Announcement

Collapse
No announcement yet.

Gluing up a panel - demonstration of failure!

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Gluing up a panel - demonstration of failure!

    This is an interesting blog series that I thought you guys might find informative. Stemming from a post on Reddit where one user was wondering what happened to his glued-up panel, this guy decided to do a test on simple home center lumber to see what would happen.

    Part I

    Part II (7 Days Later)

    Part III (Conclusion - 3 weeks on)

    So, that's a hell of a demonstration! It definitely shows you how the simple act of adding glue to a construction can send it in completely the wrong direction... frankly I was quite surprised, but then I haven't been at this all that long

    It begs the question for me - should you always have a batten of some kind? And if you're not doing a breadboard end, what are the options? I'm looking to do a fairly contemporary coffee table and I definitely don't want the walnut to cup and warp! I thought a local restaurant had a good idea when I went to their opening... the tables were made of solid wood, but it's cross-laminated. Two layers aligned with each other, and a perpendicular layer in between. I was there for dinner on Saturday and guess what... they're all cracking, some along the glue joints and some in the wood. There goes that idea!
  • Thread Continues Below...

  • #2

    Re: Gluing up a panel - demonstration of failure!

    Re: Gluing up a panel - demonstration of failure!

    If you buy properly kiln dried wood, acclimatize it to your workshop, properly straighten it out, and then laminate it - your panels will stay flat. I do it almost every day, year after year.

    Problems occur if your wood is not dried enough, or you bring it into a heated shop and work on it too quickly - or do not straighten it properly and clamp it with stresses introduced. Also, if you have it too close to a localized heat source (like under a heat vent, etc).

    You never, ever, ever glue solid wood cross grain. That is one of the fundamental things to learn in woodworking solid wood. It never turns out well.

    I never use battens - I think they are unnecessary if you follow good work practice as I suggested. They are bulky, unattractive and introduce their own issues.

    AJC
    Andrew J. Coholic
    Joe Coholic Custom Furniture Ltd.

    www.joecoholiccustomfurniture.com

    Comment


    • #3

      Re: Gluing up a panel - demonstration of failure!

      Re: Gluing up a panel - demonstration of failure!

      Originally posted by Ajcoholic View Post
      You never, ever, ever glue solid wood cross grain. That is one of the fundamental things to learn in woodworking solid wood. It never turns out well.
      So I take it you're not a great believer in the likes lap or M/T joinery?
      In order to understand recursion, one must first understand recursion

      Comment


      • #4

        Re: Gluing up a panel - demonstration of failure!

        Re: Gluing up a panel - demonstration of failure!

        Originally posted by darius View Post
        So I take it you're not a great believer in the likes lap or M/T joinery?
        You have got to be kidding right?
        "Do it Right!"

        Comment


        • #5

          Re: Gluing up a panel - demonstration of failure!

          Re: Gluing up a panel - demonstration of failure!

          Originally posted by Rusty View Post
          You have got to be kidding right?
          Not at all.
          Why would you think so?
          In order to understand recursion, one must first understand recursion

          Comment


          • #6

            Re: Gluing up a panel - demonstration of failure!

            Re: Gluing up a panel - demonstration of failure!

            Well to start with you're nit picking if you're serious. The two joints you mention involve 2 pieces of wood each. The thread is about trying to hold several pieces of wood in a flat alignment with a cleat glued across grain.

            If you're still not kidding then perhaps you could explain how the panel construction as described by guitarchitect could be successfull. If on the other hand you can't save the panel then you will understand why I thought you were kidding.
            "Do it Right!"

            Comment

            • Thread Continues Below...

            • #7

              Re: Gluing up a panel - demonstration of failure!

              Re: Gluing up a panel - demonstration of failure!

              Originally posted by Ajcoholic View Post
              If you buy properly kiln dried wood, acclimatize it to your workshop, properly straighten it out, and then laminate it - your panels will stay flat. I do it almost every day, year after year.
              well here's something that I have always wondered (and keeps me gun-shy about gluing up a panel... even though I shouldn't be).

              I've got a shop that is un-insulated. Currently my wood is standing up, on top of a garbage bag on the concrete floor to mostly prevent moisture from seeping in (planning to build a rack soon). It's very dry here in the winter, so the wood must be fairly dry by now. Presumably it's going to be in a more humid environment for the rest of its functional life, indoors... so should I be keeping the wood inside and then be bringing it out there as needed to work? Or am I over-thinking the issue? I definitely don't want to be saddled with storing wood indoors in my tiny 1B apartment, but I also don't want my first coffee table (and dining table soon after) to crack/split/bend!
              Last edited by guitarchitect; 03-26-2013, 10:52 PM. Reason: to add quote

              Comment


              • #8

                Re: Gluing up a panel - demonstration of failure!

                Re: Gluing up a panel - demonstration of failure!

                Originally posted by Rusty View Post
                Well to start with you're nit picking if you're serious. The two joints you mention involve 2 pieces of wood each. The thread is about trying to hold several pieces of wood in a flat alignment with a cleat glued across grain.

                If you're still not kidding then perhaps you could explain how the panel construction as described by guitarchitect could be successfull. If on the other hand you can't save the panel then you will understand why I thought you were kidding.
                I am responding to a hyper-emphasis which eliminates any exceptions, which is the meaning of "never, ever, ever". Heck, even the Old Testament and the Soviet penal law are not as categorical - they settle for just "never". These sort of statements are dangerous as it takes just one single exception to prove you wrong. So it's more about language than woodworking. You know, the language you think I find difficult to understand.

                Still, there are more exceptions to the "rule" and many of us use the exception. It all depends on an individual circumstance such as the moisture variations, dimensions of the cross glue-ups, lumber species, the adhesive materials used and the details of the actual work that differentiate the scenarios.

                As for the panel construction this is my take:
                1. pine of that relatively skinny (for the wood species) thickness will frequently pose a risk of cupping
                2. the boards, from what I could see in some photos, in this glue-up are not alternated so the cupping of the overall piece was almost guaranteed.
                3. in this glue-up pine should have received a finish on both sides, as well as on the end-grain as soon as the glue-up was dry and sanded.
                4. to keep the top flat I'd use blind sliding dovetail, which would still not guarantee the elimination of the cupping within the individual boards.


                In short, pine isn't the greatest material for larger flatwork.
                In order to understand recursion, one must first understand recursion

                Comment


                • #9

                  Re: Gluing up a panel - demonstration of failure!

                  Re: Gluing up a panel - demonstration of failure!

                  Originally posted by darius View Post
                  1. pine of that relatively skinny (for the wood species) thickness will frequently pose a risk of cupping
                  2. the boards, from what I could see in some photos, in this glue-up are not alternated so the cupping of the overall piece was almost guaranteed.
                  3. in this glue-up pine should have received a finish on both sides, as well as on the end-grain as soon as the glue-up was dry and sanded.
                  4. to keep the top flat I'd use blind sliding dovetail, which would still not guarantee the elimination of the cupping within the individual boards.
                  just to clarify... the intent of these blog posts was to expose the fact that simply gluing cross-grain on a panel is enough to cause the whole thing to screw up. in the original reddit thread there were 30 theories as to why the guy's table split like hell, and the author felt that the gluing of the batten was the cause. this was his proof, and he was replicating the original panels which didn't have alternated cupping.

                  the original poster on reddit used a blind sliding dovetail, but he glued it in place... so when the top shrank, it split and pulled apart rather than moved in place(and hence the severe cup in the woodworking blog, who replicated it with a glued+screwed batten to demonstrate the principle)

                  Comment


                  • #10

                    Re: Gluing up a panel - demonstration of failure!

                    Re: Gluing up a panel - demonstration of failure!

                    Originally posted by guitarchitect View Post
                    just to clarify... the intent of these blog posts was to expose the fact that simply gluing cross-grain on a panel is enough to cause the whole thing to screw up. in the original reddit thread there were 30 theories as to why the guy's table split like hell, and the author felt that the gluing of the batten was the cause. this was his proof, and he was replicating the original panels which didn't have alternated cupping.

                    the original poster on reddit used a blind sliding dovetail, but he glued it in place... so when the top shrank, it split and pulled apart rather than moved in place(and hence the severe cup in the woodworking blog, who replicated it with a glued+screwed batten to demonstrate the principle)
                    I am pretty sure the cupping you illustrated would have occurred without the cross grain glue-up.
                    It is true though that glue-ups under many scenarios will cause splitting or cupping, but usually the cupping will demonstrate itself as bucking upwards mostly in the center of the top. Your test seems to be showing moisture variations between the top and the underside as a major, though not necessarily sole, reason for the screwup.

                    I also stick to my comment about some exceptions where cross-grain glue-up will work and hold well, and not just in the m/t and lap joinery.

                    Below is an example of such an exceptions where various details of the construction make the cross glue-up work perfectly well.
                    It's a little stool that is 4+ years old now. The cross grain glue-up is 18" wide and there is no splitting or cracking of any kind, even though the stoool has spent its entire life over a heating vent - possibly the last place you want to put cross gain glue-ups. I just checked the moisture content, which as of now is below 6% - the minimum my moisture meter registers and the decimal point blinks if the MC value is less then the minimum. It was blinking. Summers here are humid so I have rather wild MC variations throughout each year. This glue-up should have failed a while ago.

                    Obviously, this piece entails some details of construction, and differences from your experiment, which defeat the no cross-grain rule:

                    Click image for larger version

Name:	IMG_1034.jpg
Views:	1
Size:	27.3 KB
ID:	886817
                    Last edited by darius; 03-26-2013, 11:17 PM.
                    In order to understand recursion, one must first understand recursion

                    Comment


                    • #11

                      Re: Gluing up a panel - demonstration of failure!

                      Re: Gluing up a panel - demonstration of failure!

                      to clarify - not my experiment! i just thought it was interesting. i agree the cupping would have happened no matter what... he was using the construction-grade big box store lumber to illustrate a point - screwed battens with room to move keep panels flat, gluing the battens across the grain don't. it's not to say it's the best solution or the only solution, it's just an illustration of the principal and a good visual indicator of what "allowing the wood to move" accomplishes

                      Comment

                      • Thread Continues Below...

                      • #12

                        Re: Gluing up a panel - demonstration of failure!

                        Re: Gluing up a panel - demonstration of failure!

                        I'm not disagreeing with the general principle at all and I apply it whenever I have too. My point was that the principle can be overcome under some situations.

                        I do have to admit though that looking at your experiment it looks like it was designed to show a failure - the choice of lumber, the grain orientation, the storage and the lack of any finish.

                        I have yet another example to show if I can get hold of a piece I made 3 years ago. It was a small sewing table, all made of HD grade construction lumber. I have no idea how it's holding up since it was a present for a friend's wife. In it I think I used cross grain glue-up for the top. I'd be curious myself to see how it looks after 3 years. I'll try to get some pix i the next couple of days.
                        In order to understand recursion, one must first understand recursion

                        Comment


                        • #13

                          Re: Gluing up a panel - demonstration of failure!

                          Re: Gluing up a panel - demonstration of failure!

                          Originally posted by darius View Post
                          I do have to admit though that looking at your experiment it looks like it was designed to show a failure - the choice of lumber, the grain orientation, the storage and the lack of any finish.
                          Um... reading the guys posts would have told you that from the start

                          Originally posted by toolerable
                          So, after putting in my two cents, I decided I would put my money where my mouth is, and attempt to recreate the failure
                          ...
                          My prediction is that over the course of the next few weeks, the panel that I glued the battens on will split, crack, and generally fail.

                          Comment


                          • #14

                            Re: Gluing up a panel - demonstration of failure!

                            Re: Gluing up a panel - demonstration of failure!

                            Originally posted by guitarchitect View Post
                            Um... reading the guys posts would have told you that from the start
                            Hence the adage "a picture is worth a thousand words". I have to admit I didn't read the many words in that story. The few pictures were enough to see how the experiment was bound to show a failure.
                            In order to understand recursion, one must first understand recursion

                            Comment


                            • #15

                              Re: Gluing up a panel - demonstration of failure!

                              Re: Gluing up a panel - demonstration of failure!

                              Originally posted by guitarchitect View Post
                              well here's something that I have always wondered (and keeps me gun-shy about gluing up a panel... even though I shouldn't be).

                              I've got a shop that is un-insulated. Currently my wood is standing up, on top of a garbage bag on the concrete floor to mostly prevent moisture from seeping in (planning to build a rack soon). It's very dry here in the winter, so the wood must be fairly dry by now. Presumably it's going to be in a more humid environment for the rest of its functional life, indoors... so should I be keeping the wood inside and then be bringing it out there as needed to work? Or am I over-thinking the issue? I definitely don't want to be saddled with storing wood indoors in my tiny 1B apartment, but I also don't want my first coffee table (and dining table soon after) to crack/split/bend!
                              My shop is unheated also. I get my wood ready for glue up and bring it inside to sit for a week at least before gluing. I have only ever had one problem and that was this winter. A table top of less than 3/4 inch thickness bowed really badly after gluing along one of join lines. I simply cut it apart and substituted another board. The subsequent process went fine. No reason not to do woodworking in the winter.

                              Comment

                              Working...
                              X