Trains.com

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Plywood, Underlayment and Homasote Sandwich for Sound Deadening

21691 views
16 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    October 2007
  • From: Fullerton, California
  • 1,364 posts
Posted by hornblower on Thursday, June 30, 2011 3:53 PM

BobH13,

You read correctly.  With the exception of a 2 by 3 frame wall down the center of my peninsula, all of my open grid framing was created by ripping sheets of 1/2" hardwood plywood into the various framing members.  I am able to purchase this high grade of plywood for about $26 per sheet.  I can rip 12 pieces just shy of 4 inches wide out of a single sheet.  Although plywood can warp in the face to face direction, it is far more stable in the edge to edge direction than dimensional lumber.  This makes it an excellent material for creating spans.  I mentioned that I have spans as long as 13 feet.  Obviously, any length greater than 8 feet long had to be doubled with staggered joints.  Instead of using plain butt joints, I cut the ends of these pieces at a 30 degree angle to give me a much longer joint and a correspondingly greater surface area for the glue to adhere to.  I did cheat a little on the longer spans by purposely building a slight crown into these spans.  My thinking was that the crown would level out as the benchwork settled.  Well, its been up about a year now with tools, supplies, and full gallon paint cans sitting on the decks and the crowns are still there.  Far less weight will be on these sections once I clear off all the junk and complete the layout. 

One thing to keep in mind when using plywood is that it doesn't hold edge driven nails or screws as well as dimensional lumber.  However, I solved this problem by adding 2 by 2 glue blocks in one corner of each 90 degree joint. 

The 2 by 3 frame wall supporting the peninsula was made by carefully ripping well seasoned 2 by 4's down to a true 3 inches wide (from 3 1/2").  I used a long, straight piece of oak screwed to my ripping fence to make sure I eliminated any crown that existed in the stock 2 by 4's.  The wall came out dead nuts straight but even the "well seasoned" studs have warped a little over the past year.  In comparison, I have observed no movement in the plywood framing.  The 22 1/2"  inch deep decks are double cantilevered off each side of this peninsula frame wall so there is nice clear storage under the layout. 

I did notice that the bare frame wall would twist a little at first.  However, my architect father advised that I add shear strength to the frame wall by gluing my 1/4" thick drywall backdrop panels to the bare studs.  Now the peninsula won't budge. 

My deck surfaces consist of 0.187" (3/16") plywood panels.  I know this flies in the face of tradition but because I was planning to cantilever everything, I wanted to keep materials light.  All of the cantilever framing members are spaced 16" on center and extra supports were glued in midway between, and parallel to,  the backdrops and the fascia panels.  Thus, no portion of the plywood decks spans an area larger than 16" by 12".  The 3/16" plywood is plenty rigid over such a short span.  Gluing the plywood to the open grid framing also increased the rigidity of the plywood decks.

Yes, I need to pay a bit more attention toward isolating the vibration in my tracks from the plywood decks, but so far everything is working out just fine.

 

Hornblower

  • Member since
    December 2008
  • From: Heart of Georgia
  • 5,406 posts
Posted by Doughless on Wednesday, June 29, 2011 9:26 PM

Thanks to both Fred and Hornblower for their responses.

I understand the practicalities of using alternative materials for layout construction.  Personally, while it sounds to me that 3/4 " ply and 2x4's would tend to vibrate less than 1/2" ply and 1x4's would if used over the same span with the same method of attachment to the grid, I much prefer working with the thinner materials.

It also seems that Hornblower has not used traditional methods, such as screws or nails, to attach any benchwork or subroadbed together.  Its been my experience that caulk holds wood  pretty well, given the minimal stresses that are generally placed on stationary benchwork anyway.  If a modeler intends on having a mobile layout, maybe not then.

It also seems that eliminating gaps in the caulk or glue would eliminate some vibration as well. Simply "beading" the caulk or glue over a surface, then clamping the surfaces without spreading the material evenly, would tend to cause gaps and allow for drumming to occur where the air gap is, even though the gap may be very small.

- Douglas

  • Member since
    February 2011
  • 117 posts
Posted by BobH13 on Wednesday, June 29, 2011 8:18 PM

Hornblower,  If I read your post correctly? you are using  1/2 ply to create the work bench and about 1/8 ply to create the top deck.. yes?

  • Member since
    October 2007
  • From: Fullerton, California
  • 1,364 posts
Posted by hornblower on Wednesday, June 29, 2011 3:58 PM

Fred,

Excellent explanations!  You certainly understand the topic.  One additional item you hit upon was the sound reflective nature of bare plywood surfaces prior to adding scenery.  Until scenery is added to a layout, it may always seem loud to the operators simply because so much airborne sound is reflecting off of the hard, flat surfaces.  Once these areas are broken up by modeled terrain, structures, ground cover and model vegitation, the sound energy tends to be dispursed at far more random angles than would occur for the bare plywood.  The result is that the layout is perceived to be less noisy. 

Doughless,

Theoretically, the more massive a structure, the more difficult it is to excite the structure with sound and vibration energy.  However, there are practical limits.  For instance, layout decks made from poured concrete would certainly be one of the most massive structures you could use.  Concrete is indeed difficult to excite with airborn sound energy.  However, monolithic concrete is far more easily excited by direct vibration than wood structures.  Based on my impact testing experience of concrete floor/ceiling assemblies, I would expect a model train running on a monolithic concrete layout to sound as though it were right next to you, no matter how far away it actually is on the layout. 

Obviously a concrete layout isn't realistic.  On the other hand, one has to consider whether building a layout using 1 1/8" plywood and 4" by 6" framing to be cost effective.  Lumber isn't cheap and we don't need to support full size trains anyway.  Proper construction techniques can make a lightweight layout structure every bit as quiet as a massively built layout.  You just have to decide which way you would rather go.  My own layout is built using 1/2" hardwood plywood ripped to the sizes necessary to create my 16" on center open grid benchwork.  Most of the benchwork is cantilevered off of the existing structure walls with no legs at all.  I have fascia spans as long as 13 feet, again with no support other than the cantilever design and the span strength of the four inch deep fascia panels.  All joints are glued together using yellow carpenters glue.  The benchwork is quite strong and very rigid, especially with the plywood decking added.  This decking is only 0.187" thick (its labeled "Project Plywood" at Lowes).  I used latex caulking to glue all of the decking to the benchwork.  Short of getting on top of the layout and jumping up and down (and why in the world would anyone do that?), my benchwork can support any realistic load expected.  All trackwork has been glued to Woodland Scenics foam roadbed using latex caulking.  The foam roadbed has also been glued to the plywood decking using latex caulking.  The loudest sounds from my layout comes from the direct sound of the metal wheels rolling on the rails, not sound drumming through the benchwork.  This sound is simply a scaled down version of the real thing.  Very little of my "plywood pacific" has received scenery so I am experiencing some of the sound reflection issue raised by Fred.  I'm sure my layout will grow even less noisy as I add terrain, structures and ground cover.

Hornblower

  • Member since
    November 2002
  • From: Colorado
  • 4,075 posts
Posted by fwright on Wednesday, June 29, 2011 2:02 PM

I'm not Hornblower, but I'll contribute some thoughts based on engineering principles.

Mass, density, and rigidity of materials all play roles in what happens to sound and vibrations at the interface.  Maximum energy transmission occurs from one medium to another when impedance (resistance to the energy) flow is equal between the materials.  So a less dense material like foam is easier for vibrations in the air to set up corresponding vibrations within the foam - the impedance of foam is much closer to air than sheet rock is to air.  When the sound in the air encounters sheet rock, more of the sound energy is reflected rather than transmitted into the sheet rock.  The same happens when the transmitted sound exits the foam or sheetrock - the greater the impedance mismatch the less sound will be transmitted to the air on the far side, with the rest going to internal reflection of the energy.

Thin rigid foam can serve as a drum head, transmitting the sound to the air under the layout more effectively than plywood.

But remember on top of the layout, the plywood (and larger wood structures) will reflect more sound than the foam will.  Since neither material absorbs much sound energy, it's a matter of where the sound energy is coming from.

Audio impedance of a material varies greatly with the frequency of the sound.  The frequency at which the impedance is lowest - easiest to excite with a sound wave - is the structure's resonant frequency.  Large mass, dense structures tend to have very low resonant frequencies, below the human audible range.  The resonant frequency for foam by itself will be higher - probably up in the human hearing range compared to 1/2" or thicker plywood - so these frequencies will be transmitted much better than the other frequencies.  1/4" plywood typically has a higher resonant frequency than 1/2" plywood - a difference that is noticeable in a boat planked with plywood.  The boat planked with 1/4" plywood will appear to be louder, and the gurgling sounds will have a higher pitch than the same boat with 1/2" plywood planking.  The reason I say "appear to be louder" is that more of the sound energy is in the frequency band that excites our ears and auditory nerves.

just my experiences

Fred W

  • Member since
    December 2008
  • From: Heart of Georgia
  • 5,406 posts
Posted by Doughless on Wednesday, June 29, 2011 12:25 PM

hornblower

 Because of its lack of mass when compared to your 1/2" plywood, the rigid foam is far easier to "excite" with sound energy.  Thus, a train rolling over the foam will tend to be amplified.  The train rolling over the plywood cannot excite the plywood and so is not amplified.  .

Hornblower,

I've been following your comments and your statement above is really interesting to me.  If I understand correctly, mass plays a very important role in limiting sound transference..

Would you say, then, in this example with all other construction and design techniques being the same, that benchwork that is constructed from 3/4" ply and 2x4's would theoretically be quieter than benchwork that is constructed with 1/2" ply and 1x4's,, because it would have greater mass?   Would 2x4 legs tend to be quieter than 2x2 legs, since they would tend to imit the vibration that is transmitted throughout the structure?  Would any difference be perceivable, in your opinion?

- Douglas

  • Member since
    January 2001
  • From: SE Minnesota
  • 6,847 posts
Posted by jrbernier on Tuesday, June 28, 2011 7:03 PM

  I am glad you mentioned a non rigid ballast connection in this discussion.  I have seen layouts with W/S foam roadbed latex glued  to plywood go for quiet to noise just by ballasting.  This my take:

Glue - White glue dries very hard and if the glue/ballast mix runs down to the plywood subroadbed - You will have noise.  Spending the money on Matte Medium for glueing  the ballast results in a softer glue/ballast mix - Less noise.  And if the ballast drips to the plywood - you have defeated your sound insulating...

  My layout uses 1/2" plywood for the subroadbed.  Homabed roadbed is glued to the plywood.  The first roadbed was glued with yellow or white glue(what I had).  I have since used latex caulk.  I use 1/2" track nails to hold the roadbed while the glue/caulk dries.  I use 'large' M-E track spikes  to hold the track to the roadbed(they do not pierce through to the plywood).  After ballasting(using Matte Medium), I pull the track spikes.  Even after plaster cloth/hard shell scenery I have not had noise issues.

  Our club layout is built using pink foam with cork roadbed and Atlas flex - all attached with latex caulk and acrylic contact cement.  We ran into the 'drumming' issue and after ballasting with real rock ballast(Arizona Rock & Mineral) and Matte Medium, the sound has been cut in half.  I suspect that and the scenery have dulled the sound.  It seems that reducing the physical attachment between the track and the subroadbed is the key issue.

Jim

Modeling BNSF  and Milwaukee Road in SW Wisconsin

  • Member since
    October 2007
  • From: Fullerton, California
  • 1,364 posts
Posted by hornblower on Tuesday, June 28, 2011 12:24 PM

markalan,

The area I'm modeling on my own layout is pretty flat so I am not using rigid foam insulation (it's difficult to get in SoCal anyway).  Thus, I don't have any first hand experience using it under model trains.  However, if the stuff reacts as I would expect it to, I would recommend gluing it down to the plywood with as few gaps between the foam and plywood as possible and using a non-hardening glue or caulking (one that remains pliable and compressable after it cures).  Silicone based glues and caulking should work well to help absorb any vibrations transferred into the foam by the trains.  Likewise, a non-hardening glue or caulk should also be used to glue the Homasote to the top of the foam to further isolate the trains from the foam.  If using track spikes, make sure they cannot penetrate beyond the thickness of the Homasote or you will short circuit the vibration isolation of the non-hardening glue/caulking.  Beyond these steps, any remaining noise would have to be accepted as a normal characteristic of the rigid foam material.

A better solution would be to use plywood or spline roadbed and use the foam insulation only to fill in the terrain between the tracks.  Leave a slight gap between the pieces of foam and the track roadbed and use a non-hardening caulking to fill these gaps prior to adding scenery. 

Also note that care must be taken during ballasting to ensure that the cured ballast/glue compound does not create a rigid connection between the track and the plywood, foam or Homasote subroadbed.  Some people recommend using matte medium as the ballast glue because it remains pliable when cured.  This is fine but an even better solution would be to lay doubled up strips of waxed paper along both edges of the roadbed prior to installing the ballast.  After the ballast/glue compound has cured, carefully slide the waxed paper out from under the ballast.  The small gap created between the ballast and whatever subroadbed you have used should effectively isolate the track vibrations from the subroadbed.

Hornblower

  • Member since
    October 2007
  • 97 posts
Posted by markalan on Monday, June 27, 2011 8:19 PM

CT

I enjoyed your comment about sound isolation mounts in submarines.  My son in law and I talked about exactly that!  He spent his 4 years in the Navy on the the USS Parche, the most decorated boat (or ship) in the history of the Navy.

 

hornblower

I appreciate your observations.  You would advocate gluing the foam to the plywood, then the homasote to the foam, correct?

 

Thanks,
markalan

  • Member since
    October 2007
  • From: Fullerton, California
  • 1,364 posts
Posted by hornblower on Monday, June 27, 2011 7:37 PM

Blind Bruce,

Rigid foam insulation has NO acoustical properties whatsoever simply because it is rigid.  The general public seems to equate insulation "R" value with acoustical efficiency.  In actual fact, open batt fiberglass insulation in the R-13 to R-19 range provides the greatest acoustical benefit (and that isn't much).  People swear by rockwool and other insulation products but insulation just doesn't provide much sound reduction. 

Resilient materials such as the WS foam roadbed and the Pergo matting mentioned are soft enough to give to pressure but also have enough memory to spring back to their original shape once the pressure is removed.  Yes, rigid foam insulation will give to pressure but will retain its new compressed shape without rebounding.  Because of its lack of mass when compared to your 1/2" plywood, the rigid foam is far easier to "excite" with sound energy.  Thus, a train rolling over the foam will tend to be amplified.  The train rolling over the plywood cannot excite the plywood and so is not amplified.  I don't know whether you glued the rigid foam insulation to the plywood.  If not, doing so might reduce the noise you're hearing.

Hornblower

  • Member since
    May 2007
  • From: East Haddam, CT
  • 3,272 posts
Posted by CTValleyRR on Monday, June 27, 2011 6:51 PM

I guess I'm just going deaf or something, because I have flextrack and Walthers Shinohara turnouts over WS foam roadbed over plaster / Sculptamold over foam or balled up newspapers.  Everything is fastened down with latex caulk; ballast is held in place with dilute matte medium.

I don't have any issues with sound being transmitted to the tabletop or elsewhere.  There is some motor hum, and a very slight "whizzing" noise which you have to be right there to hear.

For about $50K a pop, you can buy the huge sound isolation mounts that they use on submarines.  Mount your track on those, and that will remove any issues.

If that's too expensive for you, Hornblower's method will work.

Connecticut Valley Railroad A Branch of the New York, New Haven, and Hartford

"If you think you can do a thing or think you can't do a thing, you're right." -- Henry Ford

  • Member since
    January 2011
  • From: Alberta, Canada
  • 624 posts
Posted by chochowillie on Monday, June 27, 2011 6:40 PM

Am I missing something or is it because I have a significant hearing loss without my aids but for the life of me I don't understand what all the "fuss is" about a little bit of perceived excessive sound transferance from a Model RR what with all the Sound Equipped loco's plus all the rest of it.

Choo Choo Willie

CDN Dennis 

Modeling the HO scale something or other RR in the shadow of the Canadian Rockies Alberta, Canada

  • Member since
    January 2005
  • From: Winnipeg Canada
  • 1,637 posts
Posted by Blind Bruce on Monday, June 27, 2011 5:04 PM

First, I agree with all that has been said. So,, how do you explain the FACT that my trains are much noisier on that portion of the layout that is 2" foam over 1/2" plywood as opposed to the portion that is only 1/2" plywood? Both instances are Atlas flex on cork roadbed and latex caulked in place.

No screws or nails. ?????

73

Bruce in the Peg

  • Member since
    December 2009
  • 32 posts
Posted by Broken Tie on Monday, June 27, 2011 2:39 PM

I agree; keep an mechanical connections from penetrating from one strata to the next in roadbed construction.  I use the plywood/homasote roadbed "sandwich" and this is the one place I still use chemical based contact cememt; the stuff that stinks to high heaven, but grabs and holds like no tomorrow if it is used correctly.

Another thing I have found I need to be careful with is how I mount the edges of my scenery base to the roadbed.  I make sure than the scenery is attached to the plywood, not the homasote.  I use no solid covering between the plywood and homasote; this area is ballast on my railroad; and I place the ballast using as little white glue as possible to minimize 'coupling' between the two layers.  I have found over the years that this really helps to keep transference of sound to a minimum.  Lately as I have needed to rebuild some ancient roadbed, I have also been contact cementing fibreglass insulation to the underside of the plywood.  This has eliminated the final noise problems for me (the dreaded resonance); I can hear the freight car trucks click over track joints; I have nothing interfering with the sound coming from my steam locomotives.  I had not thought to use rigid foam between the homasote and the plywood, but common sense tells me this might be THE solution, provided care is taken to keep from having any mechanical joints between layers of roadbed (other than glue).

Steam may be a lot of saturated hot air, but diesel exhaust stinks.

  • Member since
    October 2007
  • From: Fullerton, California
  • 1,364 posts
Posted by hornblower on Monday, June 27, 2011 1:25 PM

First, let me state that I'm an Acoustical Engineer who deals with these matters all the time, especially footfall noise issues in multi-family buildings.  The Pergo underlayment you are using represents the right way to go about reducing noise on you layout although there are a number of similar products that will work just as well.  However, in order for you to achieve the sound deadening results you are looking for, there must never be any direct physical connection between the Homasote and the plywood base.  This means glue or caulking but no screws, nails or brads of any kind, ever!  If you are using brads or nails to mount the track, make sure they are sized so as not to penetrate beyond the thickness of the Homasote.  As soon as one nail penetrates beyond the isolation matting and into the plywood base, you have created a significant transfer path for sound and vibration.  I personally use the Woodland Scenics foam roadbed with latex caulking to mount both the roadbed and the track.  My layout uses plywood decks only 0.187" thick and yet I get NO drumming using these materials and mounting techniques. 

As far as "Sound Board" is concerned, there is no more worthless product on the market for noise control and yet, for some reason, this product continues to enjoy a glowing reputation with contractors as the "miracle" sound attenuation product.  The way I usually see this stuff used is sandwhiched inside a wall assembly where it can do no good.  It could only have some level of effect if it was left exposed to the sound source and then all it could do is reduce the reverberation levels inside the room where the sound source originates.  Sound board is not attractive and if you paint it, it loses all of it's arleady marginal acoustical potential.  As far as using it to reduce vibration, there are many superior products of much less thickness.  Don't waste your money.  Always remember that sound energy cannot be stopped effectively by absorption.  This means that insulation and foams have little effect on sound transfer.  Sound is far more effectively stopped using barriers of physical mass and isolating those masses from each other.  In this case, the Homasote and the plywood represent  the physical masses being isolated from each other by the Pergo matting.  Also keep in mind that the Pergo matting can only reduce the transfer of sound energy into the plywood.  It will have no effect on the direct sound level caused by the train wheels rolling on the rails. 

"Soundproofing" (a term I hate) a room using surface mounted insulation and/or foam materials is misleading.  Such materials can only reduce the reflected sound energy within the treated room.  Thus, the treated room sounds "dead" and this is often perceived to be less noisy.  However, little to no noise reduction results outside the room.  Don't believe me?  Just ask the kid down the street who lined his entire drum practice room with two feet of foam rubber.  He now hates practicing inside that room because it sounds so dead.  However, his neighbors still complain about his playing just as much as before he installed the foam.

Hornblower

  • Member since
    July 2003
  • From: Sierra Vista, Arizona
  • 13,757 posts
Posted by cacole on Sunday, June 26, 2011 6:53 PM

There's a sound deadening insulation product available called, strangely enough, "Sound Board" that is made in 1/2 inch thick 4x8 foot sheets.  It has better sound deadening qualities than Homasote and is also considerably cheaper and easier to cut and carve.

Sound Board is available here at Lowe's, Home Depot, and two independent lumber yards.  It's made from compressed sugar cane and other vegetable fibers.

If you go into a home improvement store here in Arizona and ask for Homasote, they give you a blank stare as if you just landed in a space ship from another planet.

 

  • Member since
    October 2007
  • 97 posts
Plywood, Underlayment and Homasote Sandwich for Sound Deadening
Posted by markalan on Sunday, June 26, 2011 6:21 PM

 

I wanted to try and minimize the transference of sound from my homasote sub roadbed to the plywood beneath it. Years ago, we installed and small section of laminate flooring over concrete in the basement. An underlayment was spread over the concrete first to level out minor dips and to cushion the flooring. I thought this might offer a good sound insulator. I did some searching and found the Pergo makes an underlayment foam 3/16” thick to be used in condo flooring to decrease the sound of footsteps on the laminate to the floor below. I spread it on top of the ply wood, placed the homasote sheets on top and then screwed the “sandwich” together using as few screws as possible.

I have no idea if this will prove to be useful. In fact, I'll never know unless I test it against a section of homasote screwed or glued to the plywood. I don't think I will ever do such a test but I have some big scraps of the underdaylayment is someone is interested in doing the testing.

 

Thanks,

markalan

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Users Online

There are no community member online

Search the Community

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!