The Drum Bubble - Leaks in my space

Ekimtoor1

Member
My room build got to the point where I could actually start testing. The initial results are disappointing, but fixable I think.

Originally I installed a window AC up on one wall which was inexpensive at $180 and did a great job of cooling the room. But in my tests it was clearly leaking a LOT. Not sure what I was thinking as it is plain to me now that a window AC can never be adequately sealed. Fortunately, I was able to return it.

My next attempt will be using a portable AC which has a single 6” exhaust duct and I think I’ll have a lot more control over that. I’m thinking about how to line the duct with MLV and I can envision a baffle device on the outside wall, also constructed of MLV. I’m not expecting perfection, but it will certainly be quieter than the window unit.

If that fails, then I’m onto a mini split which is very expensive compared to these other options. But if that’s what it takes…

The other disappointment is the door. My door is a wall section on wheels constructed just like the structure - 5/8”-green glue-5/8”-2x4 framing filled with rock wool-5/8”-green glue-5/8”. The walls are nearly soundproof and therefore so is the door. But the seal on the door is not good. I installed 2” wide strips of 1/8” MLV on the edges of the door and mating strips on the face of the structure. There are gaps of about 1/32” in different places and it is remarkable how much those gaps leak.

I have a couple of ideas of how to fix the door. One is to install hardware that would pull the door seals tight and eliminate the gaps. The other is a 6-8” “plug” for the inside of the opening (about 34”x60”) made of some sort of foam with the right properties.

Comments appreciated!
 

rob aylestone

Well-known member
The thing with home built doors is rarely the door, but the seals. I'm not a carpenter by any means and doors, door hanging and fine work like that means the door is always a weak link. My 'trick' which I've now done loads is to hang the door in the door frame, and make sure there is a bottom too, and then make up 4 pieces of timber 35x17mm seems the common size I choose and then I put neoprene seals on one 17mm edge. and with the door in it's closed position, and closer fitted, I push the vertical timber up agains the door and when the seal is right, fix the thing there. Then I add the other three and no matter my carpentry, the door seals. With your wall on wheels, what is the seal like at the floor. If it's on wheels, does that mean the seal there is poor?

One really good thing to try if you have sealing issues is to put an ordinary disco smoke machine in the room and turn it on. Then see where the smoke leaks out. That will help pinpoint the weak spots.

I also take the view that using loads of sheet material (and wasting money on green glue) and then punching a hole in it for your air is pointless. Those extra layers are wasted when you add ducts that offer escape routes to audio.

What I mean about the glue is that nobody seems to have ever built a studio, measured it, then removed the sheeting and added green glue. All the benefits of it seem to be figures produced in labs. My view on it on any surface where the sheet material is not resting on the ground, to carry the weight is that the 'technical' isolation figures it adds are compromised by fixings. Not sure about your door, but if you use your three layers with GG between, what did you use to prevent the sheets slipping against each other. A single screw through those three layers to timber underneath negates the green glue's claimed isolation effect because the screws holding it up couple the panels. The last studio Im built used the stuff, and the end product to my ears sounded no different to the similar one before it with no glue. An extra layer of sheet material seems to provide more isolation than using glue, at much lower cost.
 

Ekimtoor1

Member
The thing with home built doors is rarely the door, but the seals. I'm not a carpenter by any means and doors, door hanging and fine work like that means the door is always a weak link. My 'trick' which I've now done loads is to hang the door in the door frame, and make sure there is a bottom too, and then make up 4 pieces of timber 35x17mm seems the common size I choose and then I put neoprene seals on one 17mm edge. and with the door in it's closed position, and closer fitted, I push the vertical timber up agains the door and when the seal is right, fix the thing there. Then I add the other three and no matter my carpentry, the door seals. With your wall on wheels, what is the seal like at the floor. If it's on wheels, does that mean the seal there is poor?

One really good thing to try if you have sealing issues is to put an ordinary disco smoke machine in the room and turn it on. Then see where the smoke leaks out. That will help pinpoint the weak spots.

I also take the view that using loads of sheet material (and wasting money on green glue) and then punching a hole in it for your air is pointless. Those extra layers are wasted when you add ducts that offer escape routes to audio.

What I mean about the glue is that nobody seems to have ever built a studio, measured it, then removed the sheeting and added green glue. All the benefits of it seem to be figures produced in labs. My view on it on any surface where the sheet material is not resting on the ground, to carry the weight is that the 'technical' isolation figures it adds are compromised by fixings. Not sure about your door, but if you use your three layers with GG between, what did you use to prevent the sheets slipping against each other. A single screw through those three layers to timber underneath negates the green glue's claimed isolation effect because the screws holding it up couple the panels. The last studio Im built used the stuff, and the end product to my ears sounded no different to the similar one before it with no glue. An extra layer of sheet material seems to provide more isolation than using glue, at much lower cost.
In regards to GG, you may be right but GG does at least have lab data. I guess you’re either all in on GG or not. I did a lot of reading of both positive and negative opinions and decided if I was going to build this thing, I would use every device available within the reaches of my budget. In other words, I got sold on GG. Any more layers of drywall was not an option because of technical build problems that would create, not to mention the labor.

Having installed GG, I see how it works. It decouples the drywall sheets with this viscous, gooey layer that never cures. So you have little transmission from sheet to sheet. As for screws - what are you gonna do? The sheets have to be fastened to framing somehow, especially the ceiling as 5/8” weighs 70 pounds per sheet. GG says the sheets must be screwed, the glue itself is not suitable as a fastener. In fact it’s not glue at all, I don’t even know why they call it glue.

I gotta have AC of some kind. The window unit was a fail and I’ve already removed it and restored the wall. I hope the portable unit works out, a mini split will be triple the cost.

Right on about the door seal. I believe I can overcome this problem. The seal I described covers all four edges of the door and it does work, just not as well as I expected. I’ll figure it out. I love solving technical issues like this.

I knew this project would be an exercise in compromises and it has certainly been that.
 

rob aylestone

Well-known member
You’ve answered your own questions. All I know is that the concept of green glue is perfectly sound. However, there is no point decoupling a large sheet then compromising everything it does with rigid fixings. Rod Gervais’ book is a sort of bible, and he makes it plain what is worth doing and what fails. The studio I did with green glue performed no better Than ones without, and I’m sure it was the ceiling, where I had to fix with screws. And while in the roof space above doing the vents, my colleague in the studio could hear every screw fixing I dropped onto the upper side of three layers of green glued panels. One of the wall panels was accessible from the other side, and throwing a screw at that wall did NOT make a noise inside, so I put it down to the green glue working on the walls where there were no panel to panel fixings. All that money wasted because the roof performs far worse. Rod makes it clear in the book that you need to look at the weak points. If your windows have a certain performance, there is questionable sense in spending lots of money making the walls perform better? Sort out weak links first, and usually it is air handling, windows and doors. An extra layer or two is not that expensive on walls or ceilings. An ordinary fire door, with an extra layer on one or both sides is an effective sound door, but that quarter inch gap it needs to swing in is the killer.
 

Ed Fones

Well-known member
If you find a door seal that mysteriously doesnt peel it self off let me know, because I havent yet.

Instead of trying to seal your door why not incorporate a second door?
 

Ekimtoor1

Member
You’ve answered your own questions. All I know is that the concept of green glue is perfectly sound. However, there is no point decoupling a large sheet then compromising everything it does with rigid fixings. Rod Gervais’ book is a sort of bible, and he makes it plain what is worth doing and what fails. The studio I did with green glue performed no better Than ones without, and I’m sure it was the ceiling, where I had to fix with screws. And while in the roof space above doing the vents, my colleague in the studio could hear every screw fixing I dropped onto the upper side of three layers of green glued panels. One of the wall panels was accessible from the other side, and throwing a screw at that wall did NOT make a noise inside, so I put it down to the green glue working on the walls where there were no panel to panel fixings. All that money wasted because the roof performs far worse. Rod makes it clear in the book that you need to look at the weak points. If your windows have a certain performance, there is questionable sense in spending lots of money making the walls perform better? Sort out weak links first, and usually it is air handling, windows and doors. An extra layer or two is not that expensive on walls or ceilings. An ordinary fire door, with an extra layer on one or both sides is an effective sound door, but that quarter inch gap it needs to swing in is the killer.
I wouldn’t agree with “no point”. Sure, screws may be a compromise, but they don’t 100% render the construction worthless. In my very unlearned experience, this project is a series of compromises. I’d love 100% perfect, but I have to work in the constraints of budget, labor and physics. So what I get is what I get. It’s a done deal. It is doing what I want which is to make my drum kit unnoticeable in the house and on the street. And I do believe the GG numbers, many thousands of installs can’t all be wrong, screws be damned.
 

Ekimtoor1

Member
If you find a door seal that mysteriously doesnt peel it self off let me know, because I havent yet.

Instead of trying to seal your door why not incorporate a second door?
No room for a second door or money. All I had to work with was a 7’x9’ space in my garage, just big enough to hold my drum kit.

I used contact cement on strips of MLV and that seems to be holding well, but it’s early, I only just built the thing.
 

Ekimtoor1

Member
You’ve answered your own questions. All I know is that the concept of green glue is perfectly sound. However, there is no point decoupling a large sheet then compromising everything it does with rigid fixings. Rod Gervais’ book is a sort of bible, and he makes it plain what is worth doing and what fails. The studio I did with green glue performed no better Than ones without, and I’m sure it was the ceiling, where I had to fix with screws. And while in the roof space above doing the vents, my colleague in the studio could hear every screw fixing I dropped onto the upper side of three layers of green glued panels. One of the wall panels was accessible from the other side, and throwing a screw at that wall did NOT make a noise inside, so I put it down to the green glue working on the walls where there were no panel to panel fixings. All that money wasted because the roof performs far worse. Rod makes it clear in the book that you need to look at the weak points. If your windows have a certain performance, there is questionable sense in spending lots of money making the walls perform better? Sort out weak links first, and usually it is air handling, windows and doors. An extra layer or two is not that expensive on walls or ceilings. An ordinary fire door, with an extra layer on one or both sides is an effective sound door, but that quarter inch gap it needs to swing in is the killer.
In regard to the screws, here is an interesting product you may be aware of:

“In lab tests with traditional drywall panels, researchers claim Sound Screws reduced through-the-wall sound levels by 9 decibels, which worked out to about half the perceived sound when traditional screws were used.”

So perhaps there is a metric that can be used to approximate the effect of drywall screws on a project. 9db. The only comparable spec I could find in decibel form is a quote from GG saying on a floor install, an 8.5db reduction could be expected.

So maybe this: the GG effect may wash out the detrimental effect of screwing the drywall. Even if that’s all it did, it would still be worth it.
 

Slouching Raymond

Well-known member
You probably don't want to hear this, but this is how the Esmono doors work:
They come all assembled and hinged onto the substantial door frame.
There is a rubber seal all around the edge of the door frame, and the
substantial door handle is designed so that when you close it, the door is actively squashed
against the seal, making an almost airtight seal.
Now, if you're a cheapskate, you could get a similar effect by using a double-glazed domestic front door.
My front door has laminated double glazing, having a total thickness of 1/2" of glass, which is good for sound insulation.
It also features a squashing mechanism, to seal the door shut.
Tell them to leave out the letter box hole though.
 

Ekimtoor1

Member
Sure I want to hear it! I incorporated some similar ideas into my door. It rolls, sliding door fashion, into a series of adjustable rollers that force the seals together. The door weighs about 300 pounds so a hinge is not workable. I attached some pics. It’s still under construction, so it’s not yet pretty.
 

Attachments

  • F5352D9E-1A0B-470A-86CD-1B29E8FEF283.jpeg
    F5352D9E-1A0B-470A-86CD-1B29E8FEF283.jpeg
    2.4 MB · Views: 8
  • 93F4F91F-5B81-494B-B152-ECDB719E19B0.jpeg
    93F4F91F-5B81-494B-B152-ECDB719E19B0.jpeg
    2.1 MB · Views: 8
  • B62F2BE9-6C4F-4AE5-B655-1BB5BE34DA51.jpeg
    B62F2BE9-6C4F-4AE5-B655-1BB5BE34DA51.jpeg
    2.6 MB · Views: 8
  • B7DDFADD-0D94-4EC7-8D41-1B904A2A5B59.jpeg
    B7DDFADD-0D94-4EC7-8D41-1B904A2A5B59.jpeg
    2 MB · Views: 8
  • 59FDDE1F-C096-4D5F-8B54-EEA1D0AAD9ED.jpeg
    59FDDE1F-C096-4D5F-8B54-EEA1D0AAD9ED.jpeg
    2 MB · Views: 9
  • B3A4E3AA-70AA-4FC8-9CEE-97E48C75D94F.jpeg
    B3A4E3AA-70AA-4FC8-9CEE-97E48C75D94F.jpeg
    1.7 MB · Views: 9
  • 89B55EC9-5D35-4BAE-A567-AA42E3FC7ED3.jpeg
    89B55EC9-5D35-4BAE-A567-AA42E3FC7ED3.jpeg
    1.7 MB · Views: 9

rob aylestone

Well-known member
After a fruitless search of the internet on these remarkable screws I've noticed a few things.
1. The same test results appear everywhere
2. the same photos appear too
3. nobody seems to be selling them
4. nobody seems to be reviewing actualy cases where they have been used.
5. The so called research uses very strange language - "researchers claim" without any real data you can assess for scrutiny.

I suspect something is not quite right here.
 

Ekimtoor1

Member
I had the same suspicions but thought I would share it anyway. It would be useful to know what the db loss for drywall screws really is.
 

rob aylestone

Well-known member
Your pictures appeared while I was talking. How on earth is that going to work. The thickness of the door means it is substantial and has considerable mass, but it's virtually impossible to seal. It slides across the gap, but how are you sealing it. If it slides into the closed position, how do you then make it seal? You've spent all this time on the door but the design (to me an outsider) is flawed. The gap and seals spoil all the mass you stuck into the door.

On the screw front. They're claiming 9dB/10dB reduction - but reduction in what? Structure borne noise? What makes me think is the quantity of fixings. I've had builders in here and the sheer quantity of drywall screws the builder used on 12mm sheets is vastly more than I would use in my studio building. I find the notion that a screw, capable of suspending a sheet of drywall would be string enough without there being what? 40-50 on a 8x4 sheet? Any less and they will pull through. If you do two sheets with green glue, will that 50 be enough for the weight? Let us assume they are strong enough to support the weight. the tension on each one would surely not be pliable and soft, as it needs to be to not transmit sound? In a way, exactly how the always runny green glue works. The kick drum attempts to move the panel, the green glue absorbs it rather than transmitting it to the adjacent panel. That bit makes sense. but add 50 screws working hard against gravity and even with fancy springs surely they will transmit it. Maybe the 9-10dB is measured against the ability of a normal screw to transmit vibration. Is the 10dB 10dB less than a piece of solid steel? That is not the same as making the entire structure 9-10dB better - is it? If you suspended a panel with bed sized coil springs, it would dangle and be pretty isolated from the structure it is attached to, BUT, it would dangle down quite a way. If you made the springs thicker and shorter, then the dangle would be less but the path through more solid.

The physics of these screws seems to my mind (and A Level Physics) to have a direct link to the resilience of the device. A screw is designed to compress a joint via the screw effect. If you make the thing less resilient, how does it hold the joint tight? You could use more, but would there not be a 3dB reduction in performance when you add a second screw? , 4 screws would be 6dB less good and so on. I wonder if the spec is 10dB reduction in the single screw. Use 50 and the reduction has totally gone?

The lack of detail, and the fact that almost every internet google hit brings back the press release data, copied unedited convinces me it's a smoke oil product. In over a year, no independent reviews, comments on the available data or scientifc comment?
 

Ekimtoor1

Member
Your pictures appeared while I was talking. How on earth is that going to work. The thickness of the door means it is substantial and has considerable mass, but it's virtually impossible to seal. It slides across the gap, but how are you sealing it. If it slides into the closed position, how do you then make it seal? You've spent all this time on the door but the design (to me an outsider) is flawed. The gap and seals spoil all the mass you stuck into the door.

On the screw front. They're claiming 9dB/10dB reduction - but reduction in what? Structure borne noise? What makes me think is the quantity of fixings. I've had builders in here and the sheer quantity of drywall screws the builder used on 12mm sheets is vastly more than I would use in my studio building. I find the notion that a screw, capable of suspending a sheet of drywall would be string enough without there being what? 40-50 on a 8x4 sheet? Any less and they will pull through. If you do two sheets with green glue, will that 50 be enough for the weight? Let us assume they are strong enough to support the weight. the tension on each one would surely not be pliable and soft, as it needs to be to not transmit sound? In a way, exactly how the always runny green glue works. The kick drum attempts to move the panel, the green glue absorbs it rather than transmitting it to the adjacent panel. That bit makes sense. but add 50 screws working hard against gravity and even with fancy springs surely they will transmit it. Maybe the 9-10dB is measured against the ability of a normal screw to transmit vibration. Is the 10dB 10dB less than a piece of solid steel? That is not the same as making the entire structure 9-10dB better - is it? If you suspended a panel with bed sized coil springs, it would dangle and be pretty isolated from the structure it is attached to, BUT, it would dangle down quite a way. If you made the springs thicker and shorter, then the dangle would be less but the path through more solid.

The physics of these screws seems to my mind (and A Level Physics) to have a direct link to the resilience of the device. A screw is designed to compress a joint via the screw effect. If you make the thing less resilient, how does it hold the joint tight? You could use more, but would there not be a 3dB reduction in performance when you add a second screw? , 4 screws would be 6dB less good and so on. I wonder if the spec is 10dB reduction in the single screw. Use 50 and the reduction has totally gone?

The lack of detail, and the fact that almost every internet google hit brings back the press release data, copied unedited convinces me it's a smoke oil product. In over a year, no independent reviews, comments on the available data or scientifc comment?
Not so fast my friend! My design is pretty good, not perfect, but it will be effective. It’s not done yet. I don’t think you’ve understood my design. There is no gap, all four edges of the door seal against the face of the structure and on the edges of the door. The rollers press the seals together as the door is closed. In any case I’ve not wasted any time or money. My ears are telling me everything I need to know.

I thought the screw item was perhaps providing an approximation of db loss due to screws, nothing more. Measuring the loss screw by screw seems futile. Screws bad. I get it. What do you use in your builds?
 

Ekimtoor1

Member
Thank you. The screw issue is not one I thought about, admittedly. But I probably wouldn’t have done anything differently even if I had known about it.

Glueing double 5/8” sheet rock to the ceiling obviously won’t work. If a screw is transmitting, how much more would the much larger surface area of a glue patch transmit? Not referring to GG but standard construction adhesive which has no damping qualities.

The only way I can see around the screw problem is to do a staggered or a double wall. I didn’t have room for either of those. The cost wouldn’t have been an issue, my framing was only about $400 for my little 7’x9’ space. Just didn’t have the room. So I settled for the room-in-a-room approach and built it on the garage slab with a 2” space where it butts up to the house wall and the garage wall.

I think I’m getting a pretty good result once I finish sorting out the door seal.
 

Ekimtoor1

Member
Working on the seal the last two days and I’m giving up on the MLV strips. It’s very unforgiving, or rather it requires two near perfectly flush surfaces to create a good seal. I’m working with plain old lumber so it doesn’t have those. Gaps and friction are preventing the door from moving and sealing.

So I’m tearing out the current seal and starting over with 1/4” neoprene foam tape. That will give me a lot of give and shouldn’t care about small variations in the mating surfaces.

This will take a while.
 

jamesperrett

Active member
I've not been following this from the start so may have missed something but here are my thoughts... You can't soundproof something without some kind of compliant layer between the inner and outer wall. The most effective soundproofing layer in my opinion is an air gap. If you don't have room for an air gap you'll need to use something flexible like resilient channel - attach your resilient channel to the outer wall and then screw your plasterboard to the resilient channel. You need to use exactly the right screws to attach the board to the channel - too short and it will fall off while screws that are two long will bridge the channel and you'll lose all your soundproofing.

Green Glue works differently - it won't stop structure borne vibrations but it does stop the plasterboard vibrating quite as much as without the Green Glue. The BBC have an alternative to Green Glue - Fibreboard - which they use in their Camden partition design. This is quite an interesting paper on soundproof partition design -


It may also be worth having a look at magnetic door seals. Raven make the magnetic door seals that were used in my last studio.


For ventilation I'd suggested a ducted system with baffle boxes on the studio side to stop noise being transmitted through the walls. Even if you have air conditioning you will need some kind of ventilation because good soundproofing will be absolutely airtight.
 

Ekimtoor1

Member
Hey thanks for reading and commenting.

I researched all those things pretty thoroughly. I tossed the hat channel because I only have a 7’x9’ space. I did do the room-in-a-room so I’m not concerned with transmission into the house or out to the street. Hopefully. Tests are yet to be done. The structure is not attached to anything but the garage floor which is concrete.

I went with the highly controversial green glue between two layers of 5/8” drywall inside and out on 2x4 framing with rock wool on a slab. I used GG not because I think it’s the best way. I just wanted whatever few STC points it would give me and it was easy to install.

Air. Hmm. I remember reading about using air but literally no one I talked to recommended it for my project. It was either r13 fiberglass or safe and sound rock wool.

I’m going to try a portable AC and duct it into a baffle as you suggested. If that leaks too much then I’ll move on to a mini split which will be expensive and complicated as I will have to get permission from the HOA to place the condenser outside. But it is the most quiet option for sure.

A magnetic seal is interesting but I’m already pretty deep into my build and I’m not sure it would work on a 300 pound door.

Mike
 

rob aylestone

Well-known member
It’s certainly clever but if it slides horizontally what compresses the join? Surely at some point it must move towards the hole in the wall? I can’t see how this happens. My thinking is that if a heavy door can glide past the opening easily, then the seal thickness and material is the only barrier to sound. I’ve probably just not been able to see how you‘ve sealed this in your mechanism
 
Top