Testing the waters
Rob Winstone considers a question often asked about moisture testing anhydrite, calcium sulphate and gyvlon type screeds
WHEN these were developed and used on construction projects they were intended as a means of installing a levelling screed into large open-plan areas where concrete plank, beam floor or insitu cast concrete had been poured.
These were usually around 25-35mm thick, gave a surface that could be trafficked within a day or so and when installed correctly, a level, flat and smooth surface was achieved. This thin layer meant the drying times were reasonably quick and the whole process was better for the main contractor’s programme.
As the product has developed and screed contractors have become more efficient at installing the material, we have seen the product being installed into smaller areas and now into domestic properties.
This has been further driven by the manufacturers of wet underfloor heating systems who have found that a more even and efficient heat distribution can be achieved using this type of screed with their system.
Screed contractors have found it is easier to install a flowable type of screed than the hard work required to install a traditional sand and cement screed correctly.
However, this use with underfloor heating and other situations has resulted in thicker screeds being installed, in some cases at the same thickness as a sand and cement screed had it been specified. I’ve found cases where they are over 70mm thick.
Research has shown that as these screeds increase in thickness, the time to dry increases disproportionately more. It’s now not uncommon for an anhydrite screed at say 70mm thick to take as long if not longer to dry than the conventional correctly mixed sand and cement screed!
Anhydrite screeds dry from the bottom upwards forcing mobile moisture upwards. From a flooring contractor’s point-of-view it’s the top surface he’s bonding to which is of great importance.
Therefore, they need to be certain all the rising mobile moisture has gone in order to provide a dry surface with which to bond. Using Hygrometer boxes (I wouldn’t recommend the use of humidity sleeves in this situation) provides the best method to test for moisture, but the results seen will be very different from those seen when testing a sand and cement screed using this method.
With a sand and cement screed a gradual fall of the meter reading will be seen as the screed dries. With an anhydrite type screed the readings often remain high and then will suddenly drop within a day or two as the last of the moving moisture leaves the screed.
Where this makes life difficult is that testing a conventional sand and cement screed you may be able gauge how the screed is drying and even when it might be dry, but with an anhydrite screed there is no indication when this will occur.
With some experience, the use of an electronic radio frequency type meter can sometimes provide some information in the latter stages of drying but not sufficient to risk the laying of a floor covering. They’ll give a good indication of the wetter areas of the floor and these are the places to set hygrometer boxes.
The problem I see is that once the anhydrite screed has been installed the works programme is committed to waiting for the screed to dry or that someone takes a risk and goes ahead and installs the floorcovering.
With a conventional sand and cement screed there are options to use a system to protect the floor covering from moisture if the works programme is becoming that critical. There are now companies producing products to protect the floor covering from moisture within an anhydrite screed. These are still very new and I personally have no experience of them and have not inspected any failures where these have been used.
I’d be keen to hear from anybody that is producing or has experience of using these products. Feel free to email me your contact details (firstname.lastname@example.org).
Rob Winstone is a mediator and flooring consultant specialising in timber, moisture-related issues and slip testing
COPYRIGHT: Christine Rondeau Flickr