Hi all,
A few questions have been playing on my mind of late regarding semi-headed arched frames. I thought I'd get your expert opinions to untangle them.
So here's the deal... I've made a couple of semi-headed (180 degree arched in the vertical) door frames and doors recently, essentially following the method set out in George Ellis's Modern Practical Joinery. I machined the head in 2-3 flat laminations, segmented and offset across the width. In one case I made up a square section (all laminations equal width) and machined the rebate after. In another case I used a diminished outside lamination to cut out the need for rebating (again, as per Ellis).
In all cases I machined the curved segments to size using a small, banana-shaped MDF template before gluing them together, meaning that each laminated segment had to be perfectly aligned with the next in order to create a perfectly square section. I then finished off the head with a compass plane and sandpaper. The doors and frames turned out nicely.
However, since then my colleague has made another arched frame using a different method. He laminated the segments as straight lengths and glued them up into a kind of trapeze before machining the curve, which he did in one go, using a full-size curved MDF template. This has the advantage of reducing the curve work to a single process but it strikes me as being more wasteful in that it uses more MDF, and eliminates the possibility of getting two short segments out of the same width of timber (imagine the two segments drawn on the timber, one above the other, basically spooning!).
So, question 1: which is better/quicker/more reliable? How do you make your semi-headed frames?
I note that New Yorkshire Workshop do a kind of hybrid of the two methods, cutting the curved segments close to the finished width, assembling the segments then doing a final pass on the spindle moulder using a full size template. Not sure I want to go down that route as it seems like making two templates = twice the work. NYW also tenon-join the segments rather than laminating. Is there an advantage between one method and the other?
Question 2: I'll shortly be making a bunch of natural-finish arched window frames and therefore can't have any screw holes/plugs showing on the faces. Can I rely on double-sided tape to hold down the template for spindle ring-fencing? The answer to this question has a bearing on Q1, since the template screw holes can be hidden on the inside of the lamination using my technique, whereas my colleague's technique requires him to screw the template through the face of the finished arch section (he has ended up with a few visible screw holes -- not great).
Question 3: Ellis recommends screwing the laminations together through the face. But if the laminations cannot be screwed together (because of the unpainted finish), will the glue be strong enough on its own? We usually use cascamite for edge jointing and exterior lamination.
Anyway, sorry for writing an essay. If anyone has any insights into this, I'd be very appreciative!
A few questions have been playing on my mind of late regarding semi-headed arched frames. I thought I'd get your expert opinions to untangle them.
So here's the deal... I've made a couple of semi-headed (180 degree arched in the vertical) door frames and doors recently, essentially following the method set out in George Ellis's Modern Practical Joinery. I machined the head in 2-3 flat laminations, segmented and offset across the width. In one case I made up a square section (all laminations equal width) and machined the rebate after. In another case I used a diminished outside lamination to cut out the need for rebating (again, as per Ellis).
In all cases I machined the curved segments to size using a small, banana-shaped MDF template before gluing them together, meaning that each laminated segment had to be perfectly aligned with the next in order to create a perfectly square section. I then finished off the head with a compass plane and sandpaper. The doors and frames turned out nicely.
However, since then my colleague has made another arched frame using a different method. He laminated the segments as straight lengths and glued them up into a kind of trapeze before machining the curve, which he did in one go, using a full-size curved MDF template. This has the advantage of reducing the curve work to a single process but it strikes me as being more wasteful in that it uses more MDF, and eliminates the possibility of getting two short segments out of the same width of timber (imagine the two segments drawn on the timber, one above the other, basically spooning!).
So, question 1: which is better/quicker/more reliable? How do you make your semi-headed frames?
I note that New Yorkshire Workshop do a kind of hybrid of the two methods, cutting the curved segments close to the finished width, assembling the segments then doing a final pass on the spindle moulder using a full size template. Not sure I want to go down that route as it seems like making two templates = twice the work. NYW also tenon-join the segments rather than laminating. Is there an advantage between one method and the other?
Question 2: I'll shortly be making a bunch of natural-finish arched window frames and therefore can't have any screw holes/plugs showing on the faces. Can I rely on double-sided tape to hold down the template for spindle ring-fencing? The answer to this question has a bearing on Q1, since the template screw holes can be hidden on the inside of the lamination using my technique, whereas my colleague's technique requires him to screw the template through the face of the finished arch section (he has ended up with a few visible screw holes -- not great).
Question 3: Ellis recommends screwing the laminations together through the face. But if the laminations cannot be screwed together (because of the unpainted finish), will the glue be strong enough on its own? We usually use cascamite for edge jointing and exterior lamination.
Anyway, sorry for writing an essay. If anyone has any insights into this, I'd be very appreciative!