Hello Bill, Dave, Phil, Alan and TheOldFellow, (from various NGs)
Many thanks for your responses. Sorry I didn't say it before.
I was given a tip that someone, with potentially useful information, might lurking on some other NGs as above. if you are coming in on this thread, for the first time, and want to know what has gone on before, I can fill you in Yes, I did have some DIY feedback. But, basically, it didn't add to what we knew already.
I hope that I'm not breaking any netiquette rules by writing a piece, for anyone that is interested, that brings together my own experiences, for what they are worth, and those of other contributors.
Basically, you have to find the thickest flag, lay that, and then build up the adhesive under thinner ones, to get the same level.
That might seem obvious. But it is not so easy to do when the sticky labels with the thickness measurement keep coming off and you are laying a special pattern of flags of different sizes. The thickest flag may not be the first to be laid !!
If, as we have, your flags are different sizes in plan, as well as thickness, then it is a bit more complicated. I would suggest that the pattern is laid out, sufficient to include the thickest flagstone. If, from experience, you know how thin the thinnest layer of adhesive can be, then you put spacers under the thickest flagstone to mimick the thickness (or thinness) of that layer of adhesive. You then lay the thinner flagstones level with that thickest one. Then, when its turn comes take up the thickest one and lay it properly.
This all may seem obvious, when I've explained it, but wisdom in hindsite is a wonderful thing !
TheOldFellow - uk.d-i-y - said he used little wooden pegs to stop the tiles sinking before the adhesive set. Yes, I had thought the same sort of thing. If I had been laying them myself, I might have used small pieces of slate of different thincknesses for the spacers. Of which we have an abundance.
We didn't have the problems of a slightly undulating screed surface. If the wavelength of undulations were sufficiently long horizontally and sufficiently small vertically, I would not have thought it would matter. As for difference in height of adjacent flags, I have felt an error of up to 2 mm. was reasonable. Our builder suggested a wider grout, about 15 mm. to "mask" this. But, with different sizes of flags, this brings other problems of different grout widths - see below - unless you cut each flag to suit !!
So, we are on a learning curve, but are getting there. We have lost some flags in the process :-( There are other costs, like thicker layers of adhesive being needed. We averaged 3 sq.m. per 20 kg. bag, when we were advised that the spread should be 5 - 6 sq.m. per 20 kg. bag.
Another cost is the amount of grout required and flag cutting. Most of the flags, except the smallest, have been cut to fit assuming zero grout thickness. With different flag sizes this causes problems. It is impossible to get a constant grout width without cutting every flag. This takes considerably longer. So there are extra time costs, as well. For us, it is probably a "one-off", in a lifetime. We have about 200 - 250 flagstones to lay to get half way !
Our flags have nominal sizes in mm of 300 x 300, 600 x 300, 600 x 600 and
600 x 900. Obviously, these last, being much heavier, were much more difficult and also much more expensive if you get it wrong.
Finally, because we wanted coloured grout - brown - we have had to go to a company called Mapei as Bal seem not to make it anymore. We were warned by the Mapei suppliers that we should use white adhesive as the colour of ordinary grey could come through our slightly porous marble flags. Our builder, who normally uses Evostik cement based, polymer modified adhesive was sure that this would not happen with our flags, particularly as they are about 23 mm. thick. But, doubt having been sown in our minds, we felt we had no choice but not to take the risk. After all, we do not have the choice to get it wrong. The Mapei adhesive is much more expensive.
No one seems to be quite sure about this point of discoloration. Some felt it could happen with thin travertine, others felt the problem was sometimes translucency. I even managed to extract an informal and free opinion from the Building Research Establishment. They felt that it could happen !!
For both adhesive and grout for UFH, you need flexible material. We are using Mapei Kerqquick white with liquied Latex Plus, for the adhesive and Ultracolour grout.
John N.