In a parallel discussion about bandsaws that has drifted off-subject (pardon the pun) the comment is made that …
< Wider boards are more stable because they generally contain more rift and quartersawn wood >
… as part of an argument for using the widest possible boards when glueing up panels.
It does seem to me that all other things being equal, a panel of lots of narrow strips will be more dimensionally stable than one composed of a few wide strips, whatever the orientation of the individual boards.
I’m about to make a large half-circle table, and will glue up 5 and 6 inch boards (step by step, not all at once) to ensure stability. The table I’m typing on is an old kauri kitchen table, 42 inches wide by 72 inches long, made of 3 planks glued together, and it has cupped slightly.
Narrow-plank assemblies do need care with grain direction, it’s a mistake to mix boards that are too dissimilar, and it helps to alternate the growth rings (some up, some down), but the largest determinant of stability is likely to be plank width.
Malcolm
Edited 12/4/2005 7:34 pm ET by Malcolm
Replies
I don't think wider boards are more stable than narrow boards, but they look much better. I Also don't like the idea of alternating growth rings. I don't see where it adds to stability and it can cause appearance problems. If you have time go to http://home.woh.rr.com/federalfurniture/ for a photo and description on why I don't worry about which way the growth rings go.
Rob Millard
Excellent piece, Rob.
Let me restate the original comment, now buried in the other thread. I can't think of a single reason why a board would be more stable when it is ripped into two boards, unless perhaps if it has reaction wood that can be eliminated when the board is ripped and rejointed, but thats defective wood, not really related to width.
If you are talking about a glue up of very narrow strips, creating laminated lumber you might get less warp, but you still would not change the seasonal change in width. (this isn't the plywood case where veneer is glued cross grain to have counteracting forces that reduce total movement. )
Of course, I am biased toward wide boards because I tend to make period reproductions where the originals all used wide boards. And, frankly, it is challenging and creates a great deal of waste getting attractive grain patterns with narrow boards. Its also a challenge to make really good glue joints that don't show a glue line of any sort. I'm not saying that it is necessary to seek out the widest boards possible at all costs, but I am convinced that it isn't a good thing to rip lumber in order to glue up 4" strips which was the case that triggered my comment in the first place. If you glue the strips in the same orientation as before nothing has changed, and if you start flipping the rings you are more likely than not creating esthetic problems.
As far as practical limitations, it makes sense to keep boards within the limits of your jointer capacity. (That's why I have a 16" jointer, but also use handplanes which have virtually unlimited width capacity.) Boards much wider than 8" can be hard to find, too. Still, since I believe the downside is small, or non-existant, there is no reason not to use boards as wide as you can economically find. The upside of using wide lumber is mostly esthetic to be sure, though having a lot of quartersawn lumber in the panel will definately reduce the amount of seasonal change in width, though joinery should still account for wood movement if possible. The joinery also should tame a tendency to warp.
> I can't think of a single reason why a board would be more stable when it is ripped into two boards <
Me neither, Steve. That's not what is done tho (and not what anyone is proposing, as I understand it) - usually it's a question of mixing and matching 5 inch to 10 inch boards to get the required overall width with a good look and favourable grain direction.
> it is challenging and creates a great deal of waste getting attractive grain patterns with narrow boards <
Not sure I agree completely! I'd throw away a lot more timber if I always wanted to joint 12 inch boards! However it does take care, I agree, and sometimes some compromise!
> there is no reason not to use boards as wide as you can economically find <
That's the core of the discussion, I believe. Wide boards, varying from almost quarter-sawn at the edges to flat-sawn in the middle, are more difficult to retain flat. The table I'm typing on (3 boards spanning 42 inches, one 15 inches wide) has cupped, even tho it's probably 100 years old. The reasons are well-established and certainly supported by my own long experience. The most stable boards have uniform growth ring geometry (pure quarter-sawn being the absolute ideal), and it's much easier to achieve uniformity with narrower boards. Hence, a wide glue-up of a number of boards will - all other things being equal - stay flatter than the same glue-up made from only 3 or 4 boards.
Whether you like the look is another matter! I admit, I do like this old kitchen table, scars and scrapes and bow and all!
Malcolm http://www.macpherson.co.nz
I think you already knew the answer to your question.
Many factors determine stability: tree growth anomalies, thickness of timber, flat v. quarter sawn, age of tree. Wide v. narrow planks, it's of matter of hiding warp to a minimum, so unless the plank is quite old, I'll always re-cut to 6-8 inches to limit noticeable movement.
Edited 12/5/2005 2:37 pm ET by jackplane
I am the first to admit I am not a wood technologist. However, having spent some time getting as much out of Frank Klausz as I can, I can sum up his thoughts:1. Max width of each board to be glued up should generally be 4"
2. Don't alternate the grain. Keep the best face of each piece on the side where it shows.He has given several good technical reasons, but basically, when it comes to cabinet making advice, when Frank gives it, I follow it.
Frank Klausz is talented, I'm glad he's a mentor to you.
However, I was taught by a master to limit width to 6". This is a general rule, not strict, and depends on factors I previously mentioned. The poorer wood selection becomes, the narrower each piece becomes at glue-up.
Grain is alternated to limit warp. But it seems to me it depends on the application.Sometimes a better face is more important.
What about variable widths? I bought the remnants of air dried unsurfaced Cherry lumber from a guy ($1/bf with generous calc of bf) and I am having trouble getting good boards out of some of the pieces (not unexpected) - would it look terrible to have some boards narrower than others when doing the glue up?
I won't argue about what a proper width is, as species vary, so we can agree there. However, I will say Frank was emphatic about always using the best face on the outside surface, no matter the grain orientation. And I can only wish he was my mentor. He hs come to our club, the Long ISland Woodworkers Club on numerous occasions, as well as a visit to his shop. He always starts off with a wood technology speech, which is always interesting.
It would be interesting to see the technical reasons behind his recommendation.
For another view point, see the article by Louis Irion, of Irion Lumber, on the virtue of using wide boards, that appeared in the September 2005 issue of Woodshop News. http://www.woodshopnews.com
Here's my take on some of the issues. Not all issues are covered, but there's enough verbal diaerrheoa here to go on with. Read the notes in conjunction with the attached drawing if the link here doesn't work. Slainte.
View Image
Different workers have to adopt different strategies in panel glue ups. For instance, I can both surface plane (US face joint) and thickness plane up to 460 mm (18”) wide in one go. Therefore for an 18-19 mm (3/4”) thick table top I start with boards that are 20-23 mm (c. 7/8”) thick, join 4 to 6 boards up to a maximum of about 450 mm wide (17-3/4”) then surface plane and thickness plane the panel formed. This will usually result in a panel somewhere in the region of 19-20 mm thick (about 3/4"), and any steps formed between adjacent boards during the glue up are machined out. I then join two panels made like this to make a table top that is about 880 mm wide (34-5/8”.) Hopefully, there will only be a little hand planing with a jack plane on one joint to level it all off, and if I need a greater width than this, I do the job in three stages instead of just two. <!----><!----><!---->
For many this is not practicable because they don't have the machinery to cope with such wide boards, and the options to get a level surface after glue up are to hand plane, use a portable belt sander, find a wide panel sander, etc.. One or two of these last techniques can make it a little tricky to get a panel that is truly flat, with both faces parallel. Sometimes this matters, and sometimes it doesn't, e.g., it matters for a raised and fielded panel as opposed to a table top where this consideration, within reason, is less critical.
In conclusion, there are a variety of options that the woodworker can use for edge or butt joints, not all of which have been covered here. None of these methods should be considered the “best” or only one to use, because each technique has its place, and it’s a case of choosing the right one for the job in hand.
<!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!----><!---->End Notes for Edge Joints.<!----><!---->
Note 1. End grain orientation and movement.<!----><!---->
Sketch 8, left. Laying up tangentially, or flat sawn planks with the end grain arranged as at 1 (left) will usually result in a washboard effect as at 2 which is an exaggerated approximation. Without going into technicalities, think of the growth rings as normally wanting to straighten out during drying and shrinkage. Some movement will almost certainly happen if for no other reason than seasonal variations in ambient humidity. This applies to both air and kiln dried timber. At 3 all the pieces are arranged heart side up, and the glued up slab will cup towards the bottom as at 4. At 5 and 6 is illustrated the typical movement in quarter sawn stock where at 6 the material has merely become a little thinner because of the even shrinkage experienced. Rift sawn material, where the end grain runs approximately diagonally from corner to corner behaves somewhere between the two cuts by shrinking into a diamond shape.
These statements are purely generalisations, and it is always the exception that proves the rule. I’ve seen the opposite happen to what should often enough to be wary of making sweeping statements.
How you should arrange the planks is often a subject for heated debate. I use all these layouts, and it depends on the circumstance. For instance, the most attractive grain might all be on the heart side of tangentially cut boards, and if I’m making a table top I will use the material this way, i.e., 3 and 4. For a table top arranged this way, I usually incorporate a heavy slot screwed bearer on the underside. Similarly, if it’s a cabinet side that I need to do a dovetail housing in I might choose the same arrangement because then at worst I only have a wide bowed piece to bend straight prior to running over it with the router. This is often preferable to dealing with a series of ripples. I tend to use the layout at 1 more often when the strips are narrow, say 50 mm or less, or in cabinets where opposite cabinet sides are linked with tenoned rails.
It’s also a useful option to try and arrange all the grain so that it rises to one end as viewed from the narrow edge, although this is often not possible because the grain orientation varies so much. If successfully achieved, this results in reduced tearout when using the planers, as long as the material is fed the right way of course. This consideration is sometimes sacrificed for looks. Of course, the opposite applies where ease of work later is selected over finished appearance.
Then there is the thorny issue of the ideal plank width to use in glue ups, and again I can’t, nor won’t offer any hard and fast rules. As usual the decision has to be made according to the job. For the most part I use random widths in, for example, a table top. I will happily glue up stuff up to 150 mm (6”) wide if I am able to incorporate suitable bracing. However, if I decide that I want the boards to be around 80 mm (3-1/8”) wide each, I try to keep them all about 15 to 20 mm (5/8” to 7/8”) either side of this figure, i.e., 60-100 mm (2-3/8” to 3-7/8”.) This tactic prevents just one board standing out as markedly different from the rest, but I happen to like a little irregularity anyway. Richard Jones Furniture
Well, it's hard to tell from the photos, but if you go to Klausz's website there are quite a few pieces there that appear to be made from boards that are wider than 4". I wonder if he wasn't speaking about veneer substrate when he said use 4" boards max.?And if you want to throw around important names....how about the work of Nakashima? He certainly didn't limit himself to a 4 inch width.I've got some wide boards in my wood rack that have been there for years and they are flat. Now somebody tell why they would suddenly warp if I were to relocate them by using them in a table top or whatever.
Sort of agree.. I be makin' a project for my Sister-In-Law..My best friend... 3/8 inch thick birds Eye Maple.. Widest board is about 9 inches..Flat as can be.. Scrper a bit.. Finish with Tung oil ( I thing it has tung oil) .. Flat for a year in my shop till I put the finish on! DAMN!Experiments.. Had to cut it in 2 inch wide strips before it would stay flat as milled..Birds Eye.. Nice but like a new Bride.. Ya just Love it ....
but takes awhile 'to get it right!'
Perhaps it is my way of looking at all woodworking through the prism of period furniture, but I'm very surprised at how many knowledgeable woodworkers are advocating the use of narrow boards. Whenever possible I use full width boards for everything, and when I can't , I'm never really satisfied with the results. Years ago I bought a fairly warped slab of 5/4 figured mahogany 25"x 96" for a sideboard top. An older gentleman at the store, told be it would split and never stay flat. He fancied himself as as an expert, with his 50 years of experience, yet his furniture looked like that made by a somewhat talented high school student. Maybe, it was arrogance, but I paid no attention to him, and I'm glad I didn't. Today that sideboard top is not cracked and is still as flat as the day I took the last plane stroke down it. I still say that stability comes not from the width of the board, but the from properly dried stock, joinery details and finishing. I also, I wouldn't care if it did warp a little, furnituremaking is about using a beautiful natural material, and wood by its nature moves around.
Rob Millard
Edited 12/8/2005 9:02 am ET by RMillard
I am but a lowly amateur, Rob, but I just can't understand why a Franz Klausz would advocate use of narrow boards. If the wood is flat and dry, why in the world wouldn't everyone want the widest boards possible? Sometimes I'll cut a wide board in half if it's too cupped to begin with, but if I can get it flat and still maintain the thickness I want, I definitely want the table top to have the fewest possible boards and glue joints.
The only thing I can think of is that if you're a professional, maybe there's a practical answer having to do with the ipreferences of customers. Maybe the typical customer doesn't appreciate the beauty of the wide boards. Yet that same customer is likely to complain about any warpage. So if you're in business to make money, you arrange the boards with an eye toward minimizing warpage rather than maximizing beauty. Not that it's 100% one way and 0% the other, but you strike a different balance.
Thank goodness, I woudn't know because I'm not a professional. But that's the only thing I can think of.
Mark,
..."why in the world wouldn't everyone want the widest board possible?".
Simply because it does not always work like that-consider the following:-
I have had to use a fair amount of Iroko, always picked from prime grade parcels. A typical widish board of 450mm plus would often have say one third of this width with quarter sawn grain. So I want to make a top ora panel say about 900mm wide. Using two boards of mixed grain as mentioned would not look right, so it is better to take three of these boards, rip off the quarter sawn and use the flowery stuff to best effect. You set the quarter sawn ripped off aside for drawer fronts , for example, or save them to make a top or panel that is all quarter sawn- and here is where one can easily join it up so that it looks like one board....I seem to remember fighting this war before, but in a slightly different context.
Anyway , that is but one answer to your query, and not applicable to all timbers or furniture styles.I have other reasons, but am not willing to start a fire storm at this stage....
Philip Marcou
Edited 12/8/2005 4:31 am by philip
If I understand you correctly, you're saying that you choose the board based on its looks, rather than trying to make it wider or narrower just for the sake of its being wider or narrower. That makes sense to me. I was responding to a post suggesting that you should limit boards to 4" width without regard to the looks.
The key is "properly dried stock". To me that means between 6-8% EMC, unless one ships their work far away to a different climate.
The customer that marvels over wide, figured boards is not the same as the one who just wants all their furniture flat and without warp.
I know I said do what Frank said, but I have built one demi lune table with a 16" one piece cherry top, and since it's installed with buttons, it can expand and contract easily, and it has stayed flat for a few years now. However, I am now building a bed in curly maple, and the headboard will be captured on all 4 sides, so I will use narrower pieces built up.
Daryl,
If I'm reading your post correctly, your headboard won't be any less prone to dimensional changes, if you made it from one wide board, or many narrow boards ( unless quarter sawn stock is used) A panel made up of narrow boards will shrink the same amount as one from a full width board.
Pieces that are restrained like your headboard do present a problem, but proper joinery is the answer. I'm just now finishing up a curly maple desk, with the lid made from one full width board ( 12 1/2") The bread board ends could crack the top if expansion wasn't taken into consideration. What I did was to fit the center tenon tight and glue only it and an inch or two on each side. The outer tenons were left unglued and had the mortise elongated. I planed a slight hollow in the breadboard, so that it remained in compression and kept the joint tight ( a rigid glue is essential, so I used hide glue) Already, the breadboard stick past the field by a tiny amount, and it hasn't even been delivered.
Also, please see my little home page at http://home.woh.rr.com/federalfurniture/ for a description on how I approach selecting curly maple.
Rob Millard
I also, wouldn't care if it did warp a little, furnituremaking is about using a beautiful natural material, and wood by its nature moves around.
Amen, brother.
Edited 12/7/2005 2:36 pm ET by BossCrunk
I think ya talked to my GrandPa! He knew everything.. But I still loved him
Rob, I totally agree. I know for the sake of time and using the available materials and equipment, some people are happy with a glue-up from differing widths. And some clients don't care, like already stated. But if the furniture is built right, allowing for movement, wide boards with matching grain patterns cannot be compared to narrow glue-ups. They'll win hands down every time.
Most people don't know how to care for wide boards.
Sawing them is hard. Drying them is hard. Flatening them is hard. Keeping them from cracking is hard.
I have a shop full of boards upto 24" wide. They only get cut to finished width.
The table I am typing this on top of is 23"x 60" with a full length crotch feather.
I have to pick up a log of lumber next month - 33"dia 12' long - cut into 8/4 and dried. Wide boards are heavy.
Yeah. All of that is true!
I wish it was my problem as well!
For most of us, the days of sweating over wide boards is long gone. Which is one good reason why this debate (sorry - trying not to lecture ... in my business its a daily risk) is sort of useful.
Malcolm
http://www.macpherson.co.nz
Edited 12/8/2005 3:59 am ET by Malcolm
This forum post is now archived. Commenting has been disabled