CA glue fail

[Kevin Jenness]

I've never been completely sold on CA glue, though I have used quite a bit over the years. Sticking bits back on that split off, minor divot fills, quick small repairs are common. When I first used the stuff for glue blocks on the lathe and had the workpiece break off cleanly after a catch, I shied away from it for serious jobs. I know it has a reputation for brittleness, but a lot of RF airplane builders swear by it and luminaries like David Ellsworth recommend it for attaching weak wood to glue bocks screwed to faceplates. I have used it that way myself that way without incident in the recent past, but I didn't get any catches either. The lack of shear strength makes it easy to pop the workpiece off the glueblock with a chisel in the glueline but it worked to keep the work attached to the lathe spindle.

My current problem has to do with holding a piece in a vise for carving. I am working on a series of multi-axis hollow forms that require considerable work off the lathe. I use screwed-on faceplates for the shaping and hollowing, then saw off the base to eliminate the screw holes and attach a square block that can be held by a chuck in a carving vise. I could stick with the original faceplate attachment but I want to reduce the base thickness to avoid uneven drying and checking. The first two pieces held on ok even with a fair bit of mallet and chisel work, but the third popped off the glue block twice, even hit the floor (no damage!). The glue was thick Starbond, fairly recent vintage and kept in an airtight container with dessicant, used with accelerator and held in place with hand pressure. The glue block was 2 1/2" square, the workpiece about 10" high, the surfaces flat and sanded, the glueblock was dry and the work partially air dried, the failures occurred within an hour of attachment. I was removing material with a gouge and mallet with what I consider moderate force. Both joints broke clean off the hard maple workpiece with very little residue on that face and no visible penetration.

I am going to let the work dry for a bit and re-attach a glue block with yellow glue to complete the project but I my skepticism as to CA glue is renewed. Comments/suggestions?

 

[Russell Nugent]

Is it starboard accelerator? I'm guessing you put the accelerator on the maple, that's why there was no penetration into the maple. Could you skip the accelerator and clamp/weigh it down a bit? You probably already have the real solution for the issue, use pva glue instead of CA if you're gonna be wracking at it.

 

[Kevin Jenness]

Is it starboard accelerator? I'm guessing you put the accelerator on the maple, that's why there was no penetration into the maple. Could you skip the accelerator and clamp/weigh it down a bit? You probably already have the real solution for the issue, use pva glue instead of CA if you're gonna be wracking at it.

Both the glueblock and the work were maple. The accelerator was 2P10, which has worked fine in the past with various brands of CA glue, though I suppose there may be some advantage in using matched proprietary products. Glue was applied to the dry glue block and accelerator to the work. I can see where omitting accelerator would promote penetration into the workpiece. Do you think CA glue is just not suited to "wracking at it"?

 

[Russell Nugent]

I don't like to use CA glue personally when there's going to be any sort of shearing force on it.

 

[Kevin Jenness]

Fair enough. I'm ok with cutting off blocks glued with a serious enough bond to get the work accomplished. I was wondering if I am leaving out a step in using CA glue. Is the glueline thickness important?

 

[Mark Jundanian]

Based on the one pic, my thought was that your glued block looks small for the size of the workpiece.

 

[Vincent Vogel]

When I use ca glue for glueblocks I don't use accelerator. I put glue on one side and push the two pieces together and rotate them to evenly spread the glue and then clamp it for about 1/2 hour minimum, sometimes overnight. Using this procedure I haven't had any failures. I had failure with accelerator on one side and glue on the other side.

 

[Kevin Jenness]

Based on the one pic, my thought was that your glued block looks small for the size of the workpiece.

True- it is sized so that the chuck jaws allow working all the way around the base of the piece. I would like the glue bond to be as strong as the chuck's connection to the glueblock. I could wait to cut down the base and keep the work on the original faceplate, but I have had some checks in other pieces that I attributed to drying stresses caused by uneven wall thickness. Normally I can get a simple piece finished quickly enough to avoid these problems, but working with wet material over several days is presenting new challenges. Maybe I should keep it in a water bath until the rough shaping is done.

 

[Kevin Jenness]

When I use ca glue for glueblocks I don't use accelerator. I put glue on one side and push the two pieces together and rotate them to evenly spread the glue and then clamp it for about 1/2 hour minimum, sometimes overnight. Using this procedure I haven't had any failures. I had failure with accelerator on one side and glue on the other side.

Thanks, I will try that with a test piece. Also will experiment with medium viscosity glue to see if that penetrates better- a well-mated joint shouldn't need thick glue. I guess it should not be a surprise that a strong, shock-resistant, instant, easily released glue bond is a stretch. At least the piece didn't break when it hit the floor.

 

[John Snow]

I've had a bit of trouble with accelerator seeming to result in weaker bonds.

 

[Bill Blasic]

I would try the PVA glue and paper bond, have had good results with that.

 

[Lyle Jamieson]

I've never been completely sold on CA glue, though I have used quite a bit over the years. Sticking bits back on that split off, minor divot fills, quick small repairs are common. When I first used the stuff for glue blocks on the lathe and had the workpiece break off cleanly after a catch, I shied away from it for serious jobs. I know it has a reputation for brittleness, but a lot of RF airplane builders swear by it and luminaries like David Ellsworth recommend it for attaching weak wood to glue bocks screwed to faceplates. I have used it that way myself that way without incident in the recent past, but I didn't get any catches either. The lack of shear strength makes it easy to pop the workpiece off the glueblock with a chisel in the glueline but it worked to keep the work attached to the lathe spindle.

My current problem has to do with holding a piece in a vise for carving. I am working on a series of multi-axis hollow forms that require considerable work off the lathe. I use screwed-on faceplates for the shaping and hollowing, then saw off the base to eliminate the screw holes and attach a square block that can be held by a chuck in a carving vise. I could stick with the original faceplate attachment but I want to reduce the base thickness to avoid uneven drying and checking. The first two pieces held on ok even with a fair bit of mallet and chisel work, but the third popped off the glue block twice, even hit the floor (no damage!). The glue was thick Starbond, fairly recent vintage and kept in an airtight container with dessicant, used with accelerator and held in place with hand pressure. The glue block was 2 1/2" square, the workpiece about 10" high, the surfaces flat and sanded, the glueblock was dry and the work partially air dried, the failures occurred within an hour of attachment. I was removing material with a gouge and mallet with what I consider moderate force. Both joints broke clean off the hard maple workpiece with very little residue on that face and no visible penetration.

I am going to let the work dry for a bit and re-attach a glue block with yellow glue to complete the project but I my skepticism as to CA glue is renewed. Comments/suggestions?View attachment 66174

Hi Kevin. I've never had a problem with using a glue block and CA glue for turning bowls. One thing that stood out was your surface preparation needs to be concave not flat.
Hope these videos might help....

View: https://www.youtube.com/watch?v=y8b35iq4LTA

View: https://www.youtube.com/watch?v=ZBMj1egdwCM&t=31s

 

[Kevin Jenness]

Lyle, my understanding from your videos was that the concavity was to provide a definite contact at the outside edge of the glue block so there was no chance of it wobbling on a central hump. Why would two flat surfaces joined with CA not adhere well?

 

[John Ames]

Kevin - have you considered using hot glue? If you cover the entire mating piece it should hold quite well. People get in trouble with hot glue when they only use stripes or ribbons of glue. If the whole block is covered (like peanut butter on bread) it will hold quite well even with shearing force. If there’s room for a bead around the mounting block - as if you were caulking it - even better.

You’ll need denatured alcohol to remove it…

 

[Lyle Jamieson]

Lyle, my understanding from your videos was that the concavity was to provide a definite contact at the outside edge of the glue block so there was no chance of it wobbling on a central hump. Why would two flat surfaces joined with CA not adhere well?

I dont trust flat on flat. I am not an engineer so I cant make it perfectly flat and it wont stay flat, especially if using sanding disk or belts and heating up the surface. Too many variables, wet, dry, end grain, side grain, fine or porous grain. Use some judgement for tall pieces or bigger turning and use a faceplate. Dont pound on glueblock, it is brittle, but it will not fail in turning. Catches are completely preventable, get some help if you are getting catches. But even a catch will not break a properly used glueblock.

 

[Lyle Jamieson]

Yes, Hot glue would work better for carving but will cause/allow vibration when turning.

 

[Kevin Jenness]

Kevin - have you considered using hot glue? If you cover the entire mating piece it should hold quite well. People get in trouble with hot glue when they only use stripes or ribbons of glue. If the whole block is covered (like peanut butter on bread) it will hold quite well even with shearing force. If there’s room for a bead around the mounting block - as if you were caulking it - even better.

You’ll need denatured alcohol to remove it…

I don't trust hot glue on wet wood. I did a test with the offcut base of my piece (25% moisture content) and a dry piece of oak with medium viscosity CA glue, no accelerator, and clamped it for an hour. I did get absorption into the end grain piece and it took quite a bit of force to split the joint, though it did part with no wood failure. Rather than wait for the wood to dry enough to use pva glue I opted to reglue the block with CA and a bead of hot melt around the outside for insurance. So far so good.

 

[hockenbery]

Lyle, my understanding from your videos was that the concavity was to provide a definite contact at the outside edge of the glue block so there was no chance of it wobbling on a central hump. Why would two flat surfaces joined with CA not adhere well?

I’ve use Lyle’s method for many years. For turning NE bowls.
A thick bead of glue on one piece, accelerator on the other, align quickly, and a key for me is a slight twist.
The twist spreads the glue in a wide band inward because of the two slightly concave surfaces.

The CA does an excellent bond withe wet wood - but the working time is limited to when the wood moves.
A flat chisel on the joint breaks the glue. CA is one of the few glues not stronger than wood.

Had one student who didn’t twist and probably didn’t press the two parts together.
Joint failed. Inspection showed the bead of glue had dried as a bead and the contact area was the thin top of the bead.

 

[Leo Van Der Loo]

Yes a slight concave surface on the glue block and no hammering on the piece makes for a safe and strong bond, accelerator does not help, just give it some time for the CA to harden up.

 

[Roger Wiegand]

Yes, Hot glue would work better for carving but will cause/allow vibration when turning.

Why would that be? Perhaps if you tried to use it right away, just when it jelled but before the water evaporates out of it there would be some flexibility, but once dry hot glue is more rigid than PVA or epoxy, hence its use to hold things like guitars that need to vibrate as one piece together. With a thin glue line I've never really noticed any movement even shortly after putting pieces together. I put the glue on one surface, rotate them against each other to ensure a tight fit than let it sit for 15-20 minutes before applying significant stress to the joint. For small items like sequential segmented rings I only wait until the glue has "grabbed" before turning.

Fish glue is a more convenient liquid alternative that grabs very quickly; I've never tried using it on a glue block though. I find it to be as fast and a lot more reliable than CA for sticking small bits together if the glue pot is turned off.

 

[Kevin Jenness]

I’ve use Lyle’s method for many years. For turning NE bowls.
A thick bead of glue on one piece, accelerator on the other, align quickly, and a key for me is a slight twist.
The twist spreads the glue in a wide band inward because of the two slightly concave surfaces.

The CA does an excellent bond withe wet wood - but the working time is limited to when the wood moves.
A flat chisel on the joint breaks the glue. CA is one of the few glues not stronger than wood.

Had one student who didn’t twist and probably didn’t press the two parts together.
Joint failed. Inspection showed the bead of glue had dried as a bead and the contact area was the thin top of the bead.

I too have turned numerous pieces on glue blocks with thick CA glue, accelerator and a slightly concave joint surface. The first time I did it I had a catch and broke the joint, which put me off the idea. Since I stopped getting catches I haven't had that problem. I think my recent failure on the carving stand was due to too thick of a glueline, due to using thick CA on a flat surface coupled with accumulated shear stress from mallet work with a relatively brittle glue. I can see the potential for warping of a dry glue block joined to wet wood so that argues for getting the work done quickly.

With medium CA glue, no accelerator and considerable pressure I got a thin glueline, penetration into both joint faces and seemingly a stronger joint. Next time I use thick CA and accelerator I will look to see if there is any penetration on the accelerator side.

I opted for CA for this task as it does adhere to wet wood. Is that true of fish glue?

 

[Mark Jundanian]

CA glue, no accelerator and considerable pressure I got a thin glueline, penetration into both joint faces and seemingly a stronger joint.

Since CA bonds in seconds, I never really understood the "why" of accelerator.

Never heard of Fish glue?

 

[Kent Jaffrey]

Why would that be? Perhaps if you tried to use it right away, just when it jelled but before the water evaporates out of it there would be some flexibility, but once dry hot glue is more rigid than PVA or epoxy, hence its use to hold things like guitars that need to vibrate as one piece together. With a thin glue line I've never really noticed any movement even shortly after putting pieces together. I put the glue on one surface, rotate them against each other to ensure a tight fit than let it sit for 15-20 minutes before applying significant stress to the joint. For small items like sequential segmented rings I only wait until the glue has "grabbed" before turning.

Fish glue is a more convenient liquid alternative that grabs very quickly; I've never tried using it on a glue block though. I find it to be as fast and a lot more reliable than CA for sticking small bits together if the glue pot is turned off.

I think he was referring to hot melt glue, from a glue gun, rather than hot hide glue

 

[hockenbery]

Never hear of fish glue.


I don’t do much carving but have used vacuum to hold platters and spheres for carving. You might get a good enough seal to hold.

Also if you seal the wet wood with CA double sided tape might hold it if you can get enough P to make PSA

 

[Kevin Jenness]

Never hear of fish glue.


I don’t do much carving but have used vacuum to hold platters and spheres for carving. You might get a good enough seal to hold.

Also if you seal the wet wood with CA double sided tape might hold it if you can get enough P to make PSA

I meant to ask Roger about the fish glue- I doubt any of the animal glues will work on wet wood but maybe he has had success.

There is not enough surface area available on this piece for effective vacuum clamping. Interesting idea about sealing and using tape, thanks. I wonder if that would be any better than a straight CA joint. I assume you by " get enough P to make PSA" you mean applying enough pressure to get a good bond? I used a Kelton drive mandrel to press the glue block on at the last go-round.

I got the major shaping done last night so the serious loads on the glueline are done.

 

[hockenbery]

joint. I assume you by " get enough P to make PSA" you mean applying enough pressure to get a good bond?

You got it the P is for presure
Looks like you have a masterpiece in progress!

Another thing I’ve used is strapping tape with the fibers in it to hold thing that want to come apart.
This may be useful
One trick as you would know is to not cut through it while you need it.

 

[Kevin Jenness]

One trick as you would know is to not cut through it while you need it.

I would but I am afraid of courting cracks by having the base too thick during the time I am working on the piece. Same with getting the bulk of the material off from the seams between the conjoined forms. I have been misting the surface and keeping the piece in a plastic bag, then in a closed cardboard box, between sessions. The plastic bag is safer but invites discoloration which is no good once I get into the final shaping. I am not used to extended projects like this with wet wood and am feeling my way along. The material has been in log form for several years but is still above the fiber saturation point and wants to check and crack if given the opportunity.

 

[Gerald Lawrence]

Some good advise but I am wondering as you mention chisel and mallet. I would think that due to the brittle nature of CA , a mallet could break the bond . This seems to be a failure in sheer strength so an alternate adhesive would be advisable. I prefer Tite Bond for problem joints.

 

[Kevin Jenness]

Some good advise but I am wondering as you mention chisel and mallet. I would think that due to the brittle nature of CA , a mallet could break the bond . This seems to be a failure in sheer strength so an alternate adhesive would be advisable. I prefer Tite Bond for problem joints.

Gerald, I am working with material at > 25% moisture content, otherwise I would use pva glue. Do you know of any more shock-resistant adhesive that will work with wet wood? Franklin recommends using Titebond at < 16% m.c.

edit: Perhaps polyurethane glue will work. I will experiment with Gorilla Glue and PL Premium construction adhesive.

 

[Roger Wiegand]

Sorry, a hot melt glue gun is something I associate with Martha Stewart type craft projects, not holding wood together, though I do have one and occasionally use it for temporary positioning. Because of the nature of what I spend most of my time doing I have a hot hide glue pot on my bench and use it extensively for many different things. It's the glue of choice for my piano and organ work.

Fish glue is related to liquid hide glue and I get it from my organ leather supply house. StewMac sells it, as does Lee Valley. Unlike liquid hide glue it grabs in just a minute or so, you can hold pieces together long enough to glue them without clamping--great for tasks like putting caps on organ pipes where you need precise positioning and might want to get them off again some day. It's perfect for leather to wood or felt to wood. I think it is not as strong as hide glue, I've never really used it on structural joints. Think of it as old-fashioned CA, I guess.

I haven't had need to glue really green wood, so have no experience with that. Water would certainly slow/inhibit the second phase of hot hide glue curing, which requires evaporation of water.

 

[Kevin Jenness]

Sorry, a hot melt glue gun is something I associate with Martha Stewart type craft projects, not holding wood together, though I do have one and occasionally use it for temporary positioning. Because of the nature of what I spend most of my time doing I have a hot hide glue pot on my bench and use it extensively for many different things. It's the glue of choice for my piano and organ work.

Fish glue is related to liquid hide glue and I get it from my organ leather supply house. StewMac sells it, as does Lee Valley. Unlike liquid hide glue it grabs in just a minute or so, you can hold pieces together long enough to glue them without clamping--great for tasks like putting caps on organ pipes where you need precise positioning and might want to get them off again some day. It's perfect for leather to wood or felt to wood. I think it is not as strong as hide glue, I've never really used it on structural joints. Think of it as old-fashioned CA, I guess.

I haven't had need to glue really green wood, so have no experience with that. Water would certainly slow/inhibit the second phase of hot hide glue curing, which requires evaporation of water.

I am not a big user of hot-melt, but it does have its uses, Martha Stewart aside. I used it to hold a 4" emerging bowl on a faceplate once with no problem. I only added a bead around the glue block in this case as a low-strength insurance policy in case the CA bond failed again.

I would be all over fish glue but I seriously doubt it would cure on my material. Instant grab and easy disassembly would be nice but what I am primarily looking for is a strong, shock-resistant bond on wet wood.

 

[John Ames]

I will experiment with Gorilla Glue and PL Premium construction adhesive.

I'll save you some time ... construction adhesive is a poor choice. it stays soft for a very long time and doesn't have a lot of strength for resisting torque forces. gorilla glue is just a foamy mess that I try to avoid but it might work in your application.

there is one last suggestion I can make here... If you REALLY want to hold two things together instantly and if the mating surfaces are relatively flat to one another, double sided butyl tape will hold like grim death. its used to seal roofs in the RV industry and in the boating industry. it will stick to anything, anywhere, anytime including under water. and it does not want to come off. even with acetone or DNA... You CAN get it apart, but it is a mess.

 

[Mark Jundanian]

Is "Alien Tape" related to butyl tape? Has anyone tried Alien Tape, in any application? OK, I admit I watch to much late night TV on the "minor" stations, but I'm curious.

 

[John Ames]

Is "Alien Tape" related to butyl tape? Has anyone tried Alien Tape, in any application? OK, I admit I watch to much late night TV on the "minor" stations, but I'm curious.

no... butyl tape is nothing like "alien tape".

 

[Gerald Lawrence]

Is "Alien Tape" related to butyl tape? Has anyone tried Alien Tape, in any application? OK, I admit I watch to much late night TV on the "minor" stations, but I'm curious.

I got an off brand on Amazon and it works well. Have not tried on wood.

@Kevin Jenness As a last option Just try another wet piece with TB III. Allow 24 hours to cure and then hit it with a hammer to test bond.

 

[Darryn Achall]

3M 5100 Marine sealant is a moisture cured compound. Must be wet to cure. It is also removable from wet wood. 3M 5200 Marine sealant is a permanent bond and you would need to cut the block to remove it. You could experiment with these products on wet scraps. I have used both extensively on my 34' all wooden cabin cruiser. I believe it takes a long time to cure though.

 

[Kevin Jenness]

3M 5100 Marine sealant is a moisture cured compound. Must be wet to cure. It is also removable from wet wood. 3M 5200 Marine sealant is a permanent bond and you would need to cut the block to remove it. You could experiment with these products on wet scraps. I have used both extensively on my 34' all wooden cabin cruiser.

Those came to mind as well as some Sikaflex products. The 5200 is tenacious as hell, also Sikaflex. Years ago I was working in a house in Fat City (Stowe) as the carpenters were pulling up a hand-planed floor, glued with Sikaflex, that didn't pass muster. They were ripping the planks about every inch and prying them up with chisels and plenty of cussing. I will keep them as a last resort given their expense and short shelf life once opened. Thanks.

 

[Jim Reynolds]

Potentially not relevant for green wood, but Fine Woodworking did an article in 2007 about glue strength including destructive testing. Here is a link rehosted on the Old Brown Glue website. I found the conclusions in the article surprising, as did the authors.

 

[Kevin Jenness]

I have settled on Gorilla (polyurethane) glue for this application. Glue block cut from the end of the turning, end grain to end grain so low potential for dimensional conflict , two flat surfaces prepped with 80# on the edge sander, glue spread on both surfaces and clamped for a thin glueline. No failures over 4 days with light mallet work, a 2 3/4" square block released cleanly with a sharp chisel driven parallel to the glueline. It took considerably more force to shear the bond than with CA glue.