It was quite exacting work - measuring down from the laser level markings, making sure each rail was level, and making sure that the rails met accurately at the corners.

I gradually sped up but it was well past midday before I completed this work.

Sandy and Jim visited and I stopped work for an hour to have coffee and chat with them.

The next job was to work out an accurate way to establish the corners and the positioning of the soakers.

I need an accurate, straight, easily reproducible method for setting up the exact line of the soakers.

If I just guess and place them by eye then there is a good chance that the finished result will be ragged and disappointing.

The problem is that I need to run a straight edge over a distance of 2.7 metres.

A spare starting rail makes a good stiff straight edge and it is light and accurate.

However, it only spans half the height.

The other straight edge I have at my disposal is a piece of metal stud framing.

These are also, rigid, accurate and dependable. I cut a piece down to 2.7 metres.

The problem with these is that they are 75 x 25 mm in dimensions and for the concave angle on the bay windows they are too big to fit in while holding a level next to them.

An alternative suggested by Jim was to use my laser level to produce a vertical line - and I will try this if all other methods fail.

I even tried using a string line.

Part of the problem is that, with all the layers of flashing , some of the corners have lost there sharpness and the edge is a bit undefined.

The solution I have ended up settling on for the moment is based on the fact that there are 3 places going down each corner where I can feel the exact junction between timbers.

What I am doing is placing a nail at each of these postions and then checking that the starter railing straight edge is vertical between each of the pairs of nails, and that the metal stud frame straight edge accurately runs through all three nails.

Then I am using the railing straight edge to place some intermediate nails.

My intention is to use the array of nails and the straight edge to give me the exact positioning of each soaker as I work along.

I still haven't tried this yet so I am not sure if this is a workable plan.

I tried out my franken pliers to modify one of the soakers to fit my angle and cladding board size and this appears to be successful.

It looks like rain for next Saturday - so I might turn up with some tarpaulins and rig them so that I can keep working through the showers.