.

In the end, I think it's probably going to be most accurate to describe it as an reflection-reduced acoustic isolation chamber. The isolation is still the most important part.

Last weekend, Rob (the builder) and I had a detailed powwow about all the major issues and many details. By next weekend, he'll be putting in the order for most of the building materials. Before we can build, that room has to be emptied, and all the gear in it relocated. It happens to be the busiest, most densely packed test room in the lab. All this without sacrificing SPCR editorial content or too much of my life. It's a challenge... but I asked for it.

we're here to support you MikeC... and to look at the promised pictures

I certainly don't envy the position of organising it all, but I'm looking forward to seeing things progress.

Will it look anything like this... j/k. Can't wait to see it.

Actually, the bill of materials appears complete and Rob is putting the order in Monday. We should have everything delivered by next weekend.

I took a first physical step yesterday. There is a 4'x3' window in that room which faces the house next door. It's been the most direct route for external noise to get in. I blocked it up with a 1.2" thick piece of medite board. A simple operation which still took a couple of hours due to the need to minimize damage to the window and frame.

I should have done this years ago.

The effect was immediately noticeable. It's quieter. At night (when it's quieter everywhere) the new PC SLM now reads around 12 dBA in that room. The old B&K 2003 SLM seems unable to read below about 15 dBA. This may be directly related to its internal electronics noise -- ie, it's at its own bottom limit. When I turned on a fan at ~400rpm (around 4 ft away), the new PC SLM read 18 dBA -- and happily, so did the B&K. This tends to confirm my conjecture about the B&K's internal noise.

Any chances we'll get pictures of the work-in-progress?

Well, this one is easy.


Before.


After.

Those 6 black things sticking out of the wood panel are zap straps. The wood is about 2 inches larger than the window perimeter, and it has thick weatherstripping all around where it meets the wall/window, to make a good seal. Then 6 hefty UV-resistant zap straps (plastic ties) were pushed through holes on the board, hooked around the bars on the window, then back trough the wood. The zap straps were then cinched up tightly to make a good seal. Now, the air between the glass and the wood panel acts as an additional insulator, and if/when this needs to be dismantled, just cutting those 6 zap straps is enough to release the wood, without there bing any damage holes or whatever on the window, frame or wall.

Notice there's a ledge shelf across that wall, at the level of the bottom of the window. This is actually the concrete perimeter foundation of the house. It's up around 4 feet high. Sound transmission through this portion of the wall is very low. Most sound transmission from the outside occurs through the top half of that wall. All the other walls of this room are inside walls -- there's at least one room + outer wall beyond them, so sound transmission from the outside via those walls is very low.

Views from the outside.



The window can be opened from the outside -- this was deemed necessary in order to ventilate the window airspace so moisture does not build mold. It does rain a lot in Vancouver, so this is an issue in most houses.

Back to the inside -- The ledge on the window wall is about 8" deep. This 8" deep space between the ceiling and the ledge, and the two side walls will be filled with insulation. Recycled cotton batting that's a replacement for fiberglass. http://www.bondedlogic.com/ultratouch.htm Then, that window wall will end up without a ledge and we'll be ready to build the inner room-in-a-room -- with a 2~4" gap between the walls of the two rooms.

Nice house-modding Mike.

Good to see the project is started.

So... is there still a need for building this anechoic room...?

j/k, measuring without having to wait until it's evening/night will be more efficient, but it's funny to see how a small mod can make such a big difference.

From one geek to another: Awesome. The spectral display was especially interesting for the recently reviewed VelociRaptor. Very clearly illustrated the acoustic quality of the drive with and without the heatsink.

Can't wait for build progress on the isolation chamber.

frostedflakes -- yeah, SpectraPlus is very nice, gives lots of options... but its most important aspect is not easily visible. It lets you calibrate the mic for loudness level / dB at will, and even compensate for the mic's freq. response if it is not flat. The ACO Pcific is flat, however.


In case you haven't noticed, the garage is out, and the 12x10' current lab room is back in as the location for the anechoic chamber. Rob explained how to use sectional wall/ceiling assembly to build a floating room in a room that already exists.


Yes. Your second line says it all. The work needs to be done in the day as much as possible. Even test geeks need a life. And the noise isolation is not good enough to keep measuring / recording when a car is within a block or a plane is approaching or leaving Vancouver.

The 12 dBA reading illustrates the low noise measurement capability of the new mic/SLM system more than anything. It has been as quiet in that room before, but the old SLM has never given a reading below ~15 dBA.

I thought the point of the build was just for acoustic isolation. While you will want to provide some damping for reflections, it is not like you are going to be putting an aircraft engine in there. If aircraft engines are part of your plan, we need to get CraftyChicken to send me the cash so I can start building a scale model of a C130 for you .

My suggestion regarding damping - Consider the loudest device you would realistically put in the room and design the damping around that. That should save you some effort. For your situation, 1.5 foot thick of damping material seems like overkill.

No not just for it, but it's the more important of the two -- the second being reduction of reflections. This will make mic and noise source positioning less critical. You seem to have one thing a bit reversed -- it's not about keeping the noise in, but keeping it out. SPCR's self-defined "too noisy" is anything over ~30 dBA/1m -- this is already silent from outside the current room when the door is closed.

Sorry for the confusion, I guess I wasn't as clear as I should have been.

I wasn't talking about damping for acoustical isolation, I was talking about damping for reflections. In your earlier post you said:
* To eliminate the echo in a room you would need about 3 feet of damping on each wall.
* With your room, you are limited to about 1.5 feet of damping on each wall.
* "We'll start with that and see what we end up with"

The point I was trying to make was that for reflection damping, you probably don't want to "start with that". What you want to do is figure out what sort of damping you need to reduce the reflections in your room for the "~30dBA/1m" or less devices. Otherwise you could end up putting in 12 inches of damping material (assuming 6 inches of wall for acoustic isolation) when you may only need 1 or 2 inches.

It doesn't matter much whether the sound is at 100 dB or 20 dB -- the amount of echo in a given room is frequency dependent, not amplitude dependent. The amount of absorption material needed to eliminate the reflection goes up exponentially as frequency drops. At 10kHz, it might take only an inch of fiberglass on the walls to stop it from bouncing around. At 1kHz, it might take 4". At 200Hz, it takes way more. I was quoted $12,000 for acoustic wedges to eliminate echoes down to ~200Hz; they would have been 10" square in the base, and 1.8' long from the tip of the wedge to the base.

Note -- you and I might not hear the echo as easily when the level is at 20 dBA, but the test mic will pick it up w/o any problem, and the cancellations and standing waves and other effects of acoustics reflections will still be there in the room. These things do affect the SPL as well as the frequency spectrum measurements.

We'll probably start with the simplest thing we can possibly do: Just stack up 16" width batts of cotton-based in-wall insulation along the walls till they completely cover the interior walls with 16" thickness of damping, and find a way to suspend them similarly on the ceiling.

Ah, so that is what you are dealing with. I didn't know that (but its cool, learning new stuff is why I come here ).

This build just got a little more interesting...

Another update.

I've been working to empty this room for the last couple of weeks in order to get on with the soundproofing. It's involved a huge amount of manual labor. The simple fact is that this room has been the most intensive testing room of the 3 we have, and also the most packed, every wall lined with shelving and storage. Emptying has meant squeezing the contents of 3 rooms down to 2 -- necessary because once it's done, this room will only contain a table, the mic and the item being acoustically tested. It was truly a grind, and the other 2 rooms still look like a tornado came through.

The room was mostly emptied by mid-week last, and I spent the last couple of days sealing up the outside perimeter wall. That's the one with the window. The concrete block foundation wall comes up about 4 feet from the floor, and with the drywall, 2-by-4 studs etc, the overall thickness of this (lower) half of the wall is well over a foot, probably 14".

The upper half of the wall is only half that thickness, and contains no concrete, which provides much higher sound attenuation. This photo, posted earlier, showed how I sealed the window with a 1.2" thick piece of medite board. (see my June 14 post)



The photo below shows the 8" deep ledge along that wall.


Here's how I did it:


1) created framing for the new skin to be screwed into place. I used 1.25" thick pieces of wood, as well as 2x4 and 2x6 pieces for an additional center stud.




2) lined the "ledge" space with 8" thick 48"x16" pieces of Ultra Touch recycled cotton batting insulation. 7 pieces were needed. One had to be cut in half -- amazingly tough stuff, took nearly 10 minutes with a new utility knife.



3) screwed in 4 panels of 3/4" medite board to cover the 4'x10' area. I was originally going to use a single 4x10' sheet, but the 10' stuff is only available at certain commercial lumberyards.... and I because I was working mostly alone, trying to lift or handle one would be impossible. Besides, I don't have a pickup truck to transport the stuff. Even the 2'x5' pieces that were cut for me at the lumberyard are not easy to handle when you have to get them up above shoulder height -- but I'm certainly no Hercules.

Why not drywall or QuietRock? Well, both of these materials create really fine nasty dust that I want to avoid that as much/long as possible. I don't think medite board is quite as soundproof as QR, but it's pretty dense, with excellent internal damping, and there's also that 8" thick insulation.

Since I wanted the entire assembly to be taken apart relatively easily if necessary, I used no adhesives, made sure not to strip any screw heads with power drills, and applied foam insulation strips between the panels and the framing to make an acoustic seal.


This photo was taken just before I made a quick audio recording with the new ACO Pacific mic to show acoustically where the room is at this point.

Here's the recording -- I recommend you right-click and save to your own computer before playing, as it's 3 megs.

As I said in the recording, the sound attenuation in the room is good enough that I feel compelled to line the interior of the room with the 700 lbs of Ultra Touch cotton batting awaiting in the garage -- before building the room-in-a-room. If the noise attenuation is not good enough, then it's not that difficult to remove the batting to build that second inner room. But if it is good enough, then I will have saved a couple of weeks of strenuous labor. Well see. It might be done by the end of the day -- maybe around 10pm if I keep going at it all day.

Hello Mike It's nice to be able to put a face to the name at last.

It looks like your mission has begun, I wish you the best of luck, and the least of pain/injury.!

How big is that room.... It doesnt look huge to start with, and its going to be a lot smaller once you have finished - what kind of size room will you actually end up working/living in.


Andy

Andy, it's already been going for >6 years.

Actually, I'd say the reorganization needed to empty that 10'x12' room (w/8' ceiling) has been going on for about 3 weeks, maybe longer.

If the second room-in-a-room doesn't have to be built, the final dimensions will be about 9'x7' with a 6.5'~7' ceiling.

If the room-in-a-room does get built, then it will be about 8'x6' with a 6' ceiling. This is a bit smaller than ideal, but it's workable.

We'll see very soon.

Another update, later on the same Sunday --

The wire clothesline webbing for the ceiling acoustic damping is halfway done. A bit too much sag in the middle; will have to find a way to support the center. Am planning another layer below this one -- I didn't think the webbing would support more than one layer of the 8"x18"x48" batting -- there will be nearly 100 pounds on each level.



Also, just stacking these pieces will not work. Over 20 pieces are needed for a stack to reach the ceiling because of compression -- the lower ones get squeezed down to like 4" as more are stacked atop. And over time, more compression will probably shrink it further. It's also not stable.


I need to create modular frames of maybe 6 pieces each, then stack the frames atop each other. The frames will have to be rigid, yet have minimal dimensions to prevent more sonic reflections. Something like el cheapo bookcases might work... Or maybe stiff wires. Anyway, it's obviously not going to be finished tonight. Back to the drawing board for now.

BTW, the Ultra Touch batting is way better than fiberglass to handle -- you don't need gloves! -- but it's not without issues. Some fine dust puffs out when you handle them, and it causes me to sneeze. There was a haze in the room after some 25-30 pieces had been dragged in there. Maybe it's not bad for you, but it can't be good. After it's all done, I will cover the batting with some acoustically transparent cloth -- speaker cloth -- to hopefully keep that dust at bay. And keep an air purifier running in the room for a while.

Have you considered tightening another level of clothesline perpendicular to the first one to support the sag? Alternatively, a couple of wall-plugs in the ceiling might be useful as anchor-points to support the centre.

I'm a bit puzzled as to why you try to have everything ~100% reversible. Walls are quite plastic, nothing a little plaster and paint can't fix should you eventually evict the chamber again.

Yes, a perpendicular run is the obvious solution.

It's true that walls are plastic... but at this point, I want to be able to reverse my own work by myself if necessary. IE, if that medite board wall turns out to be a weak point, I can pull it apart quite easily and rebuild using Quiet Rock. And if the room-in-a-room is necessary, then all the acoustic padding and webbing needs to be easily removed without leaving holes or other damage to patch.

Over the last day and a half, I devised a simple frame that could hold 10 pieces of the blue fill (cotton batting) -- two rows of 5 so that it covers a wall area 4' wide and 6.5' high.





It's not difficult but time consuming, and like everything else involved in this project, manual labor intensive. The building materials had to be purchased, transported, and then cut, drilled, assembled etc. I've moved some 300 lbs of blue fill and wood around in small 30~60 lb lots probably 200ft for each lot. My body is aching...

Anyway enuf moaning. The room is about 50% lined -- there is nearly 300 lbs of the blue fill in there right now. I took a couple screen captures of the ambient freq spectrum in the room, with and without A weighting.


Here it is -- at 7:30pm on a weekday evening, flat, without any weighting. Note the vertical scale. The SPL at the moment of capture was 25 dB despite the fact that through most of the freq range, it's well below -5 dB. It's the low freq noise below 200Hz that's responsible for the SPL. Most of this is street traffic from 1.5 blocks away on Main St -- which I cannot hear just standing still in the room... but the mic picks it up.


The A-weighting filter cuts all the noise below 200Hz... and is much closer to what I actually hear in the room -- very little. The hollow core interior door with big gap at the bottom is an obvious weakness that has to be dealt with. Noise from the adjacent room (not that there is any) easily comes through that door, as well as some loud noise from outside the house. Note that the top of the vertical scale is now 15 dBA, and the level at most freq is below -10 dBA.

That's a crapload of manual labor.

wow....that's fascinating that 99% of the ambient noise is in the sub-200Hz region. would be neat if a measurement could be made at say, 2am in the morning (when there is minimal traffic) to definitely confirm it is noise from cars/traffic flow; but I fully understand after your Herculean labours, midnight vigils are not exactly what the doctor ordered.

Well done Mike - sounding good


Andy

Good job Mike.

Looks and sounds great.