Ceramic Tile Advice Forums - John Bridge Ceramic Tile

Ceramic Tile Advice Forums - John Bridge Ceramic Tile (https://www.johnbridge.com/vbulletin/index.php)
-   Tile Forum/Advice Board (https://www.johnbridge.com/vbulletin/forumdisplay.php?f=1)
-   -   Addressing I-joist vibration before marble install (https://www.johnbridge.com/vbulletin/showthread.php?t=130121)

Demonic 11-24-2020 01:32 PM

Addressing I-joist vibration before marble install
2 Attachment(s)
Hey all,
I've researched as much as I can about this but would still like to create a thread. I bought a house last year and am renovating a room that will serve as a formal dining room. The house was a custom build in 2007, and the room was spec'd for large format porcelain tile. It currently has 20x20 porcelain tiles, which are in great condition, no cracks even in the grout. Unfortunately I can't stand the look. I'm planning on replacing them with either 18x18 or 12x24 marble. The current tiles are on thinset on 1/2" ply, on top of 3/4" ply, on top of Boise AJS-20 I joists, spanning 16'4". The bottom 3/4" ply is glued to the I joists, and the top 1/2" ply is screwed into the lower ply.

Going by the Boise span charts for the AJS-20 series, which are 11 7/8" by 2 1/2", my deflection is over 720, and close to 960 for the 16'4" foot span. My concern is there's definite vibration when you stomp on the floor, and I've read this is a common complaint of I joist spans when they reach 16 feet. I'd really like this floor to feel solid when people walk across it. The basement area under the floor is completely unfinished and open, with access to the joists. I don't plan on finishing the basement, but would like the try to keep the area relatively open as I'd like to later use it as a basic shop space. The long dimension of the dining room is 20ft, but the open length from foundation wall to next foundation wall under is 26ft. I'm considering running a double 9.25 x 1.75 LVL beam across the 26ft length (green line), cutting the span of the 16.3ft joists in half. I'd support the beam with lally columns on either end resting on the foundation footers, and a 3rd lally column in the middle with a new footer placed in the basement slab. I spoke to a custom home builder I'm close with, and he said if I already meet the deflection reqs, then adding the beam is probably a waste of money and effort and he would instead do blocking and strapping.

I'd like to minimize transition height to the adjacent wood floor. So I was planning on keeping the existing plywood layers after removing the thinset, then covering with Nobleseal CIS using EXT, then the marble tile.
What are you opinions? Would you proceed with the effort of installing a midspan beam and the new footer in the middle to support the lally column, or would you just block and strap? Or something else?


Ps. I'm removing that insulation, not adding it in.

mykcuz 11-24-2020 01:59 PM

Maybe a small steel I beam could save you the middle basement column, leaving the basement fully open.

You can infill the tgi web with solid lumber that fits in the web height wise. SSI essentially each"beam" would be a sandwich with two lengths of solid lumber on the outsides and the tgi in the middle. Then you could add cross blocking in between.

You can also use steel strapping used to strengthen plates that get cut out for ductwork. About 1"width by 20"long, but with 16" screws holes for attaching to 16"centers. You could run a row of these perpendicular to the tgis, crossing the whole span screwed up to the underside to lock tgis together.

Sent from my SM-G950U using Tapatalk

cx 11-24-2020 08:14 PM

Welcome back, Austin. :)

While cross bracing is sometimes specified for those engineered joists, they're frequently designed with a full ceiling below that might help the "feel" of the floor. My first recommendation would be to contact the manufacturer with your concern and see what advice they have. You can be sure yours would not be the first such call.

And while your intended beam might certainly dampen the vibration you don't like, I think it might be overkill. Were I to experiment without contacting the manufacturer, the first thing I'd try would be to install some long 1x4s screwed into the bottoms of and perpendicular to the joists. I'd run two rows at the 1/3rd points of the joist span and see if it made any difference.

The manufacturers of those joist systems sometimes specify some cross bracing, which would be better than the struts I suggested, but not as easy and quick to try. Again, talking to the manufacturer would be a better first step.

I would not recommend the sistering recommended by Mike above. What you'd be doing, if I understand his recommendation, is double sistering engineered josits with dimension lumber, which is not gonna do anything significant for the existing joists. While it might have some dampening effect, it's a hellofa difficult and expensive method to experiment with, 'specially since you don't need the additional structure for your deflection requirements.

The metal strapping across the bottom of the joists is the same principle as the 1x4 "stuts" I recommended.

My opinion; worth price charged.

mykcuz 11-24-2020 09:24 PM

Demonic cx is incorrect in stating sistering your tgis is ineffective, it doubles their strength. Yes there is a cost associated with purchasing that much lumber, but it is one option to easily diy double the strength of the 16' floor. You can also RIP osb and attach to the sides as a stiffener, which is usually the choice of manufacturers when cantilevering.

26' width is about 20 tgis
If you installed 1 per joist it's about$500 in lumber. 2 per joist$1000.
Ripping lengths of osb would be half that, or less.

Different options, different cost, different results.

Metal strapping is my preference over wood strapping, because you can add sheetrock right over it in the future. Which you may find installing plywood or gypsum to the ceiling to be attractive one day, to separate your workshop noise/ dust from your dining room. That load should be factored into the equation before choosing a solution.

Blocking/ strapping allows the joists to somewhat share the load, however is more effective in dimensional lumber framing, than with tgis.

You probably have 1/16" deflection over 16' , osb rips could probably bring that down to 1/10,000" or basically nothing.

Any architect/ engineer with the manufacturer's tables could tell you how to best get to your desired deflection. Or of course the manufacturer can guide you directly.

Sent from my SM-G950U using Tapatalk

jadnashua 11-24-2020 11:00 PM

The core or rib of a I-joist is mostly to help hold the top and bottom pieces in their proper orientation. Blocking can help limit twist when a load is applied, that helps maintain the vertical deflection and share the load between adjacent runs. You can get similar results by adding strapping, along the bottoms, tying them together. Gluing and properly attaching the subflooring on top of the joists keeps each of them aligned and working in conjunction with each other.
Adding drywall to the bottoms of the I-joists can perform a similar action, but may be better done with the strapping first, as the screws will just eventually elongate the holes through the drywall slightly.

The I-joists work by the top cord being compressed, and the bottom one being stretched. The middle just holds them in position, so sistering doesn't do all that much. That's one reason why you can cut holes in the webbing more easily in an I-joist...it's not providing all that much strength to the assembly.

mykcuz 11-24-2020 11:52 PM

Sistering properly, doubles the strength of the tgi. Been there, done that.

The web absolutely contributes to the overall stiffness/ strength of the joist, otherwise hole size wouldn't be limited. It helps resist movement, and distributes forces in addition to aligning the cords.

Sistering can be performed several ways. Adding plywood or other sheathing material is called for by manufacturers specifically in cantilevered applications. Doubling of I joists, joined by solid filler in the web with glue and screws is another option. Doubling solid lumber to sandwich the web and glue and screw/ through bolting, is yet another option.

Each assembly has its own rating which also considers span. Either the manufacturer, or an engineer, would have easy access to these rated assemblies charts.

Sent from my SM-G950U using Tapatalk

Demonic 11-25-2020 07:59 AM

Thank you everyone, for the taking the time to weigh in. I should have also importantly asked - what are the chances the perceived vibration would increase the chance of the marble cracking? That was actually a bigger concern to me than something like wine glasses being able to rattle.

Regarding the beam and adding the center column - I don't mind having the column in that space, but having to cut the slab, then have a building inspector come to inspect the footing after digging, and then pouring the footing is kind of a pain that would be nice to avoid. I had thought about the possibility of a steel I-beam without the center column, but I think the cost and weight for 26ft would be prohibitive.

Boise does give details for nailing OSB or ply against the webbing to act as web stiffeners. They only mention nailing with 8d nails though - I would think I'd also want construction adhesive against the webbing, no?
And likewise wouldn't the strapping be more effective if screwed 'and' glued against the bottom flanges? Perhaps the web stiffening combined with 1x4 strapping would get me there.

I'll definitely contact Boise after Thanksgiving, and will report back with what they say after I pick their brains.

mykcuz 11-25-2020 08:50 AM

Definitely talk to bc rep or architect. Steel is a commodity, I'm in nj and last year I purchased a 24' steel w beam 12"height with drilling for bolt holes for about$3,000.

Since it isn't needed to be loaded bearing you could use something smaller, but ask an architect to spec it. You could possibly get away with a simple flitch plate and double lumber, through bolted, as small as 4 or 6"

Glue and structural screws are superior to glue and nail, which is superior to nail alone.

Sent from my SM-G950U using Tapatalk

wwhitney 11-25-2020 08:57 AM


Doubling of I joists, joined by solid filler in the web with glue and screws
This is the one proposed method that would double the stiffness of an I joist (plus a bit more for the solid filler). The other proposed methods are going to increase stiffness less than 100%.

Halving the span with a beam would increase the deflection resistance by at least a factor of 8.

If the bottom of the joists are clear and you didn't mind closing off the joist bays, adding a layer of plywood as a ceiling would certainly improve to the composite action of the joists to resist point loads. It could also create something like a box beam. I'm unclear on whether a single strip of plywood 4' wide or 8' down the middle would give you most of the benefit of a full ceiling.

Cheers, Wayne

All times are GMT -6. The time now is 04:45 AM.

Powered by vBulletin® Version 3.7.4
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.
Copyright 2018 John Bridge & Associates, LLC