capillary break and floor slabs

[fattdad]

Dear Discussion Forum (first posting, sorry it's so long),

Greatings from Richmond, Virginia USA. My company's practice is to use a dense-graded aggregate (i.e., minus 3/4-in with about 5 percent passing the 200 sieve - D10 of about 0.15 mm) for the "capillary break" beneath industrial floor slabs. Many in the industry (years ago) used to use an open-graded aggregate for the same purpose (i.e., minus 1-1/2 in with less than 3 percent passing the No. 8 sieve). From my understanding (I'm a geotechnical engineer not a structural engineer) they both serve the same structural purpose (i.e., enhancing the subgrade modulus and also providing a more uniform subgrade condition) and are also intended to serve as a "capillary break" between the subgrade moisture and the underside of the concrete floor slab.

Here's why I write: Looking at Terzaghi's book, the standard equation for capillary rise is Hc=c/(e*D10) using consistent units, where C is in the range of 0.1 to 0.5 cm. When I use this equation for the dense-graded aggregate, the calculated range for capillary rise is somewhere in the range of 23 to 115 cm (9 to 45 inches). As you can imagine this does not seem to offer a very effective "capillary break" when a typical layer thickness is 6 inches.

Being a born skeptic (and hopefully a passible engineer), I decided to do my own "bench test". Here's what I did. I make a column of the aggregate materials in a cylinder and acheived a compacted density of about 120 pcf. Not having a "Proctor" yet (I'm still forumulating my strategy), I figured that this was about 95 percent relative compaction). I submerged the lower inch of the 15-in tall cylinder and waited for capillary movement. Went home for the weekend and when I returned, the water had fully reached the top of the cylinder.

Looking at the post-test moisture contents, I noted that the moisture content in the submerged interval was about 12.2 percent and with increasing height I recorded moisture contents of 11.4 to 9.95 percent. Using the 12.2 as "fully saturated", I concluded that the capillary rise provided from 93.3 to 81.4 percent saturation (again with height).

Here's why I write. While it is undesirable to have fully saturated conditions on the underside of a slab, is there any guidelines to acknowledge whether it's o.k. to have 70 percent saturatation? What about 82 percent saturation?

I typically expect the optimum moisture content for compaction is in the range of 85 to 90 percent, so the typical post-compaction moisture contents would be in this range anyhow.

Does anybody on this forum understand my ramblings? Any comments? Feel free to post or otherwise provide insight.

Sincerely,

Carl "fatt-dad" Benson
Geologist and Engineer

http://www.fatt-dad.com

 

[cvg]

preventing capillary rise is required mainly where moisture in the concrete is undesirable. For example, where the floor is to be tiled, carpeted or painted, moisture should be prevented to enhance the performance of the flooring. Otherwise, the moisture may not be so detrimental and may even enhance the long term curing of the concrete. If it is desired, why not use a polyethylene vapor barrier instead? Specifiy it sufficiently thick to prevent punctures during the concrete pour. Note that the open graded aggregate may not compact to form a good uniform surface and may be disturbed during the pour.

 

[fattdad]

Yes, the use of a slip sheet/vapor barrier is also instrumental to our design. From what I understand there is an upside and a downside to the use of a plastic sheet, with respect to uneven curing, unless the moisture-cement ratio is carefully controlled. This can lead to slab curling.

Regarding the floor finish, I would think even for an unfinished industrial floor slab, the prudent design would be to use a vapor barrier (i.e., plastic sheet), as you never know the uses for the floor in the future.

I'm really cogitating on what are the "needed" engineering qualities for the subbase (i.e., dense- or open-graded aggregate). For the purpose of this thread, when I use the term "open-graded" aggregate, I'm referring to crushed stone that's free of fines. Not uniform sand, where construction disturbance can lead to un-uniform slab thickness.

Thanks for your comments!

f-d

 

[Ron]

fattdad...congrats on the skepticism and the testing.

Dense graded aggregates as you have shown, do not make great capillary barriers.

Additionally, one factor that Terzhagi and others did not consider is the absorption of the aggregate in its contribution to capillary rise. If you do your test with a relatively porous calcareous aggregate and then again with a siliceous aggregate, you'll get different results. Nevertheless, your question of a tolerable percentage of saturation is a good one, though difficult to answer.

For concrete, the defining factors for floor covering placement have been vapor transmission and the relative humidity of the concrete to about 1/2 its thickness. The relative humidity process is becoming more accepted as the appropriate quantifying procedure as it is more accurate and repeatable than the vapor migration/transmission tests.

The typical limits for subsurface relative humidity are 75 to 85 percent in the concrete. How does this relate to subgrade saturation? Difficult to answer because of the variations in concrete porosity and diffusion, given a stated source of moisture below the concrete.

 

[fattdad]

Ron and others. So then the question remains, why expend the extra cost for a granular subbase (other than to enhance the subgrade modulus). If the plastic sheet preserves the relative humidity to acceptable levels in the concrete and notwithstanding the curling problems, there is no real benefit as a vapor barrier. I have a followup bench test to run in the coming weeks and will post results.

 

[Ron]

fattdad...plastic sheeting does provide a vapor "barrier" in that its perm rating is lower than that of an air gap between the concrete and partially saturated soil.

As for enhancing the subgrade modulus...that's usually not necessary for a typical slab on grade.

I'm not a fan of plastic sheeting, but it has nothing to do with vapor transmission or water. It has to do with the fact that it is almost never flat, the folds roll up into the concrete, and cracks result from the "control" joint created by such conditions.

 

[fattdad]

Bad writing on my part: I was referring to what extent does the subbase aggregate (whether dense- or open-graded) serve as a vapor barrier, if and when a plastic sheet is used? No doubt the plastic sheeting will.

Regarding the subgrade modulus, if somebody determined the soil subgrade modulus to be 200 pci and used a 6-in granular subbase, the design modulus value whould then be 225 (or so). I figure that Ron's perspective is correct, this would have a negligable effect on the structural design of an industrial floor slab, but it is a (somewhat trival) factor. . . .

f-d

 

[Ron]

FD--the subgrade, in any form, does not serve as a vapor barrier. Some type of water inhibiting course is required if vapor permeance is to be significantly inhibited.

 

[DRC1]

Water will climb as long as it has a path and 5% passing the # 200 seems to give the water an ample path. The variations in water content may be due to a) variations in the grading, b) not complete saturation. Ifit is b, that may increase to 100% given enough time. As far as controlling the water, I am a great fan of clean (washed) 3/4 inch stone. It is very permiable so water does not climb, is easily spread and requires no compaction beyond light tamping. just be sure that the stone can drain
By the way great job on the testing and very interesting results.

 

[fattdad]

DRC1 - thanks for the comments. From my review of capillary rise (Terzaghi and Peck, "Soil Mechanics in Engineering Practice" p. 133), the height of capillary rise in soils is based on an effective diameter, which is correlated to the D10 grain size. When you calculate the height of capillary rise you are equating D10 essentially to the diameter of a capillary tube. The reality is that there are other "tube" diameters that are larger (and some that are smaller). While you may truely realize the calcualated rise in a bench test, you will never realize complete saturation over the capillary height.

I had an interesting discussion with my boss on this matter. He claims if you don't have full saturation than maybe it's not that much to worry about. Granted at some point he's right, but also bear in mind in many fine-grained soils you don't have complete saturation even when you are below the ground water table. What is at issue is whether there is the tendency for free water to "move" through the soils and whether these flow paths are interconnected.

Being a fan of washed 3/4 in stone may work great in your part of the world. In Virginia (where I practice) as well as other parts of the U.S., there is a growing desire to use something else for reasons referenced above (i.e., if there is a thunderstorm or rain event you end up with full saturation of the stone layer where a dense-graded aggregate will allow for greater surface water runoff).

Ron, I continue to type too fast. You are again correct that the subbase does not serve as a vapor barrier. It is considered a moisture barrier. I guess I continue to wonder if we have (my firm and other's as well throughout the country) successful projects sitting on dense-graded aggreagate, the subbase is really not doing much to provide a moisture barrier if capillary rise is not impeded. It implies to me capillary rise is not the real issue (i.e., saturation may be the real issue).

f-d

 

[oldestguy]

fattdad: In interesting experiment you ran.

If you really want to get into the mechansims of capillary rise, etc., go to a university library and read up on "Moisture Tension" or as the Britts say "Suction". The soil scientists have gone very deeply into this subject. Check out what the unit Pf stands for. It very likely can be applied to concrete as well as soil.

Floor systems coming loose from concrete can be related to variation in soluble salts in the concrete, in additon to humidity.

A real complicated subject you brought up.

 

[fattdad]

oldestguy, Even though I'm trained as a geologist and an engineer, I've had to dabble in characteristic moisture curves in the past. I'm familiar with soil tension and it's hysteritic parallels with unsaturated hydraulic conductivity (this is real complicated stuff as well). At some fundamental level, all of these equiations must fall apart by a tension. That said, I'm now formulating my next experiment. Here's the early thought:

I'll get two 4-in diameter pipes that are 5-ft tall. I'll make joints every foot or so to enable filling the pipes with dense-graded aggregate. In one, I'll just use the dense graded aggregate and the other I'll use an initial one-foot layer of open-graded aggregate. I'll start both tests with air-dry materials. At 6 to 8-in intervals, I'll drill a hole and insert a wick that'll "dangle" out the side of the pipe. This will be a capillary "tell-tail". I'll submerge both pipes and wait. . . . . .

After I'm done waiting I'll quantify the amount of water that issues from the tell tails and I'll disassemble the pipes, measuring the moisture content in each tube. I would expect the tube with the open-graded aggregate to have increased moisture content by virtue of vapor transmission. I would expect the tube with out the open-graded aggregate to have greater moisture content. I'd also expect to have some dripping from the tell-tails. Hopefully, there will be some height where no dripping will occur. This marking the highest point of free water movement.

Further comment welcome - not sure the practical application at this point, but that's o.k.

f-d

 

[oldestguy]

fattdad

Good deal about tension stuff.

OK now let's see if you can complicate it with a third tube identical to either one. Put some rock salt in the very top and close it off from the air above.

Under the rock salt put in a layer of concrete or mortar, sealed to the sides. Give it a while to act. My former experiments show that concrete is an ideal osmotic membrane. Osmosis explains a lot of floor failures that cannot be explained by vapor pressure theory alone.

Ain't it fun?