I’ve recently been looking back at some old notes on project work I carried out whilst working in the laboratory at Laybond (was it really 17 years ago?), prior to the company being bought by Bostik. This brought back memories of working in conjunction with two major manufacturers, Lafarge Gyvlon and British Gypsum, on a ‘new’ type of screed that was not cement based but actually used a form of gypsum as the binder. We now know this to be a calcium sulphate screed.
The relationship with the two manufacturers worked by bringing together knowledge of flooring products such as smoothing compounds, primers, adhesives etc. with an in-depth understanding of screeds. It was a fascinating learning curve for us all and yielded a great deal of information, which highlighted concerns but also recommendations relating to laying floor products.
There are great benefits to be had by utilising calcium sulphate screeds, mainly due to their ability to be pumped, therefore reducing the time that screeders need to be on site. Other benefits include the use of recycled components, the ability to apply thinner than equivalent cement screeds, minimal bay joints due to reduced tension and the ability to flow around underfloor heating (UFH) systems. These UFH systems can subsequently be brought online in a shorter time frame than cement screeds and allow acceleration of the drying.
So what were the original concerns and the subsequent remedies relating to the use of calcium sulphate screeds?
The first concern was that the calcium sulphate system was not directly compatible with cement, which accounts for the vast majority of smoothing compounds used on the market. After significant test work we concluded that the use of a water based epoxy barrier primer provided by far the best priming option. This created a chemical barrier to protect the cement from various salts in the calcium sulphate and so enabled the two materials to work happily in combination. PVAs were a disaster, whereas acrylic primers needed to be of a microporous nature and required several coats. These also struggled if there was a hint of dampness remaining in the screed. To this day I can say we have never had an issue when the epoxy primer has been used correctly.
The next concern was regarding the strength of the screed and what products could be used over them. We soon realised that applying too thick a layer of a high strength cement system could indeed cause a problem. It appeared that the curing tension in the cement systems were putting stresses on the upper surface of the screed and pulling away from it, usually taking some screed with it. Our remedy here was to recommend a maximum thickness of cement levelling compounds, which varied depending on the inherent product strengths. Typically up to 10mm was fine – the use of chippings to reduce curing tension was incorporated for anything above this.
The third issue was not readily observed in the initial work. When carrying out test work we always ensured reasonable site conditions and also tested the screeds for dryness before continuing works. It soon became apparent that the instructions for use on site were not being followed and that the screeds were sometimes far from being dry at the predicted time. Simple things like covering the screed over with other materials during its drying time, not having watertight buildings so excess moisture was being added, leaks from plumbing, varying temperatures and airflow on site not being anywhere close to what is required to take moisture away from the subfloor were all too often preventing the product from drying out. Furthermore, the UFH systems often used with these screeds were not being connected so could not be used to accelerate the drying.
A design error I have also seen more often than I would like is for the calcium sulphate screed to be laid far deeper than specified, therefore greatly delaying the drying times. Drying times are also an issue with standard cementitious products but at the time a surface DPM was commonplace for cement based systems, whereas they were not being recommended over the calcium sulphate screeds.
So, it was decided that we needed to investigate what would happen if we were to use surface DPMs. We worked with the manufacturers’ technical support teams and evaluated and assessed levels of risk and performance under different moisture levels within the screed. We also assessed the drying times of the screeds and methods of moisture testing, which will remain our intellectual property. In essence we found that at certain moisture levels it was deemed inappropriate to restrict the moisture loss from the screed and so would not advise the use of a surface DPM. When UFH was added to the equation we understood that moisture migration would accelerate and as a result the concentration of moisture at the surface, if a DPM was over coated, was very high. The higher the moisture content within a screed then, in loose terms, the weaker it is and the more likely that subsequent product applications could fail.
With this work completed we were then able to confidently recommend the use of a modified surface applied DPM, in much the same way as on a cementitious screed, with caveats of moisture levels having to be properly tested and underfloor heating fully commissioned.
Of all the occasions we have used this specification the only issues we have had are in areas where the moisture levels were higher than we recommended or where preparation of the screed was poor, leaving a weak upper layer, often due to over watering of the screed during installation.
If contractors are still nervous about these ‘new’ screeds then hopefully they can take comfort from the history and work carried out on these products to enable a fully functional flooring system to be installed. As with any screed, conditions play a massive role in whether they will cure and dry in the suggested timeframe. Improve the conditions, look at the tried and tested recommendations available and don’t be afraid of the screed!
Written by Martin Cummins
