Large Segmented Bowl - #5 Adding Rings & Fitting the Halves

To start working down from the top I need to build a base similar to what had been done for the bottom layers.  I keep a lot of odds and ends from various projects hoping to be able to use them elsewhere and this just happened to be an occasion where I could.  When I made the Snare Drum I built a multilayer base and could now re-purpose two of those layers to act as my base. 

Since the large face plate is in use and I am not going to take it off until the turning is done I used my smaller faceplate.  In the photo below starting at the top of the clamped stack you can see the smaller black 3” diameter black face plate just peeking out behind one of the clamps.  It has been screwed to an MDF layer which in turn is screwed but not glued to a segmented pine ring.  The MDF and Pine ring are not glued together so I can take the MDF piece off and check the joint between the top and bottom sections.  I will take a photo when I get there so if it does not make sense now it should then.  The top ring of the bowl (on the bottom of the stack) is centered on the Pine ring then glued in place which is what is going on in this photo.  Note the top ring like the bottom ring is made up of only cherry segments.  That’s a design choice to act as a visual cap on the bowl, or at least that is what it is supposed to do.

 

The photos below show, just like before the rings being added and the interior face being worked into final shape is stages.

After adding one more ring, the fourth and last one I spent quite a bit of time going back and forth between the two halves measuring and refining the joint where the top and bottom will be mated.  Not only do I need to match the inside diameter I need to make sure the curve flows across the joint.  That curve whose radius changes as is goes across the two rings to be mated has to look and feel continuous.  When I got to a point where things looked like they might match up I took the stack off the lathe, unscrewed the MDF disk from the Pine ring and set it next to the bottom half.  This photo shows how removing the MDF disk allows access to the inside which will enable me to check how well the two halves come together.

Setting the top section on the bottom and aligning them gives a rough idea of the bowl profile.

With the top in place I could see how close I got to matching the two halves.  It was really pretty close, better than I had hoped for a first attempt. 

Once I had checked out the joint and made some marks to indicate what additional shaping needed to be done I pulled the top off and screwed the MDF disk back on.  Below is just before aligning the top and screwing it back in place.  The green mark on the Pine ring and the MDF is my reference mark so things go back together the way them came apart.  There is also a blue one on the opposite side. 

You would think that using the reference marks then screwing the MDF disk back in the exact location where it came off would automatically center everything allowing me to go right back to turning.  That’s what I thought too, but was wrong.  There is just a little bit of wiggle room even using eight screws and putting the disk back in the same place.  To correct I just snugged two screws down then mounted the stack back on the lathe.  With that done the tool rest is set so it just touched the edge of the 4th ring.  Then as I spun the stack I could see what adjustments needed to be made to center the stack.  It took a little time but ended up just fine. 

I worked to match the exterior diameter of the bottom ring of the top stack to the top ring of the bottom stack which is pretty easy.  It just requires a light touch with the tools and frequent stops to measure so as not overshoot.  The inside diameter also needs to match up and the aforementioned curve has to flow across the joint.  It is sort of like a real marriage.  When it’s good it’s very good and when it’s bad it’s a disaster. 

A couple of fittings which required going through the steps of taking the MDF disk on and off got me very, very close.  The inside of top stack has just a tiny lip overhanging the bottom.  I think I will leave it that way and do the final work after the two stacks are glued together.  Below is a photo that shows how the two stacks match up with the red arrow pointing at the joint.  This is the photo I was referring to when I was talking about gluing the 1st ring on in the second paragraph.

At this point I am almost ready to glue the two halves together.  To help minimize any problems with glue removal I am going to put two coats of finish on the inside before gluing the two halves together.  With the finish on it will be much easier to do any cleanup afterwards.   However, this means that I have to have the inside completely done and sanded.  I start with 120 grit sand paper and work my way up through 320 grit.  Here is what the halves look like ready to finish.

I debated on the finish for a while but finally decided to use Danish Oil Finish.  It should give me a nice low-luster finish bringing out the grain of the cherry and walnut.  Below is the top stack showing the first wet coat of the oil on half of the piece.  I will put two coats on then glue the halves together.

Next up – Gluing the Halves Together & Turning the Outside

Large Segmented Bowl - #4 Rings & More Rings

With the second layer glued in place and the clamps removed it’s back to the lath to turn the corners off and make the layer round.  The left photo below is before I started turning and the one on the right is after I have turned the blank round.

I centered the 2-ring bowl blank on the next layer, marked the centering circle and the orientation lines as before.  Here if you look close in the center you can see one of the orientation lines marked with an “A”.  With the blank still in place I clamped the alignment blocks where they go. 

Below is after the bowl blank was removed.  This makes it a little easier to see the centering circle, where blocks are and how the bowl blank will drop in place.  Next is to add glue, drop the bowl blank in place, rotate it back and forth to spread the glue then make sure the orientation lines are in alignment, put the clamps on and finally check to make sure the layers are still in proper alignment.

The photo below shows how I have several rings in various stages of completion.  In the back is the rough-cut walnut and cherry strips for ring #7 setting to see if they are stable.  In the front left is a row of cherry and a row of walnut segments for ring #6 ready to glue together tomorrow.  At the left setting on the yellow pyramids with the metal clamp is the ring #5 that I glued together today.  To its right is ring #4 that I flattened today and will glue on tomorrow and finally at the right is the bowl blank with ring #3 that I just got done gluing on.

The next day after taking the clamps off I mounted the bowl blank on the lathe and turned the ring round.  The left photo shows that done.  Now with three rings glued on I can start to do some turning on the bowl’s inside.  On the right the bottom of the bowl is pretty well done and the second ring is roughed out.  There is a lot of refinement yet to be done especially at the transition between the bottom and the 2nd ring.  The top ring just has the inner face rounded and a small radius cut on the inner edge.  Lots left to do here but that will wait until I get more rings on.

I am not going to go through all the steps in making the rings and gluing them on since that has already been covered.  If anything that I have not covered comes up I will go over it but, in the meantime, here is where I added the 4th and 5th rings.

Making progress albeit slowly.  Here on the left is the 6th ring added and set up ready to round the outside edges.  The right image shows the 7th ring in place and the inside shaping in-progress.  The bottom layers are pretty much done while the outer three or four rings are still a work in progress.  The closer I get to the last ring added the less finished the piece is.

This is the 8th ring glued and clamped in place.  It marks the half-way point only 8 more rings to go.  If you look close at the bottom ring you can see the step between it and the layer above is getting smaller.  That is starting to cause me some problems in that I don’t have very much room to clamp the centering alignment blocks on.  It’s only going to get worse as the step continues to get smaller.

The left image has the 8th ring on with the outside rim rounded while the right image shows the 9th ring after rounding the outside rim and doing some interior tuning.  The pile of shavings is the result of interior turning and just sits there sliding along the bowl surface as I turn.

As I talked about when gluing the 8th ring on I was running out of room to clamp the centering alignment blocks on.  Well the left image is the 9th ring with blocks barely attached and the right is the 10th ring.  As you can see there is no way I can clamp the blocks on this ring.  That means that I will use a rub joint to help get an initial “tack” to reduce the piece slipping around as I clamp them up.  A rub joint is where you slide or rub the two surfaces back and forth until the glue becomes tacky and grabs.  It makes it easier to keep things aligned as the piece is clamped.  

Here is the 11th ring.  This ring represents the largest diameter of the bowl.  From here on in the rings will get smaller. 

The left photo is with the 12th ring in place.  It is just a bit smaller than the previous ring and that’s because I am working my way down toward the opening at the top.  Because the interior surface of the bowl is curving back I am having to get into some contortions to do the turning.  All along I have been finishing turning on the inside.  Now the interior surface is about 95% done up through the 8th ring and 90% done through the 11th ring while the 12th ring is about half done.  The right photo shows just how tall the stack has become.  There is right at 12” between the faceplate and the outer face of the 12th ring.  Unfortunately, as I was working on this ring I started getting some vibration and minor chattering while turning.  Changing the RPM’s, using a light touch and sharpening the tools helped but this means that I have gone as far as I want adding rings using this setup.  There are a couple of techniques that I could use to steady the bowl and dampen the vibration but I would still have the awkward turning position.   Then there is the problem of weight, as it sets including the cast iron face plate the blank is pushing 50 pounds.  Holding it up at the lathe in proper alignment with one hand while using the other to start the spindle threading into the face plate is getting pretty close to making my eyes bug out.  To resolve all these problems, I will turn the remaining four layers separately then marry the two halves together.  In retrospect I probably should have made the split between the 11th and 12th rings right at the widest point.

Next up – Adding Rings & Fitting the Halves

Large Segmented Bowl - #3 Face Plate Sandwich, Making/Attaching Layers & Turning

Setting the base ring aside I stated working on the sacrificial sandwich that acts as the transition from the cast iron face plate to the actual bowl.  It is made from two layers, a piece of 5/8” plywood glued to a ¾” piece of particle board.  The plywood layer is screwed to the cast iron face plate which is why I used plywood as it holds screws better than particle board.  One thing I do not want to happen is for the screws to work lose and potentially let go allowing the in-progress bowl to launch itself at me.  I had a bowl break free once and that made for a very bad day.  Here the photo shows the two layers glued and screwed together.  The screws will be removed once the glue sets overnight. 

Once the glue cures the sandwich is run through the thickness sander to make sure the faces are parallel.  The bottom ring of the bowl will get glued to the particle board layer.  When it comes time to separate or part off the bowl I will be turning through it.  Since the screws only go into the plywood it will be easy for me to not hit one accidently.  Hitting a hardened steel screw with a lathe tool is not a good idea.  Next the sandwich is centered and securely screwed to the face plate with a whole bunch of screws, 15 to be exact.  

This assembly is then mounted on the lathe and each layer is turned to a true circle.  Next an MDF reference block in the center of the bottom ring is attached using double face tape.  I used the ring joints to mark it to give me my center then measured the particle board diameter and drew a couple of concentric circles on the ring face.  One the diameter of the particle board and a second just a sixteenth or so larger.  These will allow me to center the bottom ring on the particle board when gluing the two together.  

A layer of glue is spread on the base layer then the face plate assembly is carefully centered on the pencil rings.  Eight clamps hold things in place, but I have to be careful when tightening them as the glue acts as a lubricant which allows the pieces to slip around and out of alignment.  It just takes some time keeping an eye on things and making small adjustments as the clamps are tightened. 

Once the glue had cured I mounted the assembly on the lathe and was ready to go.  The photo on the left is taken from the lathe headstock side while the right one is from the face where rings will be added and the turning will take place.

Below the left photo is a little closer look before I started turning and the right one is after I have knocked off all the corners and trued up the ring. 

Next up is the second ring.  I am fortunate to have a pretty good stack of thoroughly dried lumber.  Some is rough sawn, some is surfaced on the two wide sides and some also has one edge ripped straight.  Well, sometimes it is straight.  Unfortunately, a fair amount of the lumber I will be using for this project does not have a straight edge to reference off of.  This piece of walnut is a case in point as while it was flat and not warped both long edges were nowhere near straight.  They were way too far out to even think about using the hand plane to clean up.  The solution is to take an 8’ level and use the straight edge to mark a reference line.  I then carefully cut along the line using the band saw getting close to straight and then clean up with my long hand plane.  Last, as shown below is to clamp the level to the rip fence on the saw to give me a long surface for the walnut piece to register against.  Flipping the board over after each pass ended up giving me straight parallel edges to work from. 

To make the second ring and all the subsequent rings except for last one I need 12 cherry pieces and 6 walnut ones all the same size.  The process is the same as what I went through to make the bottom layer with one important difference.  I do not have a cut slot on one edge.  Not having that slot allows me to make more efficient use of my wood as there is not a specific Inner or Outer Face.  In the images below, you can see what I mean.  The top photo shows the cuts required for the bottom layer where each segment has to begin with a new edge so the slot is always on the inside.  The bottom photo is for the second layer whose segments can be cut in a continuous string with no waste.

Once I get the 12 cherry and 6 walnut pieces cut I can clamp them together to do a test fit to check for gaps or any other problems.  In this case everything looks good.  Last is to chalk the sides that face up so I make sure none of them get reversed in the controlled chaos of gluing the ring together 

After letting the glue cure overnight as before I ran the ring through the thickness sander to flatten.  Next is to center the ring on the stack.  That’s done by aligning the joints with the previous layer and very carefully measuring the distance between the edge of the to-be glued layer and the turned edge of the previous layer.  If nothing is in error when that distance is the same all around all the joints should align.  If they don’t then there is a problem somewhere.  Here you can see everything is good.  Last is to draw a pencil line around the first layer on the new layer. 

The glued-up bowl stack is removed and alignment marks are drawn on the new layer.  The marks cross the just made pencil line and are centered on the segments.  I make two arcs using a compass working from each joint.  Any error will show up as a gap between or overlapped lines. 

Glue-up is next.  Note that there are alignment blocks held on by orange bodied F clamps.  They are set on the centering circle and keep the two pieces in place while torqueing the grey BESSY clamps down.  Because the glue acts like a lubricant between the gluing pieces without the blocks they will slide around as the clamps are tightened.  Without them it’s a real pain to try and keep both parts aligned.  Once the glue has taken an initial tack the blocks can be removed and the glue allowed to cure.  Looking close you can see how the joints between layers are offset by half. 

Next up – Rings & More Rings

Large Segmented Bowl - #2 Making the Base

Because there are a lot of cuts that have to be precisely made I started out by running a check of the table saw blade to make sure it was cutting square to the table.  First step is to cut a couple of scrap pieces that have parallel long edges then stand them on their cut edges like below.  The green tape is my cutting reference mark set against the crosscut fence.  Everything looks good on the left image but if you rotate the left board 180 degrees then the out of square becomes visible as shown on the right image.

Looking closer the below left image shows how much the blade is out-of-square.  Some minor adjustment of the blade, in this case rotating the blade to the right brings the pieces together and all is square.

A trip into my wood storage gave me a good group of cherry and walnut pieces to start with.

In the last post I decided to float a center panel in the bottom bowl layer.  The top image below shows a section through the bottom.  For reference the center panel is just under 3/16" thick.  To make the slot I made one pass on the table saw then moved the fence slightly to get the slot width I needed and made a second pass.  The bottom photo shows that setup.  The yellow and black thing is my magnetic fence.  I set it up to apply a slight pressure to keep the board pressed against the rip fence on the right.

Once the slot is cut I can start on cutting the wedge-shaped pieces that will get glued together to make the bottom layer.  Here is the jig with the slotted bottom blank in place.  I will use the jig to cut 18 pieces for each of the 16 rings that will make up the bowl.

This is the first cut which gives me one side of the wedge and a triangular waste piece.

Before making the second cut I flip the piece 180 degrees so the slot that was facing me now is away from me.  The piece is then slid to the right and clamped in place at the Outer Face Length I obtained from the spreadsheet. 

A stop that has a matching angle is then clamped in place so I can make all 18 pieces the same size.

 

The cut is made and I have one piece the right size.  For those of you with sharp eyes this first piece is not from the same place in the board as the previous photo.  That’s because when I checked the piece’s size it was a couple hundredths of an inch small.  I must have bumped the stop or something else.  Anyway, I reset the stop, recut the first angled side and made the cut you see here to get the right sized piece number 1.

Now it’s on to the second piece.  Because of the slot there is a specific inside and outside to the pieces.  This means that I have to flip the board 180 degrees back to its original position, cut a new first taper in the top view below then in the bottom view flip the board 180 degrees, push it up to the stop and make the second cut to get the second completed piece.  Two down and 16 to go for this ring.

In the image below, I have the first four pieces cut and arranged along with the waste triangular pieces.  The other layers without a defined inside and outside will be cut without flipping thus eliminating the waste triangle resulting in less waste. 

With all 18 pieces cut I made the floating circular panel and did a test fit.

To make the center disk float in the slot it could not be glued in-place and that presented a problem.  Gluing up the rings is a really messy process and glue gets EVERYWHERE.  There is no way I could glue the ring up and not get any in the slot which would glue the disk permanently in place losing its floating attribute.  I thought about doing a two-step gluing process where I would only glue up half the ring at a time then insert the disk and very carefully glue the two halves together.  Sounds good but if the slots are not aligned dead on then there would be no way to get the disk in place plus I would probably still have glue to clean out of the slot so the disk would fit.  A couple of other ways all had their problems too.  What I finally ended up doing was to finish the disk and the slots before the gluing all the pieces together.  Here you can see the disk and ring sections prefinished with shellac where I do not want the glue to stick.  Now when I do the actual turning I think that much of the finish will be removed but the slot to disk surfaces should not be permanently glued together.  At least that’s the theory.

Gluing up this many pieces, getting them aligned and clamped can best be described as semi-controlled chaos or a mad race against time before the glue sets.  According to the time stamp on the photos it took me right at four minutes from the time I started applying the glue until everything was together and the glue mess cleaned up.  I have to say that a couple of dry runs beforehand did not hurt my time either.

Here the first ring, which is the base with the glue-up finished and ready to set overnight before proceeding.

The next day the clamps are removed and several light passes made through the thickness sander to obtain the needed flat parallel faces.

Next up – Face Plate Sandwich, Making/Attaching Layers & Turning

Large Segmented Bowl - #1 The Beginning

In a recent trip to Santa Fe, NM I saw some large segmented bowls in various Native American motifs.  I have made several segmented pieces (below) but nothing larger than about 11” in diameter. 

Intrigued I wondered what making a bowl that pushed the limits of my lathe would entail and what kind of obstacles I would encounter.  Since my lathe has a maximum turning diameter of 20” I decided to shoot for a finished diameter of about 19”.  Working in SketchUp I created my finished shape paying no attention to the size but concentrated on what the profile should look like.  Most of the pieces I have done have had a fairly small opening.  In this piece I wanted to go with a design that is reminiscence of a wide mouth storage container.  When I got something that I was happy with I scaled it to be 19” in diameter.  With that done I measured the height and found it to be about 12”.  Here is the piece in profile and looking down  slightly.

Next is to create a section as though one had sawn the bowl in half vertically .

With that done and deciding to use ¾” thick rings except for the bottom I could lay out each layer’s  diameter and segment width.

The next decision is to decide how many segments to make each ring out of.  To give me some help in this I turn to a spreadsheet I made.  Once I enter the radius of each ring it shows the Outer Face Length for each segment by ring in a variety of segments per ring.  For this piece I chose to use 18 segments  per ring.

Taking this information, I can use another set of calculations to get the particulars of each piece in each ring plus the total blank size I will need to cut the ring segments from.

 The headings like "Outer face length" in the above calculations are based on the drawing below.

To get an idea of what the piece would look like when done I built each ring using cherry for the body  and walnut for the feature pieces.  Here are a couple of views on what that looks like.

The unturned rings give a pretty good idea what the finished piece will look like but I wanted to get a better representation as I had two different designs I was considering.  That required me to intersect the unturned rings with the smoothed profile.  I have to say for the computer to calculate  the intersection of all those facets it did take some time.  I probably could have gotten by without doing this but I could so I did.

Once I had decided on the feature pieces size it is back to the spreadsheet to set up the last two  columns to calculate the blank length needed for the cherry and walnut pieces.  While it looks like everything is set in stone that is not really the case.  As I build the bowl adding the rings and turning if things do not look right I can and will make changes.

I usually do a double check on the Outer Face Length by taking the drawing one step further to see what it comes up with.  Below is a copy of that drawing.  If you compare it to the spreadsheet there is a difference of a few hundredths of an inch but that does not really worry me as for making the rings is it within tolerances.  If there is a big error then I need to take a look and see why.  It is my guess that when  SketchUp gets into that fine a measurement then it may not be as precise as the spreadsheet which carries out the calculations to a whole slew of decimals.  It’s 15 places if you are interested.

One more detail needs to be addressed and that is where all the pieces come together in the bottom layer.  There are 18 pieces so that’s 36 cuts I have to make and have them all come at an exact center point.  Now I am pretty meticulous when it comes to setting things up but if I am off by even 3 thousands of an inch then when the pieces all go together then there is going to be a gap of just over a tenth of an inch.  I could cut or drill a circle in the center then put a plug in but have had some problems with that.  Since wood moves around with humidity changes what usually happens is the piece cracks or ends up with a hairline gap.  This is going to be a pretty big piece and as such the wood movement is going to be more so I decided to take a page from cabinet making and float the center piece in a slot.  It is similar to a raised panel door where the panel floats in a frame.  Here is what a section through the bottom would look like with the slotted pieces and disk in place.

Next up – Making the Base