Replies

1. 
   Sgian_Dubh | Jul 12, 2000 10:22pm | #1

   *
   Wow, that's a bunch questions to answer. I'll start by asking a few questions back. The stuff is 'planed' but how true is it? Do you have the power equipment to surface plane, edge, and thickness these boards if need be prior to glue up? Can you handle a No. 7 try plane to straighten up the edges prior to gluing up? I'm posing these questions to see what sort of equipment and experience you have. Knowing this would help provide answers no end.

   There is no reason that all your boards should be a standard width, but I'd balk at gluing up stuff that is 13" or so wide, but this would also be influenced by the cut of the boards. Are they quarter sawn, rift sawn, or plain sawn? I join wide boards, up to about 175mm (7") to make table tops and other slabs, especially if I can add suitable bracing, and this again depends on the cut, as well as the final usage of the piece.

   Alternating the growth rings is one option, but usually results in a washboard effect, which is probably not really significant in your application. I can advance good arguments for arranging all the boards with the heart side facing one way, usually up, in a table top.

   Biscuits might be a good idea, but gluing them in will most likely cause them to reveal themselves as rugby ball shaped distortions months, or even years later, particularly in horizontal surfaces where every blemish sticks out like a sore thumb. I mention this because I've seen it several times. Dowels also reveal themselves if glued in. If your joinery is good, usually, with a slight 'spring' and the edges true to the face, the need for biscuits should be minimal, but they certainly are a useful alignment aid.

   Yellow glue might cure too fast at this time of year- northern hemisphere summer, if you live in a hot humid climate. Polyurethane provides a long open time, but I am suspicious of this glues strength. I have quite often broken the offcuts at the glue line in my hand from slab glue-ups; This just makes me wary. I try not to use it for this job now. You might consider liquid hide glue, hot hide glue, urea formaldehyde, resorcinol, epoxy, or even plain old white PVA.

   I would definitely consider achieving the requisite full width in two, three, or more, managable stages. This would also allow you to start flattening between glue-ups with your machinery, if it's big enough, or by using a scrub plane, try plane, and/or Jack plane.

   This is only a partial answer, but at least I've got the ball rolling for you. Do get back with some answers to the questions I've asked. Sliante.

1. 
   Michael_R._Koch | Jul 12, 2000 11:17pm | #2

   *Thanks for the quick (and detailed) reply! You have shed some light in other areas that I hadn't considered. To answer your questions...How true are the boards? Well, I've only run them through a thickness planer. They seem to be straight, looking at them by eye anyway. I have a 15" thickness planer, an 8" jointer, but no No. 7 plane (yet). As for experience, well, this is, uh, ok it's my first project (see previous post in this forum on a Loose Tenon...which I was able to repair with a strip of veneer by the way). I don't have too much in the way of hand tools, but I do have all the power tools I could get a hold of. :)The table base is made with 4" x 4" turned cherry legs and a 2" x 3-1/4" apron. I have added 3 maple stretchers between the long rails of the table to (hopefully) keep the top from sagging. The table top boards are all plain sawn and some of the nicer grain patterns are in the center of the boards (hence my reluctance to cur the boards down). The "heart" of each board is pretty much centered on each board, i.e. growth rings run side to side on each board.I like your idea of gluing all boards with grain patterns in the same direction (if it's going to warp, warp in one direction I guess). In this manner, can the 12" boards be used without ripping them down? Should I add any additional bracing if I do this?I would never have thought that biscuits would end up leaving marks on boards over time. Maybe I should only use one row of biscuits ?!?! Do you think that the 1.5" width of the boards would prevent this problem? I could use the shaper to put a glue joint on the edges, but that would show on the end grain (maybe that's not a problem).Thanks again for all your input!!!mike

1. 
   William_Tandy_Young | Jul 13, 2000 12:52am | #3

   *Michael:I won't try to reply to all your questions, just some. Sgian has given you a lot of great info already. In general: the time and effort spent on preparing great gluing edges on lumber is worth it. A good edge joint goes together cleanly with a minimum of fuss and a minimum of extra stuff clamped all over the work to align and register the faces of the lumber. It's simple, direct, and very satisfying. For routine work, you can get the gluing edges off the jointer if you want, but read Lon Schleining's FWW article from several issues back about ripping great edge joints on the table saw. There is real merit to that method, and you can't argue with the results. For the very best work (heirloom furniture, etc.), nothing beats a hand-planed edge joint. Crisp, true, and they disappear when you clamp them up. Widths of boards: if there are going to be lots of boards in a top, I personally like to see some variety of width in the lumber. Otherwise the top looks like something done in a factory with standardized lumber. It's nice to arrange the boards to achieve some sort of visual order or symmetry. Put the widest plank in the middle and try to place narrower boards with similar widths on either side of that to balance things out visually.You can alternate the growth rings if you want, but I like to show the best face of a board if possible, regardless of whether it's the bark side or heart side. With cherry, you often have to bury the bark side down to hide the sapwood anyway.Worry about visual composition first and grain direction second. It doesn't matter if the grain of each board is rising in the same direction if you're going to belt sand the top after glue up. The most effective way I know to surface a top by hand is with a #7 or #8 plane, working across the grain, which is a lot less work, and lots of fun. Get into a happy rhythm with the tool, don't worry if the iron isn't dead sharp for this operation, and remove material faster than you thought you could with a hand plane. This (along with hand planing the gluing edges) is a great excuse to get up to speed with a big hand plane. Once you have the top flat and flush, get your final surface from a smoothing plane, scraper plane, or the belt sander, working with the grain (I've seen some critical work done very nicely with a belt sander, but have never attempted it myself).Biscuits definitely cause problems as Sgian mentioned when used with a moisture-bearing glue such as PVA (yellow) glue, but you can sidestep the problem. If you use biscuits and PVA, don't surface the top until several days after you've glued it up, to give the moisture-swelled wood around the biscuits and glued edges a chance to release the moisture added by the glue and shrink back to an equilibrium dimension. The problems arise when the wood is surfaced too soon after gluing, and continues to shrink after it has been levelled and flushed. If you have well-prepared edge joints, they'll be more than strong enough without biscuits. The only reason to use biscuits would be for alignment. Consider putting the time and effort expended in biscuiting into getting optimal edge joints, and doing without the biscuits.Polyurethane glue (Gorilla) should work fine if you want to use it. In fact, edge joints are one of the best uses for that glue, provided the gluing surfaces are fit tightly together and snugly clamped. There's no moisture in the polyurethane to worry about, so you can surface the top sooner rather than later. Maybe use two or three biscuits per edge (one at each end and the middle) just to locate the boards if you glue up with polyurethane, because it's slippery stuff, and the boards may slide around a bit as you clamp them up.Sequence: perhaps glue up the four narrower planks into two outer pairs, than glue each of these pairs to the wide center plank, one pair at a time.Good luck.

1. 
   TAB | Jul 13, 2000 01:18am | #4

   *Hi all:This might deserve it's own post, but I'm intrigued about the references to visible dimples left by biscuits. I've only come across this once, myself, and that was when the biscuit slot was cut within 1/8" (that's ~3mm to the binary impaired...) of the surface. Otherwise, I haven't noticed any glaring irregularities in pieces where I've relied on biscuits. For larger jobs like this, I have found biscuits extremely helpful in aligning the joint during glue-up, particularly in the absence of a second pair of hands. Long boards lathered up with glue seem to have a mind of their own, and alternating the batten clamps with the joint clamps under the ticking clock of yellow glue is never much fun...What wood species and slot depths have you guys noticed this with?As for Mike's original post, I wouldn't rip down the big boards if they present nice figure. Unless you rejoin them to alternate the growth rings, they're going to bow just a much after as before. In order to get a better sense of what they'll do, I'd joint and plane them to within 1/4" of thickness, sticker them up and let them sit a week or two. Rejoint and plane right before glue-up, unless it looks like some of them are cupping, bowing, twisting, etc.. excessively. In that case, I'd consider other options (i.e. more wood, ripping and rejoining, etc...). I'd keep the larger boards towards the center of the piece, where the support will be greater.Best of luck,-t

1. 
   Sgian_Dubh | Jul 13, 2000 03:06am | #5

   *A couple of interesting comments from you William. Sidestepping the problem of biscuits telegraphing through by allowing plenty of time to dry before final surface preparation I've found to be only marginally successful. It's my experience that both biscuits and dowels will reveal themselves months b and years later.Polyurethane glue. I like the stuff in many ways, but I have my reservations regarding its strength. I've glued up, oh, I've no idea how many solid timber slabs, with all sorts of glues- a job such as this being discussed here. As I briefly mentioned before, I have my doubts regarding its strength. For example, after the slab is cut to length, I'll sometimes do some idle unscientific testing. I'll take the offcuts and flex them in my hands to see if they'll break. Joints put together with polyurethane are the most likely to snap at the glue line. Just about every other glue used requires shoving the stuff in the vice, and whacking with a hammer, and more often than not, the break will occur in the grain. Similar, and comparable, results by can be achieved by thwacking the stuff over my knee.Your comments would be appreciated by me, and surely others would benefit too, for I am aware that you have some particular interest regarding adhesives/glues. Me? I'm a tad secretive, hence the handle. Sliante.

2. 
   Mark_Levin | Jul 13, 2000 03:36am | #6

   *Michael, All the previous posts offer good advise. One thing about Gorilla glue or any polyurethane-it is not good for wood with less than 8% moisture content. This is usally stated in fine print somwhere on the bottle. That said, it does make sense to use polyurethane glue for your large table because of the open time. To use poly. glue correctly for low moisture content wood you must dampen each edge before applying the glue, this is the only way it reaches maxium strength.I have glued very large tables with poly. glue with great success and have tested the end cuts as per one of the other posts and they did not break at the joint. Some of the tops were white oak which is not a good gluing wood to begin with.I use #6 biscuts(3" long) available only from Colonial Saw & Machine and have yet to see any telegraphing in thick boards. Many times I will use several slip tenons instead of the biscuits and then dowel through the top of the table to pin each end. This is overkill but it makes a nice design detail and clients are blown away. Another glue that I love for long open time is plastic resin. Very strong, easy to use and depending on temp and humidity: 8 to 12 minutes or longer open time.If you really want long open time consider Casco Resin #585 from Bordern Chemical: 20-30 minuites open time.About $100+ for five gallong pail and a very short shelf life. But it is killer adhesive.Good luck,Mark Levin

3. 
   Adrian_Wilson | Jul 13, 2000 04:02am | #7

   *I've had the same experience as you with polyurethane, and i know won't use it in critical applications. I've never seen the problems you describe wih biscuits though, but I don't typically use them in edge joining anyway.

4. 
   CStanford_ | Jul 13, 2000 04:26pm | #8

   *I'll throw my 2 cents in very quickly. This is a tough glue-up. I'd have the boards planed perfectly flat on a drum sander. You need to joint a perfect edge and rip the opposite edge as you would any project, however edge jointing these long boards will be tough, but it's doable.Forget PU glue, use a white PVA (Garrett Wade Slo-Set). Put the boards, glue, et al. in your house two days before you do the glue-up and then do the glue-up IN THE HOUSE. You may need the maximum open time the the PVA will allow. You won't get that in a hot garage glue-up with any glue.As far as orientation goes, I'd arrange the boards for the best look and then evaluate the growth rings. Most likely, you'd simply leave them as is.You don't need biscuits or dowels, what you do need to do is a dry-run to make sure you've got your clamping sequence down pat and just generally have your shi* together before you do the wet-run.You'll hate yourself if you use PU - it's too damned runny to allow rearrangement of the boards during clamp-up. The GW glue I mentioned does a fairly good job of staying put, especially if it's at room temperature. You don't want your glue to run readily off the edges of the boards, that is for sure.Keeping the top flat is not a problem if you have Bessey clamps. Simply refrain from showering down on the pressure, but I would clamp the panel on the top and bottom. You'll need seven clamps across the bottom and four to five across the top. That's about $600 worth of Besseys, but you'll be glad you have 'em.

5. 
   William_Tandy_Young | Jul 13, 2000 05:52pm | #9

   *To respond to the comments from Sgian and Tab, and to add to my own previous post a bit:Sgian's experience with biscuits and dowels telegraphing through to the surface months or years after a glue-up is well-supported by similar tales of woe from other woodworkers, unfortunately. I should let someone else comment in depth on this problem, as I simply don't have enough direct long-term experience with doweled or biscuited edge joint myself. Here's an idea someone should try: using a PVA or other moisture-bearing glue, make up a biscuited edge joint, and put the biscuits in dry. Use some slightly overthick biscuits (every box of biscuits has some of those) that have to be pushed in with a little effort. I'd be glad to hear from anyone who has some results on that.Here's another reason not to use biscuits, dowels, shaper-cut glue joints or any other added detail to edge joints on heirloom work: if your work survives for some decades, the edge joints may need a bit of repair down the road, and anything other than a plain, unadorned edge joint is a bear to work on. I spend much of my time on repairs, and I grit my teeth every time I have to reglue a doweled edge joint. I am often unable to make an optimal repair.If biscuits are used in an edge joint, I think it is indeed important how close to the show surface they are placed. The closer to the show surface they are, the more likely they are to eventually reveal themselves one way or another. For Michael's table, the top is generously thick at 1 1/2 in. A single row of biscuits located either mid-thickness or nearer to the downside face should help locate the lumber with less chance of telegraphing through to the show surface. Where boards are 1 in. or less in thickness, I'd think twice about using biscuits to edge join them. If I did use biscuits, I'd get them as close to the downside face as possible. The ironic problem is, the thinner the lumber, the harder it can be to get it aligned well during an edge joint glue-up, so from a practicality standpoint, I would understand someone being more inclined not less inclined to use biscuits for edge joining thinner lumber.Another good reason for locating biscuits near to the downside face: that "B" face should be well-aligned after the glue up, and you can use it as a reasonable reference surface if you decide to take the top to a millwork shop and have it sent through a large belt or drum sander (which is worth looking into).This raises another point for Michael: if you happen to end up with a finished width of 42 in., you can put the whole top through somebody's giant drum sander, which will do up to a 42 in. width. These ancient beasts are considered obsolete by large plants and millwork shops, so they're cheap for medium sized shops to buy, and there a lot of them out there. A well-maintained drum sander does a nice job. On to the polyurethane glue: this stuff only works well when its particular set of demands are met. For instance, like epoxy, it doesn't do well with a dead smooth, hand-planed edge joint. It requires a smooth, accurately made, but roughened surface (this is starting to sound like a James Bond cocktail recipe). That's why someone walking into my shop as I'm prepping edge joints for polyurethane gluing will see a curious sight: edge joints being lightly scuffed with 180 or 220 grit paper after having been carefully hand planed. Never thought I'd find myself doing something like that, but it's whatever works.Item two: unlike most other glues, polyurethane requires a good deal of clamping pressure. Assuming you have well-fitted joints, most other glues only require enough clamping pressure to bring the parts together. Epoxy requires even less: only minimal contact pressure. But poly requires more, and you'll lose ultimate bond strength if you don't crank down on the clamps a bit. One informal test I did bears this out. I took three pairs of 4 in. long blocks, each 2 in. square and edge glued them with poly. Pair one was rubbed together aggressively to get the parts as close as possible by hand and left unclamped. Pair two was lightly clamped with a couple of mini Quick-Grip clamps. Pair three was clamped hard with a large Wetzler clamp. After the glue cured, pair one came apart by hand. The glue had foamed up into a visibly thick layer between the blocks. Pair two came apart readily when I put one half the joint in a vise and put a handscrew on the other half and leaned on it. Pair three never did come apart. Item three: don't know if anyone else has noticed, but polyurethane glues are faster-acting now than they were when they were first introduced. Once applied, an open glue film will begin generating visible micro-bubbles sooner, and foam-out occurs sooner after assembly than it once did. This is generally a good thing, because the first poly formulas were unbearably slow-acting. But some folks may not be getting their work assembled as quickly as they should once they have spread a film of polyurethane glue. I don't have any hard data on this yet, but I think that the more an applied film of poly begins visibly reacting (forming micro-bubbles) before assembly and clamping, the greater the chance is that the cured bond line will not develop full strength. Think of the formation of the visible gas bubbles as being like a PVA or urea glue starting to skin over in the open air. Not desirable. If people are doing all the right things with polyurethane and are still finding that joints fail at the glue line, I'd like to know about it. Also, if the problems are specific to certain brands of the glue (which I'm guessing shouldn't be an issue). I've had good luck with poly thus far for certain jobs. The important thing is to ignore all the polyurethane hype, bold claims and nonsense (of which there is plenty) and find the ways in which the glue works best for you.

6. 
   rob.kelly | Jul 14, 2000 12:06am | #10

   *Michael, these guys are generally correct, especially Sgian, Adrian and CStanford who post a lot. That said, I made a cherry table 3 years ago, 30" wide, 112" long from 5 boards, 7/8" thick, and 5" to 8" wide. I glued one board at a time, Titebond II glue, about 7 biscuits per glueup, only to keep it lined up. Edges were jointed flat on a long bed jointer. For each glueup I used 1/2" pipe clamps and Pony double pipe clamps. When the clamps were tight, I used a straightedge to check flatness, and if it was bowed, I drove a tapered wedge of scrap hardwood between the board and the pipe, all the way along, as needed on each side, then released pressure a little, and then tighened it all up. I ended up with a 2 board and a 3 board glueup, and put them through the drumsander, and glued them together as previous, finishing the seam with a belt sander and a 1/2 sheet sander. Finish was Watco natural oil and Watco wax. No problems so far. It's held to the base with wood buttons.

7. 
   Sgian_Dubh | Jul 14, 2000 01:34am | #11

   *William, to quote you,i "I simply don't have enough direct long-term experience with doweled or biscuited edge joints."I noticed this phenomenom(sp?) many years ago when my employment involved more restoration and repair work than I do now, in particular at that time with dowelled table tops and the like that came in for repair. I still see it when such tops come in, and more recently I've sometimes observed the same characteristic in table tops biscuited together.I eventually realised that a common factor is that glue was applied to the dowels, and more recently, biscuits were getting the same treatment. My solution for preventing this when using either dowels or biscuits is to use them solely as an alignment aid, not as a means of holding the edges together. Prepare edge joints as for a normal 'sprung' joint. Cut the slots, or bore the holes, but don't apply glue to the hole, dowel or the biscuit at assembly. If a little glue dribbles into the hole, or onto the biscuit, I don't worry about it. Also use fat biscuits where possible as you mentioned. I can't say that this approach is right. It's just my attempt to find a solution, but this seems to work with my pieces,.......so far.Polyurethane has a tendency to push the joint apart due to the foaming action, and my experience concurs with yours in that firm pressure is best. Similarly, I too have made the 'perfect' edge sprung joint with the overhand surface planer (jointer) followed by the try or jack plane if need be, and done a bit of roughing up with abrasive paper on a block. A spritz of water if it's a dry day, which is seldom down here on the Gulf Coast, and I've never worked with timber- in this part of the world, that reads less than 10% MC in the workshop. It may have been kilned lower than that at some point, but it gains moisture around here until it's put inside a home, office, etc.. Those offcuts continue to break on the glue line. Granted, it's not a fair scientific test, but it occurs. The brand was Excel. I don't use polyurethane glues for this job anymore. I use it for plenty of others, but not this one. And tomorrow, I have a bunch of panels to do just like the topic here, except they will be vertical when installed. The old pot of hide glue is coming out for the job, and no biscuits. All the PVA's are trouble for this, and other traditional joinery, at this time of year, in this location. I've appreciated your contributions to this topic. I hope we haven't gone off on too much of a tangent for Michael Koch. Sliante.

8. 
   Adrian_Wilson | Jul 14, 2000 04:56pm | #12

   *FWIW, Excel was the brand I've had trouble with also (only one on the shelves here). I admit, when edgegluing, I didn't abrade the surface like William suggested, but I did use lots of pressure, and had the same questionable results in situations other than edgegluing. I just decided to stick with tried and true adhesives, but I'll probably give it another try sometime.

9. 
   Sergio_Piery_II | Sep 22, 2000 05:57am | #13

   *If alignment is an issue, though it shouldn't be, you can try the use of a spline made of the same wood with the grain running in the same direction. If you do not want the splines to show, stop the 'dados' short of the ends. I prefer this method to biscuits.Good luck, Sergio.