Replies

1. 
   VtwoodchIP | Jan 19, 2007 06:34am | #1

   I have used both plastic resin and West Systems Epoxy for laminates.  I have used the plastic resin primarily for straight laminates.  I used the West systems for creating replacement stems for wooden canoes using ash and oak.  Cut to about an eigth of an inch, steamed and bent over a form to dry, then reassembled with the epoxy and shaped to fit.  After shaping, it is obviously a laminate but works fine.  It can be messy. 

   have fun.

1. 
   mrbird90 | Jan 19, 2007 06:40am | #2

   Will the epoxy reduce the spring back or will I have the same problem?

1. 
   wiskytango | Jan 19, 2007 06:58am | #3

   Plastic resin glue is the best choice. I think that if you use 4 1/8 inch strips you will have less spring back. I could be wrong but i thinkg that the more thinner plies the less spring back you will get.

2. 
   VtwoodchIP | Jan 19, 2007 08:00am | #6

   I did not have a problem with the final laminate holding the curve - it was steamed/boiled first and allowed to dry for several days before the epoxy was applied.

1. 
   user-265628 | Jan 19, 2007 10:09am | #7

   "1/8" Italian Bending Poplar" as it's called sometimes is ideal for laminatingUnibond glue is the absolute best laminate glue, no spring back. It's sold here:http://www.vacupress.com/veneerglue.htmor just call them at: 1-800-382-4109

1. 
   VtwoodchIP | Jan 21, 2007 05:07am | #14

   ok, I will check it out.

2. 
   DoubleK | Jan 21, 2007 08:00pm | #15

   I know I'm joining the discussion late but a few important things to consider. The tip about the resourcinol leaving a purple glue line in your finish was right on, the unibond tends to create a solid brown line, I tend to use West System because its glue line seems to be more compatible with solvent based finishes. The west may bleed through thinner layers which scares people but it does sand out ( I use finer grits) and again is compatible to solvent based finishes. If you are using water based finish then you need to choose a glue line closer to the color of your finished project. And I think rarely do any of these glues stain, which is why I said that about the west.

2. 
   hammer1 | Jan 19, 2007 07:34am | #4

   I think you are using the wrong material with 1/4" Baltic birch. The strength in the multiple layers will work against you. There are a number of products that would make the job easier. To reduce springback and use your existing form, you could use 1/8" bending plywood. By running the inner layers cross grain, with staggered joints, and using a cold press glue like Unibond, the joints will help reduce the pull. Make sure you leave it in the form long enough for the glue to completely cure. Glues like Tightbond can have a tendency to creep under stress. You may want to avoid water based glues like plastic resin since they can cause some warping. If you use a vacuum bag, moisture can take a while to dissipate.

   1/8" bending plywood is normally poplar. I think there are some Italian plywoods in other species. It would be easy enough to veneer the poplar if you needed birch on the face/faces.

   Beat it to fit / Paint it to match

3. 
   oldusty | Jan 19, 2007 07:43am | #5

       Try sanding or planning a ply off of one side of each piece before gluing to your form . The material will naturally start to bend , orient all the pieces the same way with the bend facing  the correct position , in my experience doing this you can make a tighter and smoother radius and finished product with less spring back .

                      good  luck            dusty

4. 
   Jfrostjr | Jan 19, 2007 05:16pm | #8

   I suspect the problem is not with the glue but the number of laminations.

   This was news to me until I went to a bending seminar by Michael Fortune. The "nuetral axis" is neither in compression opr tension. (Your glue line.) To hold the curve you need something on the outside and inside to hold the cureve. I made two 60" diam. arches with 5 laminations. I glued the first 3 around my form (with Gorilla Glue - what a MESS!) and got 6" of spring-back. Then I added the outer two lams, bent the whole thing around my form agaain and got NO springback. Try thinner lams and more of them.

   Frosty

5. 
   tom77 | Jan 19, 2007 08:26pm | #9

   Having read all of the replies, I would add the following.  Springback of glued laminations is inversely proportional to the square of the number of laminations.  So using a higher number of thinner laminates will reduce springback.  If you double the number of laminates, springback is reduced by 1/4.  (you can get the same reduction in springback by just doubling the number of laminates, but you get a doubling in thickness)  I know this sounds odd, but there is a theoretical basis for this relationship.

   The choice of glue doesn't affect the initial springback.  But, PVA glues do creep somewhat under stress, and are not recommended for bent laminations that must hold their shape.  I have use polyurethane glue, epoxy, and urea formaldehyde all with success.

   Good luck, Tom.

   
   

   ---

   Edited 1/19/2007 6:31 pm ET by Tom77

1. 
   Rich14 | Jan 20, 2007 05:41am | #13

   "Springback of glued laminations is inversely proportional to the square of the number of laminations."Omigosh! complex mathematical formulas(with exponentials!!!!!!!)have crept into art.". . . inversely proportional to the square of the number of laminations . . ."I just bend it 'till it looks kinda like I want it.". . . inversely proportional to the square of the number of laminations!!!!!!!!!!!!!!!. . ."ARRRRRRGGGGHHHH!!!!!!!!!!

6. 
   Quickstep | Jan 19, 2007 11:29pm | #10

   I recently did a lamination of 11 pieces of 3/16 poplar using resorcinal and had virtually no springback. I got the glue from Jamestown Distributors. (http://www.jamestowndistributors.com) I've also used epoxy and gorilla glue on laminations with good success. I agree that two pieces of 1/4" is probably a bit too thick. Also, I read somewhere that you should always laminate odd numbers of plies. Something about having a solid mass at the centerpoint to eliminate a place for a crack to start. I don't know how true that is, but it worked for my lamination, so I went with it.  

7. 
   Curtis | Jan 20, 2007 04:02am | #11

   There has been a lot of good ideas here.

   Bendable plywood is great stuff.

   but there is a company that a cohort of mine highly recommends. I've never tried them but there web site looks out of this world.

   http://www.flutedbeams.com/home.html

    

   good luck on your project.

    

   C.A.G.
1. 
   jeanlou | Jan 20, 2007 04:30am | #12

   Curtis,

   I see that in one of the posts you were given a source for resorcinol glue. It is a glue that will withstand steam heating, making it good for heat bending. It is a two part glue that requires mixing and sets rigid. It is often used in marine applications. The only problem with it is that it leaves a purple glue line. If you want to work cold, you can also use urea formaldehyde glue, which also sets up rigid and is waterproof in cold or room temperature conditions. It leaves a less obvious glue line. I tend to stay away from epoxy glues for this type of glue up, though epoxy definitely has its place in a woodshop for repairs and for gluing unlike materials together...good luck.   JL

8. 
   DonStephan | Jan 22, 2007 03:12am | #16

   Some glues dry "rigid" meaning they will not flex when cured, and some have some flexibility after full curing. Let a thin layer of PVA dry on a piece of 6 mil plastic overnight - the next day the cured glue can be flexed somewhat. Let a similar thin layer of epoxy or urea formaldehyde cure overnight and it will not flex at all without breaking.

   Perhaps I've been lucky, but with thin laminations and epoxy or urea formaldehyde I've not experienced any springback. After clamping, leave a small amount of excess glue on a piece of plastic sheet alongside the clamped stock so you know when the glue is fully cured. Both epoxies and urea formaldehydes cure faster in higher temperature.

1. 
   user-150529 | Jan 22, 2007 04:51am | #17

   Hi All,

   I used to work for a door manf. my job was to figure out a better way to do"ït". One of the products was a royal door, an elliptical stained glass insert in the large solid upper panel. It usually came in pairs and had a sidelite with a smaller width insert on either side of the entrance. We would bent lam the half moulding, finger join and shape on the overhead router to hold in the glass. We would use the same method for round top door casing, stop inside moulding and brick mould. We once did a half ellipse for a sun rise above three pair of french doors, thats over eighteen feet across. We used a pva glue acid catylized each day of use. the form was interior and pressed with the banding machine we borrowed from shipping finished with wedges drove home, at this point the glue would bead out. Solid wood off cuts from the operation so 1/8 by four quarter, edge grain on the broad surface, sanded to finished size. Anyway long story short the b-lam would would always shrink or close not spring.

   Grey Glove

9. 
   Boliou_25 | Jan 23, 2007 08:30am | #18

   One other issue, Urea formaldehyde glue is classified as a "probable" human carcinogen - you need a special formaldehyde cartridge - (not at your local store) for your respirator, a regular organic vapor cartridge wont filter it effectively. It is a great glue for bending, but no good for your health. SO, I'd make sure that you really can't live with a little long term creep before you mix some up.