Tuesday, January 5, 2016

A Ming-Inspired Cabinet (23)

Post 23 in a continuing series...


I took the opportunity to clean up those 9/4 thick, 12" wide boards which came into my hands recently. First off, of course, is jointing:

The Aigner support table, which you can see mounted at the far end of the outfeed, is handy for long and heavy sticks of lumber.

Some readers have expressed concern about my placing pieces of wood on the jointer tables for photography, however these sorts of chunks of wood, around 75 lbs (34kg) each, place far more load on the tables -- a load that the machine is completely able to handle without any issue whatsoever.

The chain-sawn stock, as might be expected, did not yield quite as much usable material after clean up as you would get with stock sawn by other means. I think  that when all is said and done, the most that 3 of the 4 boards will yield is going to be 1.75" for thickness.

After planing, a little bit of mottling to be seen for sure:

The left hand board (above) has a bit more flatsawn in it than I might like, along with a few small knots. Not sure how I will make use of this stock at this juncture, but it sure is some fine looking material.

I recently picked up an i-Gaging 12" (300mm) DRO for my SCMI planer, and today I tackled the install of that device:

I modified one of the brackets which came with the kit so it could be attached to the existing visual depth gage on the side of the planer's table:

Before calibrating the readout, I planed some stock and checked dimension:

Then the DRO could be set:

The i-Gaging DRO attaches with a magnet at the back directly to the metal plate behind, and comes with a long connection cord so it was easy to mount on the planer.

Now, some may be wondering why I would want to put a second digital readout on the SCMI planer. Normally one would think one readout should be sufficient. Ah, but it isn't. This is SCMI. It's not German.

I have found that the SCMI's built-in readout is not reliable. Sometimes it jumps settings by just a smidgeon, and when taking a final pass to target dimension you find you have accidentally gone 0.01" under. It's annoying when that happens. Sometimes the SCMI readout jumps an inch or more out of adjustment. Sometimes it is perfect for weeks on end, and then suddenly it isn't. I don't trust it, let's put it that way. I have learned to run test sticks through before final passes, but that gets a little wasteful after a while, especially in light of the fact that it is a work-around for a defective technology.

That's not the only shortcoming however. The other defect with the SCMI readout is that to change settings from inch scale to metric, you actually have to remove four screws from the front panel and tilt the entire mounting panel forward so as to gain access to some tiny switches mounted on the back of the unit. A 5-minute job. Not what you would call a well-designed or convenient arrangement. If you're working within only one measuring system, this would be no issue, however I work in both systems and even when I'm working in one I may occasionally need to produce items, like jig parts, etc., that conform to measures in the other system. The hassle with the SCMI readout in how one changes scale has lefty me simply having to use a pocket calculator and converting inch to mm, or the other way. A little tedious.

With the new DRO, I'm hoping to gain improved accuracy, as the readout is taking data right off the linear scale and the table's actual height, instead of the stock device which takes data off of an encoder connected to the end of the main drive screw. I will be able to move back and forth between metric and inch as I like. The i-Gaging unit cost about $90.00, which was rather inexpensive if you ask me. We'll see how it works out over time.

The next order of business joinery-wise was to process the shachi sen, which in this case are parallelogram in section, and tapered both for length and thickness. In the past I've formed these compound tapered parts using a combination of router table and hand plane. This time I decided to get the milling machine in on the action, in the hope it might lead to a cleaner and more predictable result. If there's an avenue to be taken in quest - or promise - of a better result, I'll take it and see where it leads.

I needed to make a fixture to hold the wedging pin stock for milling work, and this fixture needed to attach to the sine plate. The sine plate comes with 3/8" NC bolt holes, 2" on center. I took the pin fixture, a chunk of bubinga, and milled a groove to hold the shachi sen. Then I marked off the bolt centers on the piece and clamped it into the Kurt vise. I eye-balled the drill center spur to one of those center marks and punched a starter hole through with a 7/16" brad point drill.

Then I used the DRO to move the table exactly 4" over:

And drilled the second hole, exactly 4" from the first:

Then I moved the other axis lead screw to offset 2":

Third hole:

And a final reset for position, heading 4" back the other way to return the DRO to 0.000:

4th of 4 could now be drilled:

Drilling holes on exact centers is one of those woodworking jobs to be tackled with care otherwise, and it usually takes some time and attentiveness to realize clean results. With the milling machine and its DRO however, the task of drilling holes on precise centers is a walk in the park.

The bubinga fixture was then bolted onto the sine plate and the plate shimmed so as to incline the block with its affixed shachi sen blank. Milling could then commence:

Another view:

A little piece remained in the middle to clean up:

The clamp, and small screw at the other end, is a simple means of keeping the pin from moving while the cutter is engaged.

Here's the first wedging pin, with one of the faces tapered down about 1/32nd, just a check to see how close it is getting to a fit on the trench:

It was a starter.

After milling one face of each pin, I needed to flip the pins over in the fixture and reset the sine plate, doubling the slope:

Doubling the slope is easy to do as all that is needed is to double the thickness of the shim stack.

Now onto round 2, the sine plate re-angled, and working the other face of the first pin to taper:

At the end of the day I had all the pins tapered on both sides, but still a hair fat for thickness:

I'll be tapering them down for width next, and when that fit is getting close to the mark, I'll revisit the tapering for thickness until I have achieved a crisp fit of the pins into their mortises. They have to fit well on all four faces. If the taper across the width is wrong, the joint won't draw up properly, and if the taper for thickness is too little the pin won't fully insert, and if the taper is done too much, the fit will be loose and the pin could crown crosswise when driven in hard. Fitting these is a bit of a dance.

All for today my friends. Hope to see you next time on the Carpentry Way. On to post 24


  1. Re: accuracy of gauges on power tools. As a complete dude some 30 years ago, the shop supervisor told us, "And don't trust the measuring indicators on the table saw or jointer; they are only put there to fool you". Heh. (Granted, they were not digital, but the principal holds.)

    1. Kerry,

      appreciate the comment. I think there are ways to get into a position of greater trust with machine measurement indicators, but I don't have all the answers quite yet.

  2. Kerrry, I think it comes down to being familiar with the machines, in a shop of several people, then the "Measuring indicators" should not be trusted and always checked with test cuts etc. , but on your own machines, not abused, then you become trusting on the "Measuring indicators" whether they are digital, analog or just tick marks on a tape measure.
    Chris...with measuring down to the .001ths have you needed to account for moisture in the wood around the joints? Exposing new wood surfaces to the atmosphere. Tight fit at the time of machining, but later after some time, when glue-up takes place, have you encountered any problems?
    Lastly, the planks you just surfaced look real nice. From the photos, they match each other well. How do they match up in color to the wood you are using now?
    Great work Chris, Keep the posts coming!

  3. Joe M,

    thanks for the comment and questions. I appreciate your regular readership.

    Your question about moisture and wood movement - - I'll bounce it back at you. If you cut a joint sloppily, and the moisture content changes, do you think it likely that the fit of the joint will somehow improve, or get worse, or what? Does being able to measure wood dimensions accurately somehow affect how one would design and cut parts in relation to wood movement?

    Moisture release, when a joint is cut, does happen, however, the effect is only pronounced when working with green lumber. The material I am working with is very dry and barring sudden immersion, is not going to experience significant dimensional change in the time I am working it. Just the same, I tend to keep joined parts together when they are in a period of stand-by prior to final assembly. That way they move together and any little micro-compressions can occur within the joint and be accommodated.

    And what's this talk of "when glue-up takes place"? What's that? Do people do that?

    I jest of course, but when is the last time you saw a glue-up on this blog? Not to say one won't be happening in the future, mind you.

    As for the recently-machined planks, I may or may not make use of them on this project, and if they are used at all, I'm imagining they will be part of some re-slicing for drawer parts or the like. I already have a mix of VG and curly bubinga on this project, and while I want to keep the mix composed of fewer rather than more elements, I suspect the material from the 9/4 slabs could be incorporated into the cabinets fairly seamlessly if the time comes for that. I think the color matches well as it is, and if it didn't there are solutions available for mitigating color differences when applying finishes.


  4. Chris, "when glue-up takes place" was a goof-up...I should have said assembly.....In my work...american and english period furniture, I feel there is a different level or type of precision. in that the level of precision that you do is of less tolerances. I agree that a sloppy joint is never desirable and of course avoided. But the period type of joint (dovetail, through, pegged mortice etc) are more forgiving than the keyed,wedged joints in your style of construction. Which I find very interesting. I "tip my hat" to you for your skill and craftsmanship.
    I was just wondering if the changes in moisture did effect the fitting of the keys and joints over time and if you found yourself "adjusting" the parts because of the change. I think I remember that your shop does get very cold during the winter. Wood movement effects us all, learning to deal or compensate with it is the trick. But usually shows it's head at the most inopportune time...after finishing or delivering!!!
    Thanks for taking the time to answer.
    Be safe

    1. Joe,

      thanks for the follow up. A cold environment has no effect upon wood movement. Only relative humidity, and the rate at which a given wood takes in or releases moisture will contribute to wood movement.

      As for 'period types of joints', I use all the joints you listed regularly and I have no concept of that use having some sort of 'period' aspect, any more than I would self describe as a 'blended woodworker' a term which I find odd. The joints you list are within the scope of ones I use, and I also happen to use a lot of joints which are outside the scope of Western woodworking traditions. Those nonwestern traditions employ all the joints you mention as being 'period'. To me, any sense of a 'period' aspect to joinery would be a choice to use an archaic form of a joint for some reason. You'll see that in Japan with certain temple rebuilds, where they keep the joinery to an archaic pattern because that is what was used previously.

      The frames I am working on right now use draw-bored pegged tenons, and there are through tenons in a number of places in this cabinet. I generally avoid the use of single dovetails outside of a couple of specific contexts, as I view them as a mechanically poor, glue-dependent connection.

      My impetus to reduce my dependence upon glue and to rely upon the joinery instead, a personal quirk of mine to be sure, has lead me to employ/develop certain types of joint types over others. If I was using glue, my joinery choices would be different, as would fit tolerances. Fit tolerances for glued joints, however, also require precision work.

      I have several pieces of furniture in my home which I have built - described here on the blog - and which employ the sort of joinery you are seeing in the current project, pieces like the tsuitate, coffee table and the vanity, along with the andon. I get to see how locked miter joints, etc., behave over the year, and the seasonal rh shift in New England is fairly pronounced. I have not seen an open miter once.

      To me these joints are very strong and yet able to accommodate seasonal movement without breaking. I would describe these type of connections therefore as resilient. Isn't that another way of saying they are forgiving?



All comments are moderated:

Anonymous comments will be deleted

Spam will be deleted

Comments containing links unrelated to blog content will be deleted