Question from Melanie Reynolds, via email:
We have Wavin underfloor heating (UFH) installed on board (no screed) with plywood on top and then tiles. A problem arose recently when one tile was loose. When lifted, the underside of the tile was very wet. We thought there was a leak but pressure tested the system and it stayed at 1.5. Then we lifted six other tiles and the plywood was rotted from moisture. When we put a temporary tile down the next morning, the underside of the tile was wet, but the plywood we put down was dry.
The system has been installed for a year so the builder is saying that the problem is condensation which has started on the tiles and over the last 12 months has dripped down to rot the plywood (i.e. not the other way around). The sand is also dry. Have you got any ideas as to what is wrong?
Answer from Anthony White, technical and product development manager at Thermoboard:
Without further detail regarding floor type and floor construction it is difficult to identify the direct cause of the problem. Therefore, if this response does not provide the answers you require please contact us directly.
As your pressure test clarifies that the UFH system is not leaking there are three other possible explanations for the moisture.
1. Water is seeping down through the tiles and grout. If the floor is regularly wet, such as in a walk in shower room, and the grout isn’t waterproof then this is a real possibility, but for a normal tiled floor, which may occasionally get wet, it is not very likely.
2. Water is coming up through the floor. The email mentions that the sand is dry. Assuming this is a binding/levelling layer laid over a concrete sub floor, it is possible that the new concrete sub floor was not allowed to dry properly before the floor finish was laid. A minimum of one day per mm of screed/concrete drying time should be allowed, although this should be increased if the floor is laid over the winter period when ambient temperatures are low.
3. Water is coming up through the floor from outside. This could be caused by omission of the vapour barrier/DPM from the construction, or there may be another source of water ingress from the building perimeter e.g. a grassed slope on an exterior wall with no external drainage channel.
As tiles are typically around ten times more thermally conductive than the plywood used in your test and less absorbent, condensation or moisture is more likely to collect beneath them if it is trapped within the sub-floor.
The outcome of the test you have carried out would therefore suggest that the moisture is trapped within the sub-floor, and is either being caused by explanations two or three above.
Assuming you do have a concrete sub floor beneath the sand, please test its moisture content using a hygrometer before continuing, as you will need to ensure it is dry before laying a new floor regardless of the original cause of moisture.