One of these days I could have an operating layout in the garden and I'll need some coaches. For me, building coaches is a lengthy process so I thought I might make a start now so that there was a possible chance that something might be ready in time. I actually made a start last year but I had just doubled the sizes I had used for S scale coaches a year or two ago and I wasn't too happy with some of the compromises I had made in S scale since they were amplified by the up-scaling. The one thing that did worry me was the depth of the inset of quarterlight glazing. In the S scale coaches I had taken the easy way out and made the glazing insets for droplight and quarterlights the same to eaase the construction of the body sides. However, the more I looked at pictures of Caledonian coaches, the more it became obvious that quarter light glazing was actually getting close to being flusher. One of my drawings of a MacIntosh caledonian coach had quite a clear cross section of a window ...and I used this to trace a drawing in Draftsight and work out dimensions in 1:32 scale. It worked out that the quarterlight glazing was inset by 0.9mm from the outer surface and the droplight frame was inset by 1.25mm. So I opted to try something I haven't done before on the CNC mill and to mill both sides of a piece of material. It means you have to flip the material in register so that the machining of both sides lines up. I started on the rear of the coach side and machined the recesses for the droplight glazing - 0.35mm deep, the difference between the droplight and the quarterlight glazing insets. I have only machined the areas necessary to save time - the unmachined parts will disappear when the quarterlight apertures are cut out. The part was also machined to be 3mm oversize all round. Then the sheet was flipped over on the horizontal plane and the alloy guides are set to -3mm on X and Y axes to match the 3mm oversize of the sheet. The alloy strips are aligned with my cheap and cheerful toolmakers buttons - thick brass rings with oversize holes for the roundhead woodscrews. I use a 6mm diameter rod in the mill spindle, slacken off the woodscrews holding the buttons, place the spindle in approximate position for the X axis and move the button until the alloy strip bears against the 6mm rod, then tighten the woodscrew. Repeat this for the other button on the X axis and set the X axis value to zero. This sets the face of the alloy strip at Y = -3. Do the same for the Y axis and that strip's face is at X = -3. The material is held by double sided tape. The first big cut is to surface the material to set the correct overall thickness. The raised parts are the material for the bolections - similar to David Jenkinson's bolection method where he stuck rectangles of styrene over quarterlight apertures to form the bolections by cutting out the window aperture and shaping the inner sides. The lower panel has also been cut. The next operation cuts all the upper panelling and also cuts out the droplights and quarterlights with the waste material being left in situ. The material to form the bolections is also apparent. This is a closeup showing the bolection material round the quarterlights. The bolections have now been formed into their final shape. The bolection on the left was a cock-up. I used the code I had generated a year ago but forgot that I had altered the window shape to be more accurate, so the cut was made out of alignment. I re-wrote the code and the other bolections were the result. The final operation was to drill pilot holes for door stops, handles, grab irons, locks and hinges, and to engrave the door lines. Then the piece was machined to size, getting rid of the 3mm margin all round. It was then removed from the machine table. ...and the acid test was to turn the piece over and see if the rear apertures lined up - and they did. The shadows give the impression that they are a bit off, but they are actually pretty well spot on. Absolute accuracy is not really require for the glazing slots, but it's nice to get them where you want them. I gave the part a quick scoosh of grey primer to make things look a bit better to the camera. This is the centre section of a 48 foot coach side. The coach will finish up 18" long and my CNC mill cannot cope with that length, so the side has been cut into three sections about 6" long to coincide with door lines to disguise the joins. The support parts for the sides will be in two parts, 9" long, which my CNC machine can just accommodate. I'm waiting on some glazing material coming so I'll post a picture when the glazing is fitted to show the effect properly. This part will be treated as a test piece because of the problem with the left hand bolection, but the only way I could check my bolection code was to try it and by the time I had found out it was wrong it was too late. Jim.