John Sayers' Design Forum

John Sayers' Recording Studio Design Forum

A World of Experience
Click Here for Information on John's Services
It is currently Tue Jun 15, 2021 8:15 pm

All times are UTC + 10 hours




Post new topic Reply to topic  [ 649 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8 ... 44  Next
Author Message
PostPosted: Tue Feb 06, 2018 9:32 am 
Offline
Senior Member

Joined: Fri Aug 29, 2014 2:34 am
Posts: 397
Location: California USA
LOL,

That was a funny read...Alrighty, got the plastic strip measurements done and sent to you. :poke: (trying on some new emojis for size)

I've told Stuart that I'm ok with him sharing design detail as long as he is....soooo, good luck out there.


Report this post
Top
 Profile  
 
PostPosted: Tue Feb 06, 2018 12:17 pm 
Offline
Senior Member
User avatar

Joined: Thu Aug 21, 2008 10:17 am
Posts: 11938
Location: Santiago, Chile
Quote:
Alrighty, got the plastic strip measurements done and sent to you
Got it!

It's working, and at the correct frequencies, but not enough yet:

Attachment:
Frank-REW-RT---1-mil-and-6-mil-plastic-strips--initial.png


You can see that the top end has come up a little, to around 130 ms, but without affecting the mids and lows. The effect is where I wanted it, above 2k, so that's good. But not enough.

Let's add two more strips of 6 mil, one in the upper section and one in the lower, like this:

Attachment:
Frank--CA--USA-V11-S181-CL--SKP-2017--S314.plastic-strips-3.jpg


So leave a 2" gap in both cases, then put a 10" wide strip of 6mil across the top section, and a 5" wide strip of 6mil across the bottom section. That should leave a 28" section of bare insulation in the middle, roughly at head height.

That should gain us another few ms in the high end.

Do another set of L, R, and LR REW tests, and please also post a photo of your handiwork!

With that, I'll be able to figure out some slats for you, so we can start getting even more life back into the room. Have you decided what type of wood you'd like to use for your slats? It needs to be at least 5/8" thick, but 3/4" would be better. Even 1", if you feel like it. Any type of wood is fine, acoustically. Solid wood can look really nice, but plywood is an option too. I've even heard of people using ordinary 2x lumber planed down to around 1" thick. The advantage of that is that's its relatively easy to get in various standard widths: at least as 2x2, 2x4, 2x6, 2x8, and 2x10, perhaps with some others as well. Or you could buy all 2x10 (even 2x12), plane it, then rip it down to whatever widths we need. We are going to need quite broad widths in places, but we can also build up wider slats from two narrower slats butted up against each other. Since this is not going to be a tuned resonator slot wall, it is not necessary to have things sealed air-tight.

Send the updated MDAT when you can.

- Stuart -


You do not have the required permissions to view the files attached to this post.

_________________
I want this studio to amaze people. "That'll do" doesn't amaze people.


Report this post
Top
 Profile  
 
PostPosted: Tue Feb 06, 2018 2:13 pm 
Offline
Senior Member
User avatar

Joined: Thu Aug 21, 2008 10:17 am
Posts: 11938
Location: Santiago, Chile
Frank OK'd sharing some more details of the design, so here goes:

Quote:
1. MIDDLE PART
As I understand it - this is a diffuser of sorts, is that correct?
Right. It's mainly a poly-cylindrical diffuser, with a broad curve, designed to deal with low frequencies, but it is also slotted, so it is not specular at higher frequencies. So it's a "two in one" design. The main purpose is to help break up the SBIR issue, but it also has some effect on modes, and at the same time is slightly diffusive in the mids and highs, while also reflecting the top end. But that's not too useful here, as it is behind insulation. See below:

Quote:
Do the slats in the middle section match the placement of the slats in the top and bottom resonators?
Right. Basically just to make it easier for Frank to build. This center module is NOT a tuned Helmholtz device (like the other two are), so the spacing and dimensions of the slats is not critical.

Quote:
Is it placed specifically at ear level to diffuse or reflect back into the room?
Not to reflect, no: to break up the SBIR issue a little, and since the most critical part of the SBIR problem is at ear height, that's where it sits. It's a diffuser, not a reflector. It also sits behind a layer of OC-703, so any reflections coming off it won't be an issue, in the high end of the spectrum. And in the mids, it isn't specular, so reflections will be diffuse not sharp.

Quote:
2. TOP AND BOOTOM
Are these resonators identical? I assume using a superchunk of 703 in those corners would be much less effective, correct?
Yes, and yes. But actually there is still 703 in the corner, behind the top and bottom modules! Triangular supechunks if 703. The modules don't go all the way into the corner, even though it looks like they do. The superchunks are in the corners, behind the tuned devices. More on that below.

Quote:
3. GENERAL RESONAT QUESTION
When you designed these tuned resonatos - does the box have a wall inside it to match the curvature of the front or back of the slats?
There's a rear wall inside the box, yes, but it is not curved: it is flat and runs straight across. See below.

Quote:
What im trying to wrap my head around is if the shape of the enclosure is a specific design element or not.
The shape isn't as important as the depth of the cavity that the resonant slug "sees" behind the slats. It's a complicated thing to explain, since the basic math for calculating traditional resonators uses the volume of the sphere to figure out the frequency, as well as the volume of the slug in the neck. But that also uses the diameter of the neck... and with a slot resonator, there is no "diameter"! For the original equation, you need to know the mass of the air in the neck of the device, so that's why you use the diameter of the neck, and the depth of the neck. But you can't do that with a slot resonator. The "neck" basically runs across the entire cavity, so there's only the height of the slot as a parameter: there's no width that you can use to figure out the mass, because it works at the exact same frequency if you make it one inch wide or ten feet wide. So clearly, the width of the slot is irrelevant. It turns out mathematically, that the width of the cavity behind the slot is also irrelevant! It's only the dept that matters, for the same reason. In fact, you can sort of think of this as if the slot consists of an infinite number of "micro slots", all the exact same height, but just a tiny fraction of an inch wide, and those are over an infinite number of "micro-cavities", divided up in the same manner. So each "micro-slot" has it's own "micro-cavity" behind it, and that's all that it sees. So it doesn't matter how many "micro-slots" you have across the width of the device, since each one only sees it's own "micro-cavity". Therefore, the width of the slot is irrelevant to the frequency, and so is the width of the cavity. It sounds crazy, but it's true, mathematically. So each slot only "sees" the part of the cavity that is directly behind it, and only the depth and height matter, not the width.

In this case, the slots run vertically, not horizontally, but it's the same principle. Just take the above explanation and turn it sideways!

Quote:
In this case, the distance between the back of the resonator and the slats gets less the further you go from the middle. Does this have an effect of variating the resonant frequency?
Yes, and that's part of the design. Each slot is tuned according to the depth behind it. Since I just put in a flat rear wall, I adjusted the size of the slots to get the right frequency, based on the depth of the cavity behind the slot.

Quote:
My understanding is that the priciple of this box is that air resonates at a frequency corresponding to the volume of the cavity and the slat with, depth and space between them, regardless of the variating distance between the slat and the back wall. Is that correct?
Sort of, but not really! See above explanation. It's not the air in the cavity that resonates, technically: the air inside the cavity acts as a spring. It's the air trapped in the slot that resonates. That "bunch" of air held in the slot between the slats, all moves together as a single unit, for rather complex fluid dynamic reasons, and it is often referred to as a "slug" of air. So as a pressure wave hits the front of the slot, if the wave happens to be at the right frequency, it causes the slug of air to vibrate at that frequency, moving in and out of the slot something like a piston moves in and out of the cylinder in a car engine. The air in the cavity is the spring that keeps the slug in place. The slug "bounces" in and out on that spring.

Here's what happens, in slow motion: the pressure wave on the outside pushes the slug of air into the cavity a bit, and that compresses the air in the cavity, which makes it act like a spring, that pushes the slug of air back out again. Since the air has mass, it "overshoots", going beyond the front of the slot a little, but that causes a "decompression" or "rarefaction" of the air in the cavity, making it act like a spring that is being pulled apart: so it drags the slug of air back in again. If that all happens in synchrony with the sound pressure wave on the outside of the box, then the slug of air vibrates in and out very well: it is a Mass-Spring system that is resonating at its tuned frequency. If the pressure "pushes" and "pulls" that the outside sound wave applies to the slug are not synchronized with the way the slug of air and the spring of the cavity want to move, then nothing much happens: no resonance.

So the system only resonates at the frequency to which it is tuned. If it "hears" any other frequency, the slug can't move much since the "pushes" and "pulls" happen at the wrong time. It's like pushing a kid on a swing: if you push the kid just a little at the exact right moment in the swing cycle, then he goes higher and higher. but if you don't synchronize your pushes with the swing, and instead push at the wrong time, or at random times, the kid stops swinging very quickly. If you pull when he's going away from you, and push when he's coming towards you, you will STOP his motion, very soon.

OK, so we have this slug of air that is vibrating away merrily in exact synchrony with the sound wave: So how does that help? How does that improve the room sound? Surely we have to attenuate that sound, not just vibrate along with it? What next? Clearly, there's a lot of energy that has gone into making the slug vibrate (if you don't think so, go stand in front of a small kid who is swinging high, and see how hard he hits you! :) ). So all we need to do, is to take away some of that energy: out something in the way that the slug hits! If you place a thin wisp of porous absorber, such as light-weight insulation, right behind the slot, then each time the slug of air overshoots the end of the slot, it will have to move through the fibers of that insulation, and that will "slow it down" so to speak. Think of the kid on the swing dragging his feet on the ground each time he goes past the middle point of his arc, and you have the right idea. If he drags his feet heavily, then he slows down and stops swinging very quickly. So in the slot resonator, if we put in just enough insulation to "drag the slug heavily" as it rushes out of the slot, then rushes back in again, then that puts a damper on the slug, removing a lot of its energy. Think of it like friction on the air slug. The insulation "takes away" some kinetic energy from the slug. So where does that energy go? It gets converted into low-grade thermal energy (heat) in the fibers. But you won't notice it getting hot! There isn't much real energy in a sound wave anyway, and we are only removing the energy from one frequency, so there's only an infinitesimal increase in temperate. Not even measurable, really.

So that's the principle here. There's a slug of air trapped in the slot, and it is bouncing in and out of the slot on a spring, which is the air inside the cavity. It resonates at it's tuned frequency, which depends ONLY on the mass of the slug and the resilience of the spring. And since the width of the slot and the width of the cavity area always the same, they cancel out mathematically, so that only the depth and height matter.

Some people don't get this principle, and try to fill the cavity with insulation, instead of putting just a bit behind the slot. But that's a huge mistake: If you fill the cavity, that's like chaining down the legs of the kid on the swing! His legs are chained to the ground, so he can never even start swinging! No resonance like that. Filling the cavity prevents the air spring from working, so the device will never resonate.

Here's a better analogy: if there is a nice puddle of water on the ground right under the swing, and the kid drags his feet through the water, then he slows down very much. However, if you keep on pushing at the right moment, he can still swing, but it's really hard: you push him on the up cycle and he swings, but as soon as his feet hit the water, he slows down, and you have to shove him again to keep him going. That's the correct operation of a Helmholtz trap: just enough "water" or insulation to give the slug a hard time, absorbing all of the energy that it picked up on the "swing" cycle, as it "drags its feet" through the damper. But if you put the entire swing inside a swimming pool full of water (=fill the cavity with insulation), then it's never going to work! No matter how hard you push, the kid is NEVER going to swing, since there's just too much resistance.

So filling the cavity is the wrong thing to do, as it totally kills the entire Helmholtz effect.

Not sure if that helped to explain how these things work, or just confused you even more! :)

OK, now I can show you the "innards" of the device, and you should be able to understand it:

Attachment:
Frank--CA--USA-V11-S181-CL--SKP-2017--S314-slotted-poly-interior.jpg


There's still a triangular superchunk in the rear corner, which is NOT part of the device: there's large holes in the top and bottom panels of the device to expose that to the room air. And you can see the rear wall of the actual cavity, right in front of the superchunk. That's the back of the device, and it runs straight across. It is not curved to match the front, and does not need to be. What maters for tuning each slot, is ONLY the distance perpendicular to the slot from the rear face of the slat to that back wall. It does not matter if the back wall is parallel to the back of the slat, or not. That isn't important. Just the distance.

You can also see the thin, light insulation on the back of the slots, which is what does all the work here! That's the "puddle of water" that slows the swing down as the kid drags his feet through it.

But you can ALSO see a thin piece of insulation on the rear wall of the cavity, and another on half the floor of the cavity! What's that doing there, when I just told you what a terrible idea it is to fill the cavity with insulation??? !!! That's there to "de-tune" the device. Or rather, to lower the Q. The problem with all Helmholtz devices is that they are tuned very, very tightly, so you have to design and build them perfectly, with great precision, if you want to hit a particular problem in the room. Even a slight error in building the device can push the tuning off so far that it' won't actually resonate at the right frequency at all. And since room modes are also very tight, high Q, if the tuning of your device does not line up perfectly with the mode, then nothing it happens. It doesn't work. So what I did here, is to lower the Q of the device a bit, such that it is tuned more broadly (works across a wider frequency band), but at the expense of it not being so effective. Lower Q means broader frequency, but lower intensity. So it WILL work on the mode, even if the center of the tuned frequency does not like up exactly with the mode, because it will still resonate for a few extra Hz each side of that frequency, due to the insulation.

If you want to have a mental picture of how this works: the insulation on the rear wall makes the location of the rear wall "fuzzy". So the distance from the slot to the rear wall is no longer precise, and you can say that thee are several distances now, or a range of distances, and the device will resonate for all of them.

I also de-tuned it further by not having the slots perfectly parallel on the inside: the faces of the slats on either side are not parallel, so the slots are slightly wedge shaped. That messes with the slug of air a bit, forcing it to change dimensions as it moves in and out, which also messes with the frequency of the tuning.

So I deliberately "damaged" the tuning as part of the design, to ensure that it will work! That sounds strange, yes ("I broke it in order to fix it"), but it was necessary. If I would have designed it with the normal tight tuning, it would NOT have worked, since it's very unlikely that it would have been tuned spot-on for the exact right frequencies of the modes I'm trying to damp. Not because Frank is a lousy workman! Not at all: you can see the excellent quality of his workmanship. It would not have been tuned correctly because the equations assume perfect materials, perfect sizes, and air at the exact correct temperature, humidity and pressure. Since there's no such things as perfect materials (wood always has variations, warps, ripples, dents, etc.), and the weather changes all the time, it would only ever have been correctly tuned occasionally, when all the conditions where perfect. As the humidity rises and falls, the wood swells and then dries out, thus changing the dimensions an the tuning. As the air temperature, air pressure, and air humidity change, so does the speed of sound, and therefore the tuning changes. Etc. So it just makes sense to design and build the device slightly out of tune, to cover a broader range of frequencies ALL the time, even if it is less effective at actually damping.

And of course, you can see the results... :) It does work, doing what it is supposed to do. Each slot is tuned to a specific frequency (or small range of frequencies) that coincides with one of the room modes in the two horizontal axial or tangential directions, but not the vertical axials or tangentials, nor the the obliques, since those modes never see the rear corner of the room.

Quote:
After you measured the room and saw the problem you need to fix, you based the design on the measurements you took in the listeners position, correct?
Right, but because these are modal problems, they occur everywhere in the room at the same frequencies, not just at the listening position. Each mode is a standing wave, and the peaks and nulls always fall at certain locations in the room, but the wave moves through the entire room. And since the pressure peak of a standing wave is always highest at the wall of the room, that's that best place to treat them with pressure operated devices! And since the two side walls meet in the corner, by putting the device in that corner, you can be certain that it "sees" all the modes associated with those two walls. And since this is a corner control room design, it is basically square and symmetrical, so ALL of the horizontal axials and tangentials are present in those corners... In other words, even though the measurement was taken at the mix postilion, the treatment must be in the corner to be effective.

Quote:
If i was to measure in the corners themselves and design a resonator based on those measurements - would it make sense acoustically or would it be stupid?
You could measure in the corners if you want, and that would work very well too. For live rooms, I often do measure in diagonally opposite corners for this very reason. ALL modes terminate in corners, so if you put the speaker down on the floor in the far left corner of the front wall, an the measurement mic up high in the top right corner of the rear wall, then you are guaranteed of both triggering and also detecting every single mode in the room. That is, indeed, a really good way of seeing what potential modes you might have to deal with. That's for live rooms: However, this is a control room, and the most critical spot is the listening position, so What I really want to know here is: which modes are having the worst effect at the mix position? Those are the ones I want to attack most. There might be two dozen modes that show up in the rear corners, but only four or five of those will be a major problem at the mix position, and those are the ones I want to hit hardest.

Quote:
... measure in the corners themselves and design a resonator based on those measurements - ...
You need one resonator for each problem. Measuring in the corners will show up dozens of problems. There isn't enough space in a room to build a resonator tuned to each of two dozen modes, so you have to "pick and choose" the worst ones, and deal with them, so go with the ones that are affecting the mix position most. In theory, each resonator needs to have an internal volume of about 1% of the total room volume, so clearly if you wanted to hit 30 modes, you'd need to use up 30% of the total room space to do that! :shock: Not feasible. So it is far better to go after all the modes with general broad-band bass traps that hit all frequencies to some degree, then use a very small number of specifically tuned devices to deal with only the very worst issues. The broadband absorption in this case is the sueprchunk in the rear corner, as well as the hangers in front of that, between the tuned devices. Those are undoubtedly also having an effect on the same problems that the tuned devices deal with, for sure! But the devices are tuned to the five specific issues that I suspected were going to be the most stubborn and hardest to subdue. So for those five specific frequencies, I'm getting the combination of tuned absorber, plus superchunk, plus hangers. I'm killing them in multiple ways at once, from every angle I can think of!

And the strategy seems to be working, as you can see from the first set of "before" and "after" graphs... :)

- Stuart -


You do not have the required permissions to view the files attached to this post.

_________________
I want this studio to amaze people. "That'll do" doesn't amaze people.


Report this post
Top
 Profile  
 
PostPosted: Tue Feb 06, 2018 3:29 pm 
Offline
Senior Member

Joined: Fri Aug 29, 2014 2:34 am
Posts: 397
Location: California USA
Okay I'll get this finished up by tomorrow evening and get an mdat over to you.. won't be able to get back to it till tomorrow afternoon my time. I'll post some pics along with some ideas that I'm thinking about.


Report this post
Top
 Profile  
 
PostPosted: Wed Feb 07, 2018 2:36 am 
Offline
User avatar

Joined: Wed Nov 22, 2017 1:15 am
Posts: 25
Location: Moscow
Hey boys,

What a great reply! Thanks for getting it all layed out for me :)

Soundman2020 wrote:
The superchunks are in the corners, behind the tuned devices.

Yes, I saw that when I read the whole thread. I assume that superchunk this small isnt doing much in the low end, correct?

Soundman2020 wrote:
What maters for tuning each slot, is ONLY the distance perpendicular to the slot from the rear face of the slat to that back wall. It does not matter if the back wall is parallel to the back of the slat, or not. That isn't important. Just the distance.

WOW.. So basically the size of device isnt that much important, correct? I guess a larger device would still be more effective because it would cover more volume for sound to be absorbed, but generally then it doesnt matter if you make the slat 0.5 meter or 2 meter long - it would only resonate to the calculated frequency, right?

Another thing: say I would deisgn a square multi-band resonator and want to target several frequencies at the same time with this device. What would make more sense to choose: build a fixed frontal dimension unit and put differentiated slats inside the cavity or a fixed depth device with the slats variating from the front? Also, when calculating the frequency of such device, how important are the slats that stand next to each other? I have seen one thread where John used a design of paired variable slots. Do the slats that stand next to one another have to be identical for the calculation to be correct? Did you implement such principle or all of the slats are different?

Soundman2020 wrote:
You can also see the thin, light insulation on the back of the slots, which is what does all the work here!

Is this 703 or an even less dense product?

Soundman2020 wrote:
But you can ALSO see a thin piece of insulation on the rear wall of the cavity, and another on half the floor of the cavity!

THat is actually very smart.. How much effectiveneess do you sacrifice by doing this?

Soundman2020 wrote:
So it is far better to go after all the modes with general broad-band bass traps that hit all frequencies to some degree, then use a very small number of specifically tuned devices to deal with only the very worst issues.

So basically after you measure the room, you pick the deepest dips in RT60, waterfalls, spectrograms etc in the low end and then design an multiband low absorber to those frequencies and a wider absorber for mid lows and mids, right?

And in the end do diffusion and reflection for the highs, right? I saw you advised Frank to stick some film to the insulation. As I understand, this will eventually be covered with fabric of some sort. I saw some guys here cover every surface of their studio with film before covering it with cloth. I assume it is not the best thing to do as the studio would eventually become too reflective in the high end, correct?

You both guys are great :) This thread is definitely helpful in wrapping my head around certain things..

Thank you Stuart for your time and effort and thank you, Frank, for being kind sharing your design details :)

This is a great community!
Cheers,
D

_________________
Vladimir said it - Vladimir did it.


Report this post
Top
 Profile  
 
PostPosted: Wed Feb 07, 2018 6:33 am 
Offline
Senior Member
User avatar

Joined: Thu Aug 21, 2008 10:17 am
Posts: 11938
Location: Santiago, Chile
Quote:
Yes, I saw that when I read the whole thread. I assume that superchunk this small isnt doing much in the low end, correct?
At 19" along the two 90° sides, it isn't exactly small! Nearly 16" deep in the middle (tip of rear corner to front face). Don't forget there's stud framing behind that, which also has insulation in the bays... it all adds up. It's not as deep as a pure superchunk, no, but still plenty deep.

Quote:
WOW.. So basically the size of device isnt that much important, correct?
The size (in terms of total internal volume) actually is important, since that defines how effective the device can be. If this unit were only 1" tall, it would do practically nothing. But between them, the combined units are nearly 4 FEET tall, in order to have as much effect as I needed.

Quote:
I guess a larger device would still be more effective because it would cover more volume for sound to be absorbed,
Right. And also more surface area on the front, and more slot area.

Quote:
but generally then it doesnt matter if you make the slat 0.5 meter or 2 meter long - it would only resonate to the calculated frequency, right?
In relation to the resonant frequency, yes, that's correct. The equation for calculating the frequency does not consider the length of the slot at all. Here it is:

F=54.59 * SQRT (height-of-hole / (height-of-volume * depth-of-volume * effective-neck-depth) ) * 1000

All of those measured in millimeters.

The last term, "Effective-neck-depth" refers to the thickness of the slats plus a "mouth correction factor". Remember I mentioned the slug of air trapped inside the slot? Well, the slug that moves in and out is actually bigger than the slot: it includes some air on either end, just outside the slot, so the "effective depth" of the slot is the thickness of the wood plus 20%, because it extends about 10% more in either direction.

As you can see, there is no term for the length of the slot in that equation: just the height of the slot, the height of the slat, the thickness of the slat, and the depth of the cavity. So the frequency will be the same for a slot 1cm long, as it is for a slot 10 m long, but the longer slot would absorb much more effectively.

If you want the FULL form of the equation, including length of the slot, then this is it:

F = 54.59 * SQRT (length-of-device * height-of-hole / (length-of-device * height-of-volume * depth-of-volume * effective-neck-depth) ) ) *1000

As you can see, there's a "length of device" in the numerator, and also a "length of device" in the denominator, so they cancel each other out. That's the mathematical benefit I mentioned yesterday: Since the slot is the same width as the cavity, the frequency does not change if the slot-and-cavity are very long, or very short: if the device is 10m long, then there would be a factor of " 10,000/10,000" in that equation, which of course works out to "1". And if the device is only 5cm long, then there would be a factor of " 500/500" in that equation, which of course also works out to "1". No matter how long or short it is, you always end up dividing the length by the length, and since the answer to that is always "1", it is irrelevant.

Quote:
What would make more sense to choose: build a fixed frontal dimension unit and put differentiated slats inside the cavity or a fixed depth device with the slats variating from the front?
It doesn't really matter too much, as long as the depth of the cavity is greater than the half-wavelength. So if you set your slats back too far, you might find that the cavity is now shorter than the half-wave, and that would be a problem.

Quote:
Also, when calculating the frequency of such device, how important are the slats that stand next to each other? I have seen one thread where John used a design of paired variable slots. Do the slats that stand next to one another have to be identical for the calculation to be correct? Did you implement such principle or all of the slats are different?


You can vary ANY of the parameters, if you want. Some people like to make slot walls where all of the slats are the same size, but the gaps between them vary. Others like to keep all the gaps the same size and use different sized slats. Others like to vary both. Other people keep both the slot and the slat the same size, and vary the depth, by angling the front with respect to the back. As long as you stick to the "rules of thumb" about designing them, then you are OK. Keep the depth large with respect to the wavelength, aim for 1% of the room volume in the cavity, keep the total open area of the front face below about 10% if you want the devices to act individually, etc.

If you have adjacent slats that are different thicknesses, then you only consider the part where there are faces opposite each other inside the hole. In other words, if the slats are both 10cm thick but one is offset to be 3cm deeper than the other, then there is only 7cm of that depth where there is a face on both sides, so that's what you would use. HOWEVER! If you do this, you also have to take into account a different mouth correction factor, and the de-tuning due to the slug NOT being able to expand as it leaves the slot, and also having friction on only one side, not the other... lots of extra stuff to think about. That's why I angled the slats here, to keep the depths roughly the same, rather than staggering them, as John does. That's harder to calculate, and my brain is lazy when it comes to complex calculations!

Quote:
Is this 703 or an even less dense product?
For this we used "pink fluffy" insulation, which Frank pulled apart even further with his fingers, to make it very light. 701 would have worked, but the thinnest you can get is 1" thick, and that's too much for here.

Quote:
THat is actually very smart.. How much effectiveneess do you sacrifice by doing this?
I didn't calculate it, (lazy brain! :) ), but I estimated that the coefficient of absorption at resonance would drop from around 0.98 to around 0.75, give or take.

Quote:
So basically after you measure the room, you pick the deepest dips in RT60, waterfalls, spectrograms etc in the low end and then design an multiband low absorber to those frequencies and a wider absorber for mid lows and mids, right?
Actually, for MOST rooms that I design, I just use hangers and very thick absorption on the rear walls, perhaps with superchunks in the corners, but for this specific case, because it is a "corner control room" design, there's not enough space to do that effectively. So for this specific room I decided to add the tuned devices (especially since I also needed to deal with SBIR), and yes, I did what you said. Looked for the four or five biggest modal issues, and tuned the device for them.

But it won't be the deepest dips that I look for: it will be the modes that ring the longest (decay the slowest). Dips MIGHT be modal, but they might also be SBIR, or reflections. I don't go after dips at this stage, usually. If it turns out that Im' left with some modal dips still at a later stage, I might go hunting for them in other parts of the room to see how bad they are, the decide what I'm going to do about it.

One word of warning: It sounds real easy to do, but in reality it is not easy at all! I do not encourage first-time studio builders to go this route: designing and using tuned low frequency devices is hard, even for experts. I much more strongly recommend using only absorption for first-time studio builders, since it is so much simpler, and really hard to get it wrong!

Quote:
And in the end do diffusion and reflection for the highs, right?
Diffusion ONLY if the room is big enough! That's a major issue, and one of my "pet peeves" that I like to rant about...

[RANT-MODE = ON]
You can only use diffusion validly in large rooms. Or at least, you can only use numerically-based diffusion in large rooms. You can't use it in small rooms, for this simple reason:
Attachment:
QRD-Diffusion-lobing--pattern-graph-SML-ENH.PNG


"Lobing" is the term. That image shows what happens to the sound waves after they have been diffused by a typical numeric diffuser, such as a QRD, PRD, Skyline, BAD, Leanfuser, etc. You can clearly see that the resulting sound field is diffuse but NOT evenly diffuse! If you sit in that field, then as you move your head from side to side, or forward and backward, or even turn your head one way or the other, your ears will end up in different parts of that field: you will be getting different intensity, phase, timing, and direction information fed to your brain from each ear, and that will change as you move your head! This is not a good situation, as you can imagine. It messes up the psycho-acoustic perception of the diffuse field, which in turn can mess up the way you perceive the direct field.

It turns out that you need a distance of at least 3m before that field becomes smooth and even, so your ears need to be at least 3m away from such a diffuser. Since that is impossible in most home studios, diffusers should NOT be used unless the room is big enough. That means at least 4m between the mix position and the actual hard, solid, rigid rear wall, which implies that the room must be at least 7 meters long....

But there's more: in addition to those 3m, you might also need even more space, if the device is tuned low. Because you also need to have a distance of at least 3 full wavelengths of the lowest scattered frequency, to be sure that you are in the even, smooth part of the field. So if you have your device tuned to a low frequency, such as for example 250 Hz for the low scatter cut-off, you would need to be MORE than 3m away, because the wavelength at 250 Hz, is nearly 1.4 m. So you would have to be 4.2 m away from that device....

Therefore, you should only use numeric-based diffusers in large rooms, where there is enough space for everyone who is doing critical listening, to be at least 3m away.

However, this fact does not seem to stop home studio builders from splattering diffusers all over their rear walls of their small rooms! You see that EVERYWHERE! Ignorant people doing silly things to their studios. The problem is that they see pictures of pro-level studios with diffusers on the rear wall and think "That looks cool! If they need those thingies, then I need them too!" So they go buy them, without realizing what damage they are doing. And of course, the manufacturers don't bother telling them that it's a bad idea! They just want to sell more diffusers, so they don't care who buys them....

Sigh!

[RANT-MODE = OFF]

Quote:
I saw you advised Frank to stick some film to the insulation. As I understand, this will eventually be covered with fabric of some sort. I saw some guys here cover every surface of their studio with film before covering it with cloth. I assume it is not the best thing to do as the studio would eventually become too reflective in the high end, correct?
Once again, it depends on what you need. There are cases where you might want foil ("film" / plastic sheeting) on most of the surfaces, but you do have to be careful since all foil is tuned! That sounds strange, but it has a tuned characteristics. Foils will reflect a certain part of the spectrum, and pass through another part of the spectrum, based on the surface density. The thinner and lighter the foil, the higher up you have to go before it reflects. That's why I asked Frank to use two different thicknesses, because I want to treat two parts of the spectrum differently. And I'm only covering a certain percentage of the insulation like that, since I want to control the AMOUNT of reflection at those frequencies. And in addition, the foil is cut into certain widths, because any object will only affect wavelengths that are smaller than it is: larger waves will go right past it, as though it was not there. The strips here are horizontal, so they reflect all of the waves that correspond to the thickness of the plastic in the horizontal direction, but only some of them corresponding to the width of the strip in the vertical direction.

It's complex to tune a room, when you want it to be very accurate, as Frank does. You can't just hang a few things on the walls because you saw them in a photo of a high-end studio, and figured you need them too! For high precision, it's a long process, done one step at a time. For someone like John, who has a lot more experience than I do(!), he can skip some of those steps by just using his knowledge and expertise gained from many, many past studios. But I like to test every step, to make sure it is working as expected, before going on to the next step. Maybe when I have as much experience as John does, I'll be able to do the same! Until them, I'm a "repeat tester"... .)

Quote:
Thank you Stuart for your time and effort and thank you, Frank, for being kind sharing your design detail
:thu:

- Stuart -


You do not have the required permissions to view the files attached to this post.

_________________
I want this studio to amaze people. "That'll do" doesn't amaze people.


Report this post
Top
 Profile  
 
PostPosted: Wed Feb 07, 2018 7:44 am 
Offline
Senior Member

Joined: Fri Aug 29, 2014 2:34 am
Posts: 397
Location: California USA
OK second test results sent to you Stuart!

So what I did, was put the plastic over the covers to test because they were already up and they hold the 1" duct liner. The panels are made with 2" square lumber (which of course is really 1.5"). This gives us the ability to either have the slats over the panels or hidden inside or both. I can certainly mill lumber or look for pre sized stuff at the lumber yards. Lots of options there. Not sure whether I want to paint or stain any exposed slats yet...

I'll have a better idea once I see the slat layout.

I didn't pull the plastic tight even though it looks it. its loose.

If you want to try out any variations of the slat wall, I can tack the boards over the panels and plastic for testing before committing to a finished product.

Attachment:
20180206_132518.jpg

Attachment:
20180206_132525.jpg


You do not have the required permissions to view the files attached to this post.


Last edited by Stadank0 on Wed Feb 07, 2018 12:52 pm, edited 1 time in total.

Report this post
Top
 Profile  
 
PostPosted: Wed Feb 07, 2018 9:03 am 
Offline
Moderator
User avatar

Joined: Wed Feb 01, 2017 6:03 pm
Posts: 1501
Location: St. Albert, Alberta, Canada
This is the best thread on the forum!

I hope I'm not hijacking this thread. It is directly related to CR treatment, Helmholtz, and my question is stemmed from a comment made here.

Quote:
The size (in terms of total internal volume) actually is important, since that defines how effective the device can be. If this unit were only 1" tall, it would do practically nothing. But between them, the combined units are nearly 4 FEET tall, in order to have as much effect as I needed.


I've been planning on putting slots in the front edge of the raised floor that the couch will sit on in my control room. The extreme depth of the floor joist cavities would allow for some super low frequency tuning. But, it might only be ~5 1/2" tall. The statement above did make me question my logic. But, a little improvement here and there will add up to a big improvement in the end. Should I be researching other ways to utilize the raised floor cavity? I did try to learn as much as I could about "space coupling" as suggested by Stuart at one point, but it seems to be pretty specific to functions I cannot for the life of me relate to a floor space!

Quote:
You can only use diffusion validly in large rooms. Or at least, you can only use numerically-based diffusion in large rooms.


Non-tuned slats of wood on the walls is coming up shortly in the tuning of Franks room. These will cause specular diffusion. Is this the only type of "diffusion" that works in smaller control room type rooms? Why do "pro" designers like Wes Lachot use diffusion on the entire back wall of their designs if it is common sense that it causes havoc?

Quote:
For someone like John, who has a lot more experience than I do(!)


I wish i was around 8+ years ago on the forum when John had more of a presence on here, sure. But man, you're so awesome. He has experience, no doubt, but you also have mad skills. You're an icon and everyone here has the upmost respect for you. Thank you, on behalf of everyone for all that you do for us.

Greg

_________________
It appears that you've made the mistake most people do. You started building without consulting this forum.


Report this post
Top
 Profile  
 
PostPosted: Thu Feb 08, 2018 1:53 am 
Offline
Senior Member

Joined: Fri Aug 29, 2014 2:34 am
Posts: 397
Location: California USA
Hi Stuart,

in the interest of mapping out the rest of my week, I thought I'd ask if the soffits were going to take some time. I imagined they might....any busy work you could think of would be great!

can't wait to see the soffit detail. I'm really glad about how well the room is shaping up! :)


Report this post
Top
 Profile  
 
PostPosted: Thu Feb 08, 2018 2:52 am 
Offline
Senior Member
User avatar

Joined: Thu Aug 21, 2008 10:17 am
Posts: 11938
Location: Santiago, Chile
I'm working on slats for the rear center section right now... should be ready a bit later today. Better go buy yourself lots of 3/4" or 1" thick wood, in planks many inches wide! :)

By the way, what's the final decoration style that you want for this room? If you want to give it a rustic look, you could go with old weathered wood for the slats. Dimensions are not terribly critical here, so even chunks of tree trunk cut roughly to size would be fine. You can sometimes pick up reclaimed lumber from an old barn or farmhouse that's being knocked down, at very good prices. Examples:

Attachment:
rustic-farmhouse-lumber-ENH-SML.jpg


Attachment:
rough-hewn-stained-ENH-SML.jpg


Attachment:
rough-hewn-planks-ENH.jpg


Attachment:
Rough Cut Siding-ENH-SML.jpg


Attachment:
old-rustic-wooden-wall--ENH-SML.jpg




Attachment:
rough-log-siding-ENH-SML.jpg


Attachment:
distressed-log-siding-SML-ENH.jpg


Attachment:
Half-Log-Hewned-ENH-SML.jpg


Attachment:
Log-siding_Hewn_Cedar-ENH-SML.jpg


Attachment:
D-log-Siding-Western-Red-Cedar-PNT-SML-ENH.jpg


Attachment:
rough-quarter-sawn-logs-ENH.jpg




The rough surface, imperfections and defects can look nice, especially if you clean it up a bit and stain it, and would also be useful acoustically, as opposed to smooth surfaces.

Especially those curved log siding ones! They would be very good for avoiding specular reflections from flat planks.


Of course, if you are not into any of those styles, then get nicely square sawn, planed and sanded lumber. It's your room, so the interior decoration is totally your call!


Another wild idea: You could mount some of the planks on spacers, so they are a bit away from the cloth, and use concealed LED light strips behind them, for a really lightning effect. LED strips often come with remote controls for changing the color and for dimming, so you can set some very nice accent lighting like that, and change it to match your mood...

Anyway, stay tuned: In a couple of hours, I'll have some slat configurations for you...


- Stuart -


You do not have the required permissions to view the files attached to this post.

_________________
I want this studio to amaze people. "That'll do" doesn't amaze people.


Report this post
Top
 Profile  
 
PostPosted: Thu Feb 08, 2018 3:02 am 
Offline
Senior Member

Joined: Fri Aug 29, 2014 2:34 am
Posts: 397
Location: California USA
Hmmmm.

not sure yet on the style. probably gonna go with common pine lumber cause its cheap and I can either stain it or paint it.

I think I'll have a better idea when I see your scheme. I already have a lighting scheme and plan to have led rgb strip lighting behind the cloud as well as in the can light section in the cloud. I like your idea of led behind the lumber but probably won't be motivated to deal with the wiring issues.

it will look cool I promise. That is a must! :jammin:


Report this post
Top
 Profile  
 
PostPosted: Thu Feb 08, 2018 5:44 am 
Offline
Senior Member
User avatar

Joined: Thu Aug 21, 2008 10:17 am
Posts: 11938
Location: Santiago, Chile
OK, here's the plan. This is just for the rear corner part, not the walls (yet). We'll so how it goes with this, then decide on what to do with the sides.


Attachment:
Frank--CA--USA-V11-S181-CL--SKP-2017--S321--rear-corner-slats.jpg


Attachment:
Frank--CA--USA-V11-S181-CL--SKP-2017--S321--rear-corner-slats-2.jpg


Attachment:
Frank--CA--USA-V11-S181-CL--SKP-2017--S321--rear-corner-slats-3.jpg


Three different images of the same thing, because the dimensions and concept are not so clear form any one of those. In these images, the "wings" spread out 28" along the walls, but you could go a little more if you wanted. Maybe 30, or even 32, if that would make it easier to hit a stud behind the cloth, for anchoring the slats.

I would suggest putting those up in stages, and doing REW tests after each stage, to confirm that they are working as expected. So for example put up one third of them, starting at the top and bottom then working towards the middle, and REW test. Another third, REW test. Final third, REW test.

It doesn't look like it, but that's the equivalent of 46 square feet of slat area on there . . . I do want to make sure we are hitting the right frequency ranges without creating bad specular reflections. If we do create specular reflections, then "plan B" is to add uneven angled or curved "things" over the slats, to break then up a bit. This is targeting from around 1 kHz upwards mostly (although it will also do stuff below that), and with a bit of luck should get us up around 180 ms or so RT60. Probably not that much though. So I do want to confirm that it is working at the right frequencies and in the right range, hence the REW tests.. If we get enough effect sooner, then we can skip some of them. If we don't get enough effect, I might change the dimensions of the remaining slat that you have not yet put in at that point, or maybe add additional slats.

Let me know what you think.

- Stuart -


You do not have the required permissions to view the files attached to this post.

_________________
I want this studio to amaze people. "That'll do" doesn't amaze people.


Report this post
Top
 Profile  
 
PostPosted: Thu Feb 08, 2018 5:53 am 
Offline
Senior Member
User avatar

Joined: Thu Aug 21, 2008 10:17 am
Posts: 11938
Location: Santiago, Chile
One more view, for clarity:

Attachment:
Frank--CA--USA-V11-S181-CL--SKP-2017--S321--rear-corner-slats-4.jpg


I'd suggest putting the bottom two slats and the top two slats first (one small, on large), REW test, then add the next one in each direction, REW test, then the fourth one from the top by itself, REW test, then the last two smaller ones in the middle.

- Stuart -


You do not have the required permissions to view the files attached to this post.

_________________
I want this studio to amaze people. "That'll do" doesn't amaze people.


Report this post
Top
 Profile  
 
PostPosted: Thu Feb 08, 2018 6:13 am 
Offline
Senior Member

Joined: Fri Aug 29, 2014 2:34 am
Posts: 397
Location: California USA
OK!

So I will tack everything up in a temporary fashion. I'll do the phases of testing as suggested. You may want to go back and look at my post with panel pictures. I will have to do construction to suit the gaps between the panels. It won't be a problem. It may just vary from your drawings a bit at the angle joins. I'll head down to home depot and get the lumber right now!

I'm thinking I'm going to paint the slats black to match the LR. I think it will look sharp..I'll know better when I see the rest of the slats on the side walls. Maybe a mixture of paint and stain......hmmmm.


Report this post
Top
 Profile  
 
PostPosted: Fri Feb 09, 2018 10:41 am 
Offline
Senior Member

Joined: Fri Aug 29, 2014 2:34 am
Posts: 397
Location: California USA
Alright...Here is the next test process.

I have temporarily installed the slats and done three tests according to your instructions Stuart. New mdat also sent of course.

Test 1:
Attachment:
20180208_160052.jpg


Test 2:
Attachment:
20180208_162024.jpg


Test 3:
Attachment:
20180208_163823.jpg


You do not have the required permissions to view the files attached to this post.


Report this post
Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 649 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8 ... 44  Next

All times are UTC + 10 hours


Who is online

Users browsing this forum: lievik33, Majestic-12 [Bot] and 22 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
Powered by phpBB® Forum Software © phpBB Group