Friday, November 30, 2012

Mizuya (10)

Many hours of drawing work since the last post, however the reader likely won't note too many obvious visual changes from the previous iteration. I have revised the shelving in the main compartment, added shelves to the lower compartment, and have done a lot of detailing with joinery. It is not terribly straightforward to configure everything when the conditions that I have set are for the piece to be robustly constructed, use no glue, and be demountable without reliance upon metal fasteners. I do like the challenge, and it does come with a fair amount of head-scratching. This project reminds me of the demountable doctor's desk I built 7 or 8 years ago. That project was a bit of a saga, hopefully this one will have a smoother journey.

Here are some views:

Two shelves, will probably be of solid 3/4" material with tenoned and mitered breadboard ends, though I may do frame and panel shelves as an alternative:

Two shelves for the bottom compartment as well:

All the shelves will be height adjustable, the bottom one's though only be one or two positions up and down.

Here's a view with all the doors and drawers removed:

I have started working my way through the design component by component, sorting out the joinery layer by layer and case by case. After three days of drawing, this is how far along that process is - not far!:

The framing around the drawers with the insert runners is some of the most complex in the piece - all so that it might be repaired easily down the line.

A view from below - I'm making a fair amount of use of wedged through-tenons:

Detail of the twin tenoned connection, with mitered tongue, which is seen where the drawer rails meet the post:

The tenons go through, and will be wedged and trimmed flush.

Oh, and in other news...the jeweler has produced a prototype of the hinged door pull:

It's still raw copper and hasn't received any patination yet. The pentagon portion of the mark has been dropped. not sure how I like it yet.

All for today - have a great weekend!  On to post 11

Tuesday, November 27, 2012

Getting a Handle

I spent a few hours this morning trying to render a handle in SketchUp, and due to the shape I was after, curving and continuously changing in diameter, the drawing was not especially profitable. I did spend some time looking at various shapes and sizes of Japanese tansu drawer handles, and thought it would be worth making an account here. There are accounts out there of this form of traditional hardware in a few different publications, most notably the Koizumi book, however after a bit more research I think I might be able to fill in the picture a bit.

What follows then is meant to be a glossary of tansu handle types, or hikite kanagu (引手金具) as they are termed. Hikite (pronounced: he-key-tay) refers to 'pulling with the hand', and kanagu means 'metal hardware'. The word hikite can mean pulling in any direction, so hikite are also used to describe the recessed pulls found on sliding doors. Another term for drawer pulls is totte-kanagu (取手金具), the word totte meaning 'seize(d) with the hand'.


Kaku-te (角手) : literally, 'square/angled' (角 ) 'hand' (手) , these have an angular shape with 90˚ corners:

 Considered, by some sources, to be the oldest form of handle used on tansu.

The squared form seems especially common on funa-dansu, shipboard chest, as in this example, a Sakata-made (a town in Yamagata prefecture famous for tansu) coin and document safe, in Keyaki:

As you'll see with this and other pull forms, it is the handle's shape that is the primary categorizing factor, not the shape of the back plate, which may be absent, as in the second example above, and replaced by a pair of escutcheons.

Mokkō (木瓜): literally, 'tree' (木) with 'melons' (瓜), referring to the Japanese Quince, Chaenomeles japonica. The characters are also read 'bo-ke', especially if you are talking about the tree and not the form derived from the shape of the fruit. Confusingly, the same characters, 木瓜, are used also to describe the Papaya tree today. More confusingly, these handles are described in English-language Japanese furniture books simply as 'melon-pulls',  when in reality a quince fruit has nothing to do with a melon at all, the tree being a member of the rose family.

The Japanese quince has small, apple-shaped fruit, 3~4cm in diameter. Here's what the quince tree's fruit looks like, cut open:

And here's the handle shape based on that:

A more modern styling:

The form may be based on the interior seed cluster shape, or on the exterior shape of the fruit. Mokkō is a popular motif in Japanese art. There are several variants of Japanese crests, or mon, using it - here's one:

Kaku-mokkō-gata (角木瓜): Putting the angular kaku-te form together with the quince form yields this type:

Yama-shina (山科): might be best considered a variant on the mokkō. A cusped type of handle associated to Kyoto, Yamashina being the name of a ward in that city.


Kushi-gata (櫛型); literally, 'comb' (櫛) 'shape' (型).

For reference sake, here's a typical Japanese wooden comb:

And a drawer pull based on that form:

Another, mass produced and inexpensive:

This form is widely used in many Japanese traditional objects, like this shoulder plane:

Gunbai / gunpai (軍配): refers to a wooden, metal, or wood-metal military fan, carried by generals and used in feudal era battles to direct troop movements, ward off arrows, and act as an impromptu sunshade:

Gunbai are also currently used by sumo referees to point to the winning side after a bout is complete.

For some reason one also encounters the word 'gumpai' to refer to these fans, however as there is no terminal ~m consonant in the Japanese language ('軍' is gun, not gum), this is one of those cases where an erroneous transliteration has stuck in English language materials. A couple of western books on tansu continue to use the misnomer, as does at least one website I came across.

Here's a real gunbai, lacquered black and red and with mother-of-pearl inlay:

The drawer pull which employs the same form:

A Paulownia cabinet, kiri-dansu, with the gunbai pulls:

Hiru-te (蛭手): the 'leech' (蛭) pull. In Japan, the leech being modeled is a terrestrial leech, Haemadipsa zeylanica, also called a yamagiru, (山蛭) lit. 'mountain' (山) leech. I did a lot of hiking in Japan but never saw one, so they must be local to places in which I hadn't lived, or they are uncommon. Kinda glad I didn't meet one, though in reality most leeches dine on decomposing bodies, not living ones.

Leeches, perhaps, don't immediately bring inspiration to mind in terms of hardware:

However I am guessing that the key point is the shape of the sucker:

They have these suckers, in one form or another, at each end. Oh, and a brain at each end of their body as well. Unusual little gippers.

The is the drawer pull handle based on that form:

The handle is the leech body with suckers depicted on each terminal end. In this form, the handle always enters the back plate's loops from the inside.


Another, of a more modern cast:

One more, from a sword storage cabinet:

Warabi-te (蕨手): warabi is bracken, or fernbrake, teridium aquilinum:

The furled fronds of this plant, among several fern species are harvested as a vegetable, are referred to as 'fiddleheads' of course. The kanji for warabi (蕨) is not in common use in Japan and the plant name is often written phonetically in Japanese either in hiragana (as わらび) or katakana (as ワラビ).

The tansu drawer pull based on the plant form:

Another, by Avigal David:

I would say that the warabit-te handle is the most common form encountered on tansu, and could be considered one of the 'classic' forms, along with the squared, leech, quince, and military fan styles.

Hana-giku (花菊): Flowering (花) Chrysanthemum (菊) pattern. Here, the back plate is the primary decoration:

Kan (鐶): ring-shaped pulls. An old form, and quite common, especially on skinny drawers. Here are some variants:

This one is in copper, by Avigal David:

This one has the hana-giku style back plate:

Hangetsu (半月): the half-moon shaped pull.

I guess it might be called a 'D-shaped' pull in English.

Hamaguri (蛤 or 浜栗): a pull which is shaped like a saltwater clam, meretrix lusoria. The second way of writing the word hamaguri, '浜栗', is the more interesting to me, as it is referring to a clam as a 'beach' (浜) 'chestnut' (栗), which I find amusing.

Here's the mollusc:

Here's the handle:

Ume-komi-shiki (埋込式): recessed pulls. also called hako-hiki-te (箱引手). One of the more modern forms of tansu hardware.


Uncategorized: at the the time of publishing this blog, I was unable to ascertain the correct technical name for these pulls, however that will change after a bit more research:

This one looks a bit like the hiru-te form in certain aspects, however the handle ends are different

This one has a shaped plate for a pull:

I can't quite put my finger on what this handle form is called, however it is very similar to one used for pagoda corner chimes:

I've also seen something similar on Japanese bushi helmets:

I have a feeling the shape  relates to the pine tree. I just looked through an entire book of Japanese design motifs but couldn't find a match. Can any readers help out?

Thanks for coming by the Carpentry Way.

Sunday, November 25, 2012

Mizuya (9)

Many hours of drawing have ensued since the last post. Design is not complete yet, but is getting closer I do believe.

The point of focus in recent days for the drawing work has been on the technical/structural side of things. I find that by resolving details at that level, new information and ideas come to light which affect the aesthetic side of the equation. Sometimes I find that an aesthetic I was chasing simply proves to have no good solutions, within the design parameters I have set, and two steps back are necessary. I can't claim it's the most organized and seamless way to go about design, in fact, it's a bit messy and chaotic at times. In the end, I feel it works for me though. The main thing is to maintain patience and work through the design fully before cutting up any material.

I've been working on the drawer runners. Given the NK drawer influence on my drawer design, which resulted in my drawers having lignum vitae slips on their lower side edges, I had thought that having corresponding lignum vitae runners in the case would ensure a virtual lifetime of smooth running. If lignum vitae is good enough for hydroelectric turbine bearings, it should suffice for a cabinet - ahem!

My initial foray into the runners was to place a couple of 1"x1" scantlings on each side of the drawer, like this:

Lignum vitae is a rather precious material though, and 1"x1" is a more generous size than I would prefer to us for this application. I'm using it for its toughness and supreme self-lubricating qualities, not for load carrying- which it would excel at of course, well in excess of any need here. I thought about it for a while I decided that thin inserted strips would be more ideal for the runners, somewhere on the order of 3/8" (9mm) in thickness. Then I thought about it some more and one of the problems with using thinner runners is that, by themselves they would not be adequate to carry heavy drawer loads, and worse, they may warp. If I was going to use thin pieces of lignum vitae for runners, it made sense that those strips would be held in some sort of carriers which handled the bulk of the support task and would restrict them from moving in some manner. I was also thinking of the wear and tear issue, and how it would be nice if the insert strips could be removed, so that they could be shimmed or replaced if need be. Just on principle, this seemed like a good direction to head.  In a way it is the same philosophy as for insert tooling.

So, replaceable insert running strips it would be. This meant however that some machine screws will be introduced to fasten the strips in position. I considered various all-wood connection solutions, but the screws seem the most practical and durable in this case, and allow for easy replacement years down the line by anyone who can make a rectangular section strip of material.

I also took a look at the drawer slips themselves and reasoned that the bulk of the weight is to be carried by the undersurface of the slip, and the side surface would not be subject to quite as much potential wear. So, I decided I could slim down the outside wearing surface of the slip, and I could do so without weakening the slotted hammerhead joint running the length of the slip:

The side running surface reduces from 0.75" to 0.5".

On the front of the drawer, I have made the drawer front protrude down 1/64"  (0.015625") lower than the slip:

Why did I do this on the drawer front connection? Well, one of the things I like about metal drawer slides is that the drawer box does not rub on the surrounding cabinet members, which makes for smooth running and no wear on the cabinet framing. Inspired by that, funny enough, I thought it would be nice to have the only rubbing parts be the drawers slips and the drawer slides.

Here's a look at the drawer slides on the end wall of the cabinet, viewed from the back:

As you can see, the lower insert strip sits proud of the framing by 1/32". The drawer fronts will therefore clear the framing by 1/64" when the drawer is closed all the way. No parts to rub means the frame surfaces should stay clean and the finish unmarred for longer.

The two strips will each be fixed by 3 machine screws, not yet sketched in - and the strips will be readily removable without disassembling other parts of the cabinet. First the side insert strip comes out, then the bottom strip. Here they are removed:

Here's the scene with the drawer in place:

The corner joinery begins to get a little involved:

Those tenons will be wedged, which will contain any tendency of the short side tie piece's mitered extension to open away from the post.

The drawer runners which sit between a set of drawers are the same simple insert pieces, with a few differences in the supporting elements:

Here's a drawer runner assembly from the lower bank of drawers:

The joinery remains provisional on the drawer runner framing, but it is most of the way there I think.

Another change is that I've given the top of the cabinet's plates some extra width and formed an abbreviated crown molding:

I prefer the one-piece construction by far over the commonly-seen technique of applied molding. Note too that the upper sliding door hardware has been reduced in size by 25%.

In elevation, I think the crown gives a pleasing termination to the top of the cabinet without verging into grandiosity:
You might also note some changes to the lower pair of hinged doors and the drawer set to the right of it. Following a reader's comment in the previous post, I looked at that area again. Previously the doors had their front panels divided into thirds by battens. I made them like this because I wanted the door panel grain to run in the same direction as the drawers, and aligning the battens with the drawer banks' horizontal dividers seemed like a good move. Well, not so much. For one thing, there was no provision in that arrangement of the doors for the door pulls I had designed. So, a revamp led to a central rail in each door, and the addition of a pair of central rails in the doors which, when together, form a shape reminiscent of a belt and buckle to me. The panels can still run in the direction I want and the former 'gridded' look to the front has been dialed back, to the relief of many readers I'm sure.

On the right, the bank of drawers has been rearranged a bit. The bottom drawer presented challenges with placing the lignum vitae insert runners directly atop the sills, so I decided to drop the lower drawer altogether. It will be a hinged-door compartment instead, and will have a concealed locking mechanism for that door. Inside, there will be a storage box. Kinda like a safe I guess. I may place a drawer handle on it so that it looks like a drawer - haven't decided yet. Probably will add mitered breadboard ends to the panel as well.

I removed the back plates from the drawer hardware, and now the handles look even skinnier - too skinny. They definitely need to be thickened up a bit. I'll give some more attention to the drawer hardware in upcoming days.

Here's a look at the bottom of the cabinet in perspective:

Next, one of the doors partially opened - one can see the drawers to the right of the compartment, but that is only because the dust panels haven't been placed yet:

There will be a pair of shelves in that compartment.

Another view from a different perspective:

I pulled the interior shelving out from the last version and am reconsidering what I will do in that space. Stay tuned for more. I feel like I've made good progress since the last post and have solved some technical issues which give me the freedom to step back and focus on the aesthetic side of the puzzle for the next while.

Thanks for your visit today!  On to post 10