Mat footing idea/question

[UnneutralAxis]

This might not be original and it might not be the best idea ever. Just wanted some opinions.

I have a site that got a pretty bad soils report and a rigid mat was recommended. Pretty light loads, one story office building. 100 psf floor load. Line loads less than 1200 plf. Point loads less than 20 kips. Seismic design category B. Will be using a mat 12" thick, roughly No 5 top and bottom around 16-18" o.c.

Getting to the point now... Frost depth is 30". Owner doesn't want to step down into the finished floor. I worry a little bit about doing a turn down as I don't want to have the possibility of an increased bearing pressure occurring at the turn down. Has anyone used a trench of 57s to get below the frost depth at the perimeter? Seems to be a descent idea, the only real concern I have is with scour. A rigid insulation detail would be provided just inside the trench interior of the building.

Thanks.

 

[ishvaaag]

You may acommodate the heave from the swelling as an additional hypothesis to check and see how it affects deformation and modify if required the design. At the expected pressures on the ground even some insulation may stand well the working pressures long term and buffer differential movements.

Hence in my view adding some hypotheses may deal with your problem with scarce structural change.

 

[ishvaaag]

By the way I have seen to behave well a 14" thick mat with 3 floors above and no particular precaution for frost. Not advocating no precaution, but mats can be generous foundations.

 

[UnneutralAxis]

Thanks for your comments. I have to have some frost depth shown on the construction documents or the building officials won't approve the plans.

A rubble foundation is nothing new, just haven't seen one on a "new" set of plans before. I've never done a rigid mat for an entire building before, so I was curious how everyone else handles the frost depth at the edge. I've only done them supporting a group of columns/line loads interior of the building.

 

[DaveAtkins]

You may want to lower the mat footing below the frost depth and pour a slab-on-grade.

DaveAtkins

 

[hokie66]

If the frost protection is just needed at the perimeter, just dig a trench, fill it with lean concrete, and start your mat above. Slightly slope the top of the frost wall and isolate it from the mat with building paper or plastic sheeting to allow the mat to shrink without substantial restraint from the perimeter wall.

 

[UnneutralAxis]

hokie, yes frost protection is needed at the perimeter. I considered shotcrete but felt the subsurface drainage benefits of the 57s would serve well at this particular site. Has a very shallow water table.

DaveAtkins, thanks. Yes, I suggested this, but the owner doesn't want to step down into the finished floor, which would be approx. 30" (frost depth) - 12" slab = 18" or two steps down into the building. Plus, then you have short CMU stem wall with waterproofing to hold back soil. Makes his building look shorter from the road and all the other complaints one could think of.

Specifically, does anyone see a big problem with the 57 trench? Like I said, I only see a problem with possible scour.

 

[larsacious]

I like hokie66's approach. Lean concrete is relatively cheap and all that would be required is a trench pour. The cold will likely transfer under the building if all you provide is stone.

Also, I don't see an issue with having a "turndown" at the perimeter like you originally suggested. This will likely improve the mat's performance and make it behave more like a rigid body. Let's say the bearing stress is exceeded below the ribbing. As soon as the soil would begin to fail the stresses would redistribute by either having the ribbing span over a localized failure location or by additonal stresses being transferred to the 12" slab immediately adjacent to the ribbing.

 

[ToadJones]

If you use a turn-down, you'd have to insulate the inside of the vertical portion as well as some distance in from the perimeter under the mat as well, correct?

 

[ToadJones]

if using a turndown in cold weather, what keeps the cold from penetrating at grade level thru the mat or slab?

 

[BigH]

Why not consider placing a vertical layer or combo of vertical and horizontal of insulation (styrofoam) along the outside perimeter so that the "minimum" path for frost would be more than 3 ft? I guess this has an acronym now - FPSF

http://web.byv.kth.se/bphys/reykjavik/pdf/art_134.pdf

http://www.supergrund.se/files/document/tjale.pdf

http://www.cs.arizona.edu/people/jcropper/desguide.pdf

http://www.huduser.org/publications/pdf/frost.pdf

http://abecsouth.org/Design_Tab_02_03_30.pdf

http://www.foundationworksinc.com/techdata/frost_solution.pdf

http://www.styrotech.com/products_general.html

 

[larsacious]

@ToadJones: I look at the turndown as being no different then a frost wall at the perimeter of the building. I have seen these walls constructed with and without insulation placed on the inside face of the wall. In either case I have not heard of frost heave issues associated with them. As I understand it, the frost wall holds the warmth of the soil below the building.

You are correct that cold can still transfer through the 12" mat if the mat is exposed. This will not create a heave issue since the heat transfer will need to freeze the mass of soil below the mat. However, if exposed, this could cause cold spots in the floor, sweating of the concrete, flooring issues, etc. This is why it is better to have a floating slab on grade with expansion felt between the frost wall and the SOG.