Shortly after a contractor installed a wood floor in our newly constructed home we began to notice the floor was slightly uneven in spots and the hardwood was cupping.

The floor was installed within a 1.5 inch recessed slab to allow the wood floor to be on the same level as the tile and carpet floors in the baths and bedrooms. The contractor laid a vapor barrier, then nailed 3/4" plywood to the slab using 29 to 32 steel nails per sheet of plywood, then laid a layer of felt and finally nailed the 3/4" oak to the plywood and sanded and finished the oak.

When the contractor was called back to look at the buckling and cupping, he suggested that there might be a plumbing leak. The plumber was called and did a pressure test and concluded the plumbing system was not leaking.

Having no other alternative, the contractor pulled up the floor to see if we could determine where the moisture was coming from. There was nothing obvious under the floor to indicate water or flooding. In the process of removing the floor, the plywood panels appeared to be dry and intact (no separation of the ply layers) but many of the nails that held the plywood to the slab were completely rusted off at the slab surface.

The contractor advised us to let the slab sit for a couple of weeks and watch for any obvious accumulation of moisture. He said that we should run the showers and bathtubs and flush the toilets frequently to see if that produced any wet spots on the slab. So far nothing obvious is happening to dampen the slab and it looks and feels dry.

What would cause the nails to rust off at the slab surface? Would the failure of the nails cause the buckling? (In some spots the floor was bowed up 1 or 2 inches and it was easy to trip).
The general contractor who built the house is to send an engineer to look at the plywood panels and the rusted nails to try and determine the cause of the floor failure. None of the plywood appeared to be damp.

I am hesitant to pay for the installation of another floor until I know what caused this one to buckle. Any advice on why this happened and how to prevent it happening again?

Appreciate any suggestions.

Well, only one thing will cause steel and iron to rust.
Your challenge (or your contractor's) is to determine the source of this moisture.

I would find out whether you have copper pipes embedded in the concrete. Even though the plumber did a pressure test, the leak might either be very slow and minimal, or be part of a circuit that is not part of the system he checked. Dunno cause I am not a plumber. But I have seen similar moisture problems, and judging from the rotten wood (engineered) in my case it must have gone on for years. A slowly leaking copper pipe could saturate the concrete over time. They did that in the 80s that I know of. If your neighborhood was developed at the same time, talk to neighbors and see what they think. You can also look at the little wheel on the water meter, and mark it with white out or something, while turning off everything in the house that might consume water, especially the reefer if it has an icemaker. You would have to unplug it from the power outlet. You could then see or at least get an idea whether you are leaking water. Not sure if this helps cuz I don't have a scientific report to give, but might be something to check.

The house is 3 yrs old. All the pipes in the slab are pvc - no copper. The water is from a well and during the pressure test the plumber turned off the well pump for several hours and we unplugged the reefer and closed the valve on the icemaker.

I suppose it's possible that the relative humidity of the slab was too high when the floor was installed and caused condensation that rusted the nails and popped up the subfloor??? I'm no engineer either. This has me stumped.

Thanks for your response.

Well first off, even though your only 1.5 inches below the other slab technically that is a below grade installation. Second puncturing the plastic vapor retarder with thousands of nails does not help at all. NOFMA (which was swallowed by the NWFA) always stated that and installation over concrete like that needed at least tow layers of alternating cold roofing mastic and 6 mil plastic if memory serves correct. The mastic seals up the holes. Otherwise it's a crapshoot. But a simple floated underlayment works great in that type of installation.

You need an experienced person to test the concrete MVER with suitable industry apporved device like a calcium chloride test or RH probe meter. I use the Wagner RH equipment for a quick idea of whats under there. That should have been done prior to installation anyway.

Also you cannot simply watch for moisture emissions out of a slab. lol Just like you cannot tell the moisture content of wood without a pin meter. Read the services page on my website for more info on moisture testing concrete. http://www.tucsonazflooring.com/services.html

in the meantime you can do the redneck moisture test. cut a few 2 x 2 squares of 6mil plastic and tape them down on all four sides to the slab. Wait 24hrs and pull up and look for condensation on the back of the plastic.

Many thanks to Floormeintucson and to Kevin D. I will pass this on to the general contractor so he can get the slab tested for moisture and put together a plan for replacing the floor properly. I really appreciate the information you provided.