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!

Roadbed sound reduction?

15188 views
40 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    January 2013
  • From: PA
  • 481 posts
Posted by Schuylkill and Susquehanna on Saturday, January 26, 2013 4:22 PM

Thanks for all the help, but there are a couple things that might affect the sound deadening solution.

I have a few sound-equipped locomotives and hope to have the rest of my locos equipped with sound in the next year or two.  The layout is going to be partially built off the wall and partially supported by a box of 2x4s in the middle of the peninsula.  The layout also must be able to be disassembled in sections that are no bigger than about 30" by 5'.  I plan to do what was done on the Mississippi Alabama and Gulf (see the latest issue of GMR) and use 1x2 or 1x3 open grid modules over 1x3 stringers atop the brackets.  I also plan to have a train elevator (not powered) along the back side of the peninsula.

Naturally using concrete for the roadbed won't work because of weight, but it's interesting because the extreme mass must dampen vibrations.

Rubber isolators between the shelf and wall are a possibility, but the screw into the wall could transmit the vibrations as well as the shelf bracket.  If I do put rubber there, then the shelf and backdrop would have to have rubber behind them as well.

What I meant by using homosote insulators was that the roadbed would be nailed/screwed to one part of the homosote, and the rider would be nailed/screwed to another part.  Thinking about it now, I believe that it would be easier to just make the risers out of 1/2" homosote with risers every 16" or so.  The roadbed is going to be 1/2" ply as I said.  That, cork roadbed, and matte medium holding the ballast should dampen any vibrations.

Any thoughts now that I've clarified a few points?

 

Modeling the Pennsy and loving it!

  • Member since
    October 2007
  • From: Fullerton, California
  • 1,364 posts
Posted by hornblower on Thursday, January 24, 2013 8:16 PM

Wow!  So much good advice mixed in with worthless Old Wive's Tales that the poor OP won't know what advice to take.  As an acoustical engineer having spent several days recently actually measuring the vibrations from the BNSF main line through Anaheim/Placentia, I can help a little here.

There are few effective means of mitigating vibration transfer other than high mass and resilient isolation of the vibration source from the high mass.  Starting at the vibration source (the model train), there is no significant way to reduce vibration transfer between the train wheels and the track rails other than to make sure the track is as smooth as possible.  The first opportunity to isolate the vibration source from the high mass of the benchwork is to use a non-hardening caulk or glue to adhere the track to the roadbed.  By non-hardening, I mean a caulk or glue that stays soft and pliable after it has cured.  Something that stays as soft as silicone gasket sealant for automotive use would be best.  Besides a non-hardening caulk or glue, there should be no other physical and/or mechanical connection between the track and the roadbed/subroadbed.  Even the wire used to operate the points of a turnout could be a vibration path into the layout benchwork if care is not taken to isolate the swith motor from the bechwork.

Yes, it is true that ballasting the track can negate any vibration isolation you achieved with your non-hardening track/roadbed adhesive and massive or resilient roadbed material.  Using a ballast cement that remains more flexible than white glue can help but a better approach is to ensure that your ballast/glue mixture never touches the subroadbed.  When ballasting your track, place a few layers of waxed paper (or other non-stick material) along the edges of the roadbed.  A slight overlap of the roadbed edge is desirable.  Now apply your ballast and glue mix.  Once the ballast and glue mix has cured, carefully pull out the waxed paper.  This will maintain a slight gap between the ballast/glue mix and the subroadbed eliminating any chance of the ballast transferring noise and vibration into the subroadbed.

The first opportunity for high mass is the roadbed.  Using dense materials such as drywall and cement products might seem like a bit of overkill but they will definitely work as acoustic mass.  However, the subroadbed can also be used as an additional layer of isolation should you not want to work with less conventional materials.  Cork roadbed will work as an isolator but the Woodland Scenics foam roadbed is an even better isolator.  Whatever roadbed material you choose, use only a non-hardening caulk or glue to adhere it to the subroadbed.  Again, there should be no mechanical connections between the roadbed and the subroadbed other than the non-hardening caulk or glue.  If a rigid roadbed material is used (drywall, wood, homasote, etc.) care should be taken to ensure that the non-hardening glue completely separates the roadbed from the subroadbed.  Use some kind of thin shims or spacers to maintain a physical separation during the gluing process.

The next opportunity for high mass is the subroadbed.  Half inch or thicker plywood is a good choice although there are more massive building materials.  5/8" thick drywall and 1/2" or thicker MDF are both heavy materials that make good acoustical mass.  Although certainly not as strong as similar thickness plywood, drywall and MDF will work if well supported.  Other non-conventional materials such as cement board would also work but the skills needed to work with these materials could cause some modelers problems.  Using a non-hardening caulk or glue to adhere the subroadbed to the benchwork offers yet another layer of isolation between masses.  Just be sure that all nails and/or screws used to temporarily hold the subroadbed to the benchwork are removed after the caulk or glue has cured.  Any nails or screws left behind become direct vibration paths into the benchwork.  Again use thin shims or spacers to maintain a physical separation until the caulk or glue has cured.

Successful isolation of any benchwork attached to the walls of the room can be difficult.  As previously suggested, the best approach is to design the benchwork to be supported only from the floor with no connections or bracing of any kind attached to the walls.  Attempts to isolate building framing by sandwiching rubber between two pieces of wood is common but the nails or screws holding the sandwich together create direct vibration paths between the framing members.  True vibration isolation mounts are expensive and most modelers could not justify such expense.  Rubber grommets, though a little better than the rubber sanwich, do not constitute true vibration isolation mounts mainly because they tend to get compressed too much in the desire to make the assembly stop moving around.

If the benchwork must be attached to the wall studs, you can increase the noise and vibration resistance of the wall.  However, you need to throw out a lot of Old Wive's Tales to do it correctly.  If your walls are lath and plaster, leave them alone.  Any attempt to improve their noise and vibration resistance will likely make things worse.  If your walls have drywall, all mitigation should be done from the side of the wall opposite the train room.  Start by first removing the drywall from the side of the wall opposite the train room.  With the lights on in the train room and the lights off in the room your working in, look for light leaks through the trainroom side of the wall.  Fill all such leaks with caulking including those around the backside of any electrical outlets.  If the outlets are really bad, apply Lowry #10 putty pads or red fire pads tightly around the outlet boxes until no light leaks show.  Now insulate the stud cavities with R-13 fiberglass open batt insulation.  Do not pack any additional insulation into the stud cavities.  Do not use spray on or expanding foam insulations, and do not use any rigid form of insulation.  Forget about "Soundboard" as it simply doesn't work.  Increased insulation "R" values do not translate into increased noise reduction.  Thermal and "acoustical" insulations are piss poor sound barriers no matter how you slice them.  Adding some helps a little but adding more does NOTHING!  Put your money elsewhere.

Now, pull any electrical outlets in the modified side of the wall forward and remount to match the new wall face (about 1 3/4" further forward)  Now apply single leg resilient channels perpendicular to the studs at 16" on center with the stud side flange pointing toward the floor and the drywall side flange pointing toward the ceiling.  Attach a new layer of 5/8" Type "X" drywall to the resilient channels (and only to the resilient channels) using screws no more than 1" in length.  Be careful that no drywall screws make contact with the wood studs.  Leave a 1/4" gap between the new drywall and the ceiling, side walls and floor.  Now thoroughly caulk all seams between the drywall panels and fill the perimeter gap with a non-hardening caulk.  Now attach a second layer of 5/8" drywall to the resilient channels (and only to the resilient channels) using screws no more than 1 5/8" long.  Again, be careful that no drywall screws make contact with the wood studs.  Stagger the drywall panels so that you don't have second layer seams directly atop first layer drywall seams.  Leave a 1/4" gap around the perimeter of the second drywall layer.  Fill the perimeter gap and any gaps around the relocated outlet boxrs with non-hardening caulking.  Finish the drywall and paint.  From this point on, never hang anything on this wall heavy enough to require attaching directly to the studs.  Hang it somewhere else.

The above wall modifications will yield a good 10 dBA of additional noise reduction with similar vibration reductions if done exactly as described.  Do not shortcut the system by sandwiching the resilient channels between the existing and new drywall layers.  It doesn't work and I have the test results to prove it.

Hope these suggestions help.

Hornblower

  • Member since
    February 2008
  • From: Potomac Yard
  • 2,767 posts
Posted by NittanyLion on Thursday, January 24, 2013 5:34 PM

BroadwayLion

Homasote is a BRAND NAME, and Big Box Stores will not carry it by the brand name

They most certainly do.  For a wide variety of reasons, I was in three Home Depots and two Lowe's this past weekend.  All five stores stocked it and it was explicitly labeled Homasote in all of them.

  • Member since
    January 2007
  • From: Eastern Shore Virginia
  • 3,290 posts
Posted by gandydancer19 on Thursday, January 24, 2013 10:47 AM

I use the method that Jim described and I don't have any problems.  I also run ALL sound locomotives, and metal wheels in my rolling stock.

Elmer.

The above is my opinion, from an active and experienced Model Railroader in N scale and HO since 1961.

(Modeling Freelance, Eastern US, HO scale, in 1962, with NCE DCC for locomotive control and a stand alone LocoNet for block detection and signals.) http://waynes-trains.com/ at home, and N scale at the Club.

  • Member since
    July 2006
  • From: North Dakota
  • 9,592 posts
Posted by BroadwayLion on Thursday, January 24, 2013 10:28 AM

If you are using common shelving brackets to hold a layout to the wall, then you can put a strip of rubber or weather striping between the bracket and the wall. End of that part of the sound problem.

Homasote is a BRAND NAME, and Big Box Stores will not carry it by the brand name, and so their sales boys (er associates) will not know what you are talking about. As for either under-floor insulation material or sound-boards, and they will show you many suitable options. LION likes Celotex the best, but while the Celotex company is alive and well, this particular product disappeared with the advent of fire codes. What you want are light weight (Homasote is rather heavy) sheets that are dimensionally stable, easy to cut yet durable. Do not ask for Homasote, tell them you are building a train table and describe what you want: A sheet of soft, stable material. Or else just go to the Homasote website and find a dealer near you.

Noise? What is the problem with noise. LIONS *like* noise. Listen to the LION ROAR:

wtxsdwqv08

The Route of the Broadway Lion The Largest Subway Layout in North Dakota.

Here there be cats.                                LIONS with CAMERAS

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Thursday, January 24, 2013 6:44 AM

 Dunno where these rumors of them not making Homasote keep coming from. Just because Home Depot doesn;t have 4x8 sheets of something doesn't mean it's not being made, despite what some of the so-called 'experts' workign there might say.

 One definite key in any method actually working is not not have a mechanical connection joining layers. Caulk is great for this, easy to use, easy clean up, and it bonds nearly anything, but not SO strong like other adhesives that you couldn;t remove and reuse track. By not drying rock hard like other adhesives, it provides an additilally layer of different density between the materials, attenuating the sound.

 I'm still waiting shipment of a couple of items to have all the options available so I can make a test of 3 different bases with 3 different roadbeds to see which ends up the quietest.

                  --Randy


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    July 2006
  • From: 4610 Metre's North of the Fortyninth on the left coast of Canada
  • 9,352 posts
Posted by BATMAN on Wednesday, January 23, 2013 6:36 PM

Put rubber grommets in the screw holes of your brackets. And if you want to get really serious about no noise transmission, use cement board as a sub roadbed and make your roadbed out of topping cement and you will get zero noise transmission. The track here is caulked to the cement.

 ">

 ">

 ">

Brent

"All of the world's problems are the result of the difference between how we think and how the world works."

  • Member since
    January 2001
  • From: SE Minnesota
  • 6,846 posts
Posted by jrbernier on Wednesday, January 23, 2013 5:10 PM

  What you need to accomplish is stop the 'wheel to rail' sound from being transmitted to the layout 'structure' and then to the walls.

  • 1/2" plywood for subroadbed is good.  Just do not lay sheets of the stuff all over(drum head effect).  Homasote is messy to cut and the usual 5/8" product is not uniform in thickness.  I would buy Homabed from California Roadbed Company.  The product is cut to shape and sanded to a uniform .240" thickness.  Much easier to work with.  Your idea of Homasote 'pads' will not fly - Any carpenters glue/screws will transmit the sound through your pad and to the layout structure.
  • I would attach the Homasote roadbed to the plywood subroadbed with cheap acrylic caulk.  And attach the track to the roadbed with the same acrylic caulk.  Good sound deadening.  Woodland Scenics also has a foam roadbed that can be attached with acrylic caulk - Some folks swear by it, other swear at it.
  • When you ballast, do not use diluted white glue.  It hardens rock solid and will transmit the noise  from the track to the plywood.  Use something like diluted Matte Medium that remains somewhat 'soft' and will not allow the sound to be transmitted.  I have heard of white glues that remain somewhat soft, but I have never seen them.  Do not use diluted yellow carpenters glue - It will be noisy and be a 'mess' before you get done.

  I use 1/2" common birch for my subroadbed, with Homabed or cork roadbed attached to the subroadbed with acrylic caulk.  My flex track is actually nailed down, and after ballasting with Matte Medium, I pull the nails.  The only 'noise' I really hear is if the hard shell scenery(plaster wrap) becomes a 'drum head'.  Usually ground cover/trees seems to absorb much of the sound.

  Sound deadening has really not been much of an issue until 'sound' engines started getting popular.  Our club has 2-4" of foam over 1/4" plywood.  Cork roadbed is attached with contact cement as is the flex track.  There is significant 'noise' when 7-8 long trains are running!

Jim

Modeling BNSF  and Milwaukee Road in SW Wisconsin

  • Member since
    February 2005
  • From: Vancouver Island, BC
  • 23,330 posts
Posted by selector on Wednesday, January 23, 2013 5:02 PM

Sound is most greatly attentuated when it is met by two mediums juzxtaposed that are of varied densities.  For example, if you had drywall by itself, it might be quite noisy.  If you have wood bound to it and supporting it, it would be much less noisy, place extruded foamboard insulation on the other side of the wood, and the chances are you would hear next to nothing.

If you had a composite roadbed of vinyl underlay, say 1/8" thick, and over top, tightly mated along their lengths with almost no gaps or airspaces, a similar dimension of masonite strips for a top surface, your tracks would be very quiet.  I had a trestle made of scale dimensional lumber with Atlas Code 100 caulked to the stringers.  Amazingly quiet.  The think caulk between two distinct densities, the plastic ties and the stripwood, rendered the sound almost mute.

What makes for noisy trains is the glued ballast.  The ballast transmits vibrations and noise through its mass with its essentially homogenous density, and directly to the adjacent sub-roadbed or plywood surface.

One of the best sub-roadbeds for noise, believe it or not, is drywall.  I mocked up a temporary track system between layouts using strips of drywall as the surface on which the rails sat.  It was deadly quiet.

Crandell

  • Member since
    September 2003
  • From: Southeast Texas
  • 5,449 posts
Posted by mobilman44 on Wednesday, January 23, 2013 3:55 PM

Hi, I have a similar setup, and have no problem with the sound transmitting thru the walls. 

The cork is definitely a help, and for my latest layout I used caulk to affix the roadbed instead of nails - which also should be a help.

One other area that has to have made a difference is my under table supports.  The closer you are, the smaller the "drum area" of the plywood.   Mine are 18-24 inches, and if nothing else give me a very solid RR.

ENJOY  !

 

Mobilman44

 

Living in southeast Texas, formerly modeling the "postwar" Santa Fe and Illinois Central 

  • Member since
    February 2008
  • 8,851 posts
Posted by maxman on Wednesday, January 23, 2013 3:14 PM

You need to get a copy of the January 2013 MR.  In there is an article by Pelle Soeborg called 5 Expert Tip for Installing Track and Roadbed.  He discusses sound deadening sub-roadbed.  Much easier to read the article than for me to explain here.

Quite frankly, I don't think that any wheel noise is going to be transmitted through the walls.  If it does, than I guess you're not going to have any sound locos, either.  Probably the vibration you mention will be the bigger culprit.  But I don't think that any sort of roadbed will help you there.  You might want to investigate installing some sort of rubber isolators between the shelf bracket supports and the walls.

Another option would be to have your lower level supported off the floor, and the upper level brackets attached to some risers that are supported from the bottom level.

If you are contemplating some sound locos I'd set up a temporary shelf, not supported by the wall, and turn up the sound.  Blow the whistle and ring the bell.  If that annoys your neighbors, you may have to re-think your project.

  • Member since
    January 2013
  • From: PA
  • 481 posts
Roadbed sound reduction?
Posted by Schuylkill and Susquehanna on Wednesday, January 23, 2013 12:00 PM

Hi, I'm working on planning benchwork for a double deck layout - only problem is that it is going to be attached to the walls and I need to keep the layout rail noise at a minimum.  I plan to use 1/2" plywood subroadbed with 1/4" and 1/8" cork on top.  I have some homosote (do they really not make it anymore?), but not enough to use under the track - besides, it would be too thick.

I was thinking about using homosote pads between the risers and the subroadbed to dampen the noise.  The real problem is the vibrations getting into the walls since the house has almost no soundproofing.  Anybody got any ideas?  I'm going to use shelf brackets to attach the layout to the walls, and I model in HO scale.

Thanks

 

Modeling the Pennsy and loving it!

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!

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!