Ern’s blog

Turning a natural-edge bowl

Timber: Silky Oak (prob. Grevillea Robusta).

Cut in winter. A half-log with pith removed. Rounded on the bandsaw and sealed with wax.

Intention: turn a natural-edge bowl (ie. with bark intact), with a recess in the foot for rechucking, and by way of form an extended Ogee.

Order of work and tools used:

1. Found a ‘centre’ to mount the top, ie. bark side, of the bowl.  At the centre, drilled a 2″ hole to take small Shark Jaws mounted on a Vicmarc 100 chuck.

The high point of the half-log was straight.  The shoulders were asymmetrical.  I moved the point a little towards one shoulder to adjust for this.  The change was measured by eye and by good luck worked.

An alternative is to mount the blank between centres, rough out, and adjust the tailstock point to get the low points of the rim level, and the high points ditto.  Between centres also allows you to adjust for figure, flaws etc.

2.  The base, or the flat, was even and appeared roughly 90 degrees to the blank axis point.  I drilled the hole at the top with a saw tooth bit mounted in a drill press.  It needed 20 mm depth in sound wood.  I had to go deeper to allow for the bark.  That meant the jaw flanges fouled the bark so I chiselled away enough bark and sapwood to ensure that the jaw faces engaged positively with the base of the hole.

3. Roughed the outside of the bowl  using a 5/8 bowl gouge. Then refined the shape with a 3/8 (rod width) cutting from the bottom up, except at the rim when you have to come top down to avoid chipping the bark off.  Formed a foot with a recess to avoid three mountings.  The recess was 6mm deep. The foot however had to be fairly wide to allow for the recess and while still within the 1/3rd rule of foot to bowl width, as the outside was shaped it was clear that the proportions were ordinary at best.  The recess was also deeper than the outside of the foot – not good form but I was concerned about the strength of the wood when remounted for hollowing.

4. Sanded the outside to 320 grit and finished with nitrocellulose sanding sealer – applied, wiped off, and cut back with 0000 steel wool.  Some of the wool got caught up in voids and bark and had to be picked out.  Sanding: with a light touch and a trailing edge of the power sanding pad rotating in the same direction as the bowl, you can sand out to the high points of the bark. Go in too hard and you’ll round over some bark edges.  Some power sanding with the lathe stopped was necessary to remove a groove in the wings – again with a light touch this will not be noticeable.

(At this point it was bed time; so I wrapped the piece in a plakky bag; next morning there were a few drops of moisture on the bowl outside but not nearly as much as I expected with such a wet piece so clearly the sanding sealer also works to some extent to seal the grain.  And to add a warning at this point: when you are green turning to finish and you’ve started hollowing, wrap the piece up even if you’re just stepping out for a cuppa.)

5. Reversed the bowl: expanded jaws into the recess with two layers of kitchen wrap separating the metal from the wood to avoid black marks.  Drilled a depth hole.

6. Started hollowing with a 5/8 bowl gouge, down to below the depth of the low points of the rim.

 

7. Within an inch or so of the outside of the rim, swapped to a 3/8 bowl gouge with the barest of ‘fingernail’ grind.  The smaller tool is less likely to chip the bark off and with a bevel that is shorter than a bigger tool feels easier to control to get the right curve. Position a spot work light so just the rim tops are clear.  Even then, go in gently but confidently with your gouge flute straight up, the tool shaft pointing upwards somewhat, and the left level bevel side pointing in the direction you want to cut. Raffan proposes a different technique for dealing with bark but this works for me.

Once down to below the lowest rim point, changed tools to:

8. Proforme slightly bent shaft, mounted in an Irons toolgate.  Alternated between the Proforme to remove stock, going down perhaps 2 inches at a time, and a Sorby multi-tool tear drop scraper to level any ‘corduroy’ and in shear mode clean up the surface.  Left as much stock in the centre as possible to reduce wall flex hence going down stepwise fashion.  As you do this, you are constantly checking wall thickness with a caliper and running your finger up and down the stationery bowl wall.  With figured timber, it’s often hard to see high and low points.  When you find them, mark them with a pencil so with the bowl spinning you can see where you need to be working.

9.  Last gasp: the toolgate saddle clamp clip rotated unnoticed and ripped the bark off the two high points :-{

 

With the poor form this piece was relegated to the category of a ‘good lesson’ ;-} 

What I got right was the even thickness of the wall at the highest and lowest rim points; about 6 mm.  Any difference here shows up like dog’s balls.

Normally with a good piece of timber I’ll round it, draw the outside dimensions on graph paper and sketch out some lines.  Had this been done, the poor form would’ve shown up before shaping cuts were taken and alternatives considered.  Eg. (1) a narrower tenon foot to match the clamping dimensions of chuck jaws to avoid marking; (2) if marking was unavoidable, then remount with a vacuum chuck to clean up the foot.  Raffan uses a friction drive with the bowl mounted between centres.

One comment about the outside form of a piece like this: you actually have two lines to consider, that from the highest point to the base and that from the lowest.  In this case, the former looked OK, the latter did not.