Another novice question: I have a large bottle of yellow Elmer’s. I built a small open front cabinet from oak and the back panel separated the entire length at one of the edge joints. I used biscuits and fastened the top and shelf with screws and plugs. I bought the glue about two years ago and have never had a problem before. Has the glue simple become too old or is there another problem?
Discussion Forum
Get It All!
UNLIMITED Membership is like taking a master class in woodworking for less than $10 a month.
Start Your Free TrialCategories
Discussion Forum
Digital Plans Library
Member exclusive! – Plans for everyone – from beginners to experts – right at your fingertips.
Highlights
-
Shape Your Skills
when you sign up for our emails
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply. -
Shop Talk Live Podcast -
Our favorite articles and videos -
E-Learning Courses from Fine Woodworking
-
-
Replies
What material did you use for the back panel?
If it was solid wood then you would have a wood movement problem. If the panel dried after the piece was assembled, the shrinkage associated with that would create a split if it was firmly fastened to the sides. (Assuming the grain on the panel was vertical) The glue joint was the weakest point--suggesting that either the glue was over the hill, or the joint hadn't been adequately tight to begin with. But if the glue joint hadn't failed, you would have had a failure elsewhere. A solid wood panel of this sort would have to be set unglued in grooves around the sides or top, or in a frame, to allow it to shrink and swell when the moisture changes, as it will.
The back panel was solid oak, just under 36" wide and 20" high.. Same age as the rest of the cabinet. I bought it and it sat around in the shop for almost two years before I used it. I bought it for a project that never got underway. I appreciate the lesson on panels being set in grooves. As I said, I'm a novice. Now, if I knew what kind of grooves and where to put them, I could avoid the same problem. At least now I know the glue needs to go in the dumper.
Edited 4/24/2006 12:48 pm ET by sansmnd
Before you blame the glue, did your design allow for expansion/contraction of the failed section? Sounds to me like that may be the true culprit. Did you glue the panel in or let it float? If you glued it in, I'm not surprised that it let go.
Mike HennessyPittsburgh, PA
We should be able to get this straightened out, though it may take a new back.
First, let me make sure I have this down correctly. You have a box that is 36" wide, 20" high, with a shelf running across the wide dimension. There is no front, but you would like to have a back.
It sounds like the corners are butted together and joined with biscuits. I assume the top and bottom panels run the full width, with the sides between them. The shelf runs parallel with the top and bottom and is attached to the sides. ???I'm not entirely clear how--screws or biscuits?
Then you made a back panel and glued it to the top and sides? Were there also screws? This is where the problem arose. Wood changes width as the humidity in the air changes with the seasons. It changes only insignificantly in length. Therefore when a long grain side is attached across grain to the back, when the back changes dimension, it is attached to something that doesn't change and something is bound to fail, if not the glue, then the wood would split.
To avoid this a solid wood back panel must be allowed to float. One way is just to cut grooves around the sides, top and bottom. The grooves would be inset a small amount, say 1/4" and be wide enough to accomodate the thickness of the back, and deep enough to accommodate the change in width that might happen. Depending on the humidity in your shop during the eight weeks it rested there, it could have been as high as 12% (and probably less), and depending upon the finish and humidity in the house where it was placed, as low as 6% (and probably more) At the worst case this would mean it was attempting to shrink slightly more than 3/8" over 20". So the grooves would have to accommodate a bit more change, so if they were 1/4" deep there would be up to 1/2" of shrinkage before the panel would slip out. You would have to disassemble the sides to cut the grooves, a PITA to be sure.
The problem with this construction, is that the panel adds nothing to strengthen the cabinet since it floats. One way around that is to build a frame that fits into a rabbet around the back. It can be glued in place since its grain runs substantially the same as the top sides and bottom of the cabinet. The groove can be cut in the sides of the frame, and the panel floats within the frame while the frame itself strengthens the cabinet. This allows you to cut down the panel, not make a new one entirely.
The third way is to use plywood for the back. Red oak plywood, 1/4" thick could be fit in a rabbet around the sides bottom and top, and could be glued in place, since plywood doesn't change dimension very much.
I hope this helps, if I have made any wrong assumptions let me know. By the way, Norm had to be "hollered at" by the viewer in his first season to get the hang of accomodating wood movement.
I think I get it now. One cannot fasten long grain to end grain? The entire piece is made of 1x. The wood is well aged as I have had it for over two years allbeit in my non-weatherproof shop. I wanted a solid oak cabinet, including the back panel. I was pleased with the result until the failure.
It is a box as you described, sorta. The back panel is 33" x 29". The sides are 14" x 29". The shelves are screwed inside the box. The top is 36" x 15", screwed to the top of the back and sides, long grain to end grain, no glue. All components are glued panels. No base - it just sits on the floor with the bottom shelf about 1 1/2" off the floor. I tried to make it as simple as possible. I had NO IDEA the oak would move as much as you say but, indeed, the crack at the failed joint is 3/8" wide. (How did you know that? LOL) I don't think I can fix this one but I want to try again.
So, I need to make a frame, cut dados all around, cut splines on the back panel and have the back panel simply sit in the frame? Or, is it possible to fasten a glued up top inside the box fastening the long grain of the back to the long grain of the top? But then, the sides would be long grain to the short grain of the top and I suppose the same problem would arise. True? But, that is how the sides now attach to the shelves and I have had no failure in the side panels. Can I anticipate the same thing happening to the sides? If not, why not? ARRRGH! My head is spinning!
I hope I have explained with sufficient clarity what I'm having a difficult time understanding. BTW, I was hoping to avoid a veneer back panel since I can never find veneer that has grain that will match or be as attractive as the solid wood. Some years ago, before I knew even the little I do now, I made an entertainment center out of ash. Essentially two speaker boxes 15x20x48 with two shelves and a large 15x48x48 box with various size shelves for components. All glued up panels fastened with screws except the back panel of the large box which is ash veneer that doesn't match the rest very well. Built in the same shop and moved into the same house. No failed joints anywhere. That's the reason I thought I could use the same technique on this project. Things that make you go hmmmm!
Sorry for the long post. Thanks again for the time all have taken on this thread.
Edited 4/25/2006 3:48 am ET by sansmnd
Sans, I didn't read this thread until just today, didn't realize it had such great info in it. Just wanted to pass along an on-line calculator called The Shrinkulator that could be very helpful to you in the future. Click here. [hopefully, I didn't miss someone else posting this!] You can stick numbers in for any given piece of wood, acclimated to your shop RH%, and see how much it will move when acclimated to a particular house/office, etc., RH%.
If you have a moisture meter you can use those numbers.
I strongly recommend you get a "weather station" for your shop to measure at least RH. I have an old one picked up at an auction, tracks temp, RH and looks cool too.
I plugged in some random numbers (Red Oak, 29" wide, Initial RH 20%, Final RH 10%, and it spit out a tangential change of -.26" -- Ouch. forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)
That's a very small change in relative humidity, which refers to moisture in the air, not to moisture in the wood.
To get a more realistic sense of how much the wood might move, plug in a change in relative humidity from 20% to 70% which is what you will see in an area with cold winters and moderately humid summers, you'll get a movement of about an inch with a moisture change in the wood of 8.6%.
I just went to the Shrinkulator site and they suggest that you should make the calculation for a change in moisture content from 4% to 14% which gives a movement of 1.11 inches for a 29 inch wide board.
John W.
Edited 4/27/2006 2:32 pm ET by JohnWW
I chose my numbers by slightly modifying one of the other posters' suggestions. None of them come even close to the swings we have in the Great Northwest, ROFL!!! Just threw them out for the OP to consider.
If I run my pellet stove 24/7, I can keep the shop around 25%-30%, but it's kinda expensive. :-(forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)
Hoadley's Understanding Wood has a table (figure 1) at the beginning of chapter five titled "Coping with Wood Movement" (at least in my 1980 edition). For us in Connecticut (I'm further south, down route 8 from Newtown.) it shows approximate MC of interior wood work in January of about 7%, and the July equivalent datum is 13%. (This corresponds to an average relative humidity of 60%) However, finish can affect this quite a bit. Table 2 shows unfinished wood fluctuating from 4% in winter to 14% in summer, but with wood finished with lacquer or varnish the range is much less, varying about a percentage point above and below 8%. But with so many people using oil finishes, and wipe on varnishes with only a few coats, that would imply a much larger range than the couple of percentage points for finished wood in the figure.
Ouch, indeed! Oh well, live and learn.
forestgirl is right. There is some great information in this thread. Thanks again to all.
Now, indulge me just a bit longer, please. It appears using a glued panel of the width I tried is not feasible because of the movement. So how wide can one be before it fails? The top and shelves of my piece are about 15" and they are fine - so far. And if wide panels are not feasible, how are solid table tops made? I am really showing my ignorance but I have so many questions after reading all the messages in this thread. I would welcome suggestions for a good rererence book on this subject. I may be new at this but I CAN read. :)
Edited 4/28/2006 3:14 am ET by sansmnd
Here is a current thread that might be of some help:
http://forums.taunton.com/n/mb/message.asp?webtag=fw-knots&msg=29628.1
forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)
Solid table tops, of any size, are attached by methods that allow them to expand and contract with the seasons.
Taunton Press--our host on these pages--has a series of The Complete Illustrated Guide to that has lots of good stuff. The volume on Furniture and Cabinet Construction by Andy Rae has answers to all your basic questions I think. Other titles in the series are also pretty self explanatory.
<<It appears using a glued panel of the width I tried is not feasible because of the movement.>>
In reality, that's not necessarily true.
<<So how wide can one be before it fails? The top and shelves of my piece are about 15" and they are fine - so far. And if wide panels are not feasible, how are solid table tops made? >>
Practically speaking, as wide as you need; however....
Very short version: wood movement has to accounted for in your design and construction. IF you do that, then wood movement, for practical purposes, is largely irrelevant.
Long version: there are numerous ways to account for wood movement. Several books and numerous articles in WW mags that touch on or cover this in depth.
Some basic things to keep in mind:
Different woods move different amounts and often at different rates (there are several sites that will calculate movement under various conditions that you can input, as well as a number of books that have movement info for specific wood species)
Wood movement is influenced by many factors:
How the wood is cut (quarter sawn, flat sawn, etc);
How the wood was dried (air, kiln, etc);
What kind and how much finish is on the wood (oil, varnish, lacquer; 1 coat, 10 coats; french polished, etc);
How the wood is restrained (draw bored M&T, dovetails, breadboard, etc);
Where you live and the relative humidity changes that occur in your area;
And a BUNCH of other stuff.
Hope that this helps a little. I'll add an post later with some specific books that have information that you may find useful.
Cheers!
James
You sound like you really want to understand the issues yourself. To do this, you'll need to get knowledge about two subject areas: Wood itself, and the construction methods chosen to accomodate its behavior.
I have gotten a great deal from the book "Understanding Wood: A Craftsman's Guide to Wood Technology" by R. Bruce Hoadley. For my money, that's the reference to how wood behaves.
There's also the excellent "Wood Handbook" produced by the Forest Products Laboratory, which is part of the USDA's Forest Service. You can afford this for sure, it's free (well, already paid for by your tax dollars). You can get it at this U.S. government web site, and it includes the following chapters:
Characteristics and Availability of Commercially Important Wood
Structure of Wood
Physical Properties and Moisture Relations of Wood
Mechanical Properties of Wood
Commercial Lumber
Lumber Stress Grades and Design Properties
Fastenings
Structural Analysis Equations
Adhesive Bonding of Wood Materials
Wood-Based Composites and Panel Products
Glued Structural Members
Drying and Control of Moisture Content and Dimensional Changes
Biodeterioration of Wood
Wood Preservation
Finishing of Wood
Use of Wood in Buildings and Bridges
Fire Safety
Round Timbers and Ties
Specialty Treatments
As you can see, it's quite a reference, 463 pages, last updated in March 1999.
On that web page, you'll also find links to vendors (like Lee Valley Tools) who will sell it to you in hardcopy, if the free 14M PDF isn't to your liking.
I've read a LOT of books on structure and design. One excellent, accessible volume is Taunton's "Practical Design Solutions and Strategies: Key Advice for Sound Construction from Fine Woodworking", ISBN 1561583448, available for about $12 from Amazon. I don't recall anything specifically about the issue you're now having, but the questions you're asking indicate you need some of the same information I've needed over the past year or so as I've tried to learn.
Hope this helps!My goal is for my work to outlast me. Expect my joinery to get simpler as time goes by.
pz, Stve and John -
Now THAT'S what I needed! I'll go into research mode now with my heartfelt thnaks.
I allow about 1/16" minimum per foot of width for expansion and contraction. For a 36" panel I would make the panel 3/16" less in width and slightly less than the exact length depending on the joinery. For a panel that large I would prefer to veneer it then I know it would be unlikely to have any problems.
Oak, of all varieties, moves quite a bit with changes in moisture. Even with a conservative estimate of just a 6% change in moisture, a panel 36 inches wide will move a full 3/4". Here in New England, with dry winter and moist summers you could expect to see a panel that wide move well over an inch.
The rule of thumb is to expect 1/4" of seasonal movement per foot of width, but for oak and a number of other species, and in situations with wide ranges in humidity the allowance needs to be even greater.
John White, Shop Manager, Fine Woodworking Magazine
I am not sure where you got the idea that a piece of wood will take on an additional 6% moisture content. If the kiln operator dries it at 6-8% MC then if it takes on an aditional 6% that is soggy wood at 12-14%. For a piece of wood to go from 8% and gain an additonal 6% it would have to an additional 75% in actual water. Oak shrinks from 6-8% from green to dry for flat sawn lumber. All the shrinkage begins at a moisture content of 30%. So if a piece of wood takes on or loses 6% moisture content it will shrink or expand 6/30 x .06 x 12 to 6/30 x .08 x 12 inches per foot. That range is .144" to .192". If wood actually expanded or contracted that much you would see wood floors which would move 18.8" to 34.56" in a floor which measures 15 feet across. I have lived in Arizona in the desert, near the water and in Houston where the temperature and humidity are about three apart. I went directly from 10 years in Houston to a few years in Arizona and did see anything of the kind of movement you suggest. If wood siding gained that much moisture think about what it would look like. It would have joints that move .25" per foot and lines all around the house unless you painted it on the day it reached maximum shrinkage. Think of what the hardwood floor in a home would look like. Even the flooring manufacturer recommends .75" at each wall. That is 1.5" across the entire floor. When I install floors I only use a 1/4" washer for a spacer for every six rows. That is the thickness of a washer for every 13.5". So I do not know where you get that kind of range of .25" per foot.
I suggest you check out the Shrinkulator at http://www.woodbin.com/calcs/shrinkulator.htm
This contains the same formulae developed at Forest Product Labs and does show that when Northern Red Oak expands in the tangential (flat sawn) mode from a moisture content of 6% to a content of 12% a one foot wide board would expand to be 12.24", close enough to 1/4" per foot in my book. Your rule of thumb isn't accurate.
The six percent range is a bit on the wide side--essentially what would happen if you took a piece of furniture from a heated room and then let it spend a summer in a house that was not air conditioned. Well made furniture should be designed to tolerate that degree of stress.
Hardwood floors work because they are NOT panels all glued together, but are made of narrow boards. Changes in moisture mostly change the space between the boards, not the width of the entire floor. But if the plumbing springs a leak at the start of a two week vacation... Similarly clapboard is nailed so it can move, and almost never has much more than 6 " of exposure. The movement potential is one part of the reason that vertical grain siding is signiificantly superior to flat sawn siding.
I suspect another problem, too. LOL Perhaps that problem can be at least mollified by reading this forum. As I've said before, I don't even know what I don't know.
Yellow glues of all brands have a self life of about one year. Using it after the expiration can be successful, but risky considering the cost of glue compared to the cost and effort of a piece of furniture. I don't know about Elmer's brand products, but Titebond puts a date code on the bottle. 6BEFJ - The 6 means 2006 and the B means Feb of that year. The other letters mean plant codes and really don't apply to us woodworkers. I use Titebond III all the time, but I usually buy just the quart size to keep it fresh. I know it is more expensive that way , but I go back to my original statement about glue cost vs project cost comparison. I ran out the other day and had to go local and was I shocked, I found glue with 4A codes to 5C on every bottle. Every one was out of date on the self of the supplier. I had to drive 15 miles to my local Lowes and I found all the bottles with 6C codes. That means they were made March 2006. I like Titebond III because it has a longer assembly time and will work down to 47 degrees. I don't know of many glues that will do that.
The second post is right on about the structure issue so I wont elaborate on that. I hope this helped anyone with glue issues.
Good tip. I knew about the 1 year shelf life, but I didn't know about the date codes on Titebond bottles. I love learning new stuff....
Thanks!
All that said, I've used "old" glue often and never had it fail. I suspect something else is the culprit with the OP's problem.
Mike HennessyPittsburgh, PA
Mike,
We all have used out of date glue successfully, but I don't even bother with it anymore because glue is so cheap compared to lumber. I agree that the problem was probably not the glue but the expansion of a solid panel not allowed to expand.
"We all have used out of date glue successfully, but I don't even bother with it anymore because glue is so cheap compared to lumber. "
Yup. I agree. I just didn't want the OP to blame the glue without also considering other causes. I'd hate to see him re-do it with good glue and get another crack due to improper design.
Mike HennessyPittsburgh, PA
That's good information on the date codes -- I was not aware, thanks.
Hello: I'm going out to the shop and dumping all old glue. You are absolutely right about the cost of glue vs. the cost of wood and the time involved in a project. Does epoxy have a similar problem? Since it isn't ready for use till it is mixed wouldn't that remove age from the equation?
Thanks, KDM
Kenneth Duke Masters
The Bill of Rights December 15 1791 NRA Endowment Member
LEAA Life Member
CRPA Member
I don't know if all epoxies have problems with age, but twice now in the Fine Woodworking shop we have had older bottles of unmixed epoxy start to foam up in the bottle until they overflowed. So obviously they were starting to change chemically.
I've also had one or both halves of unmixed epoxies start to harden in their tubes in less than a year so it doesn't seem that they are immune to aging.
John White, Shop Manager, Fine Woodworking Magazine
I used to buy yellow glue by the 5 gallon bucket and keep it up to 2 years. Sometimes the bottom of the bucket would get a little thick so I thinned it with water. Never had a problem with glue failure. I kept it sealed tight and in a heated shop. Don't let it freeze or it is ruined. Your problem might be the design of the back of your cabinet. If you use solid stock, it should be made up of shiplap or tongue and groove boards. This allows for expansion and contraction. Actually, plywood panels are more practical for backs. I only use solid wood if I am being fussy about a wood match. Then I use shiplaped boards usually 3/8 to 1/2 in thick and .nailed to the cabinet (no glue here). If you use plywood you can use glue.
One more thing about plywood. The cheap oak plywood you see in the big box stores and ordinary lumberyards is likely to have its top veneers rotary cut. This gives wild grain that looks little like solid oak. But better grade plywood with slice cut veneer that looks much more like real wood. It's available in both plane sawed, and for bigger bucks in quarter sawn.
I agree about the plywood. I try to buy from a yard that caters to cabinetmakers. Two points that I omitted from my first post is that when using boards for backs, the widths can be random in size and I put a small chamfer on the edges that upon assembly creates a "V" groove. Usually I'll resaw stock from the same lot as the cabinet is made from, run them through the planer and use them side by side. This makes for an easier grain match.
A respectable answer for Glue aging can be found at
http://www.tightbond.com in the FAQs thread. Gluing technique is a whole other story.
Work Safe, Count to 10 when your done for the day !!
Bruce S.
sansmind,
As a cabinetshop wholesale supply company here, I can't say enough good things about Franklin Titebond glue products. Leave the Elmers for the school kids, and as far as a shelf life on Titebond, it can freeze and thaw 5 times according to the factory. And, if it has sat for some time, a little aggitation, (mixing ) will bring any glue back to life. Only when it hardens in the jug or bottle should you consider tossing it. Always store it in room temp, and follow mfgrs. application instructions, and you should be in good shape for all your glue up needs. Any questions, reply..........
Kurt in MPLS
To the kiddies it goes! Maybe I can get a tax writeoff by giving it to my local kindergarten. Thanks for the advice about Titebond.
This forum post is now archived. Commenting has been disabled