What is CA glue and how does it work?

[kurtvd19]

Just shear alone will not be an issue with models.  What will break a joint is a sharp rap to the piece - deck house on deck bumped from the side.

[Balclutha75]

Xtreme newbie modeler here, with a question for you guys. I just used some CA glue on my model for the first time. It's the DAP RapidFuse, one of the thicker varieties, with a 30 second set time.

 

Based on my observations, it seems CA glue will bond, at least to some degree, on painted, stained, or sealed wood. But PVA glue will not bond in these situations. Is that accurate?

 

Apologize if this is covered above, but I didn't see it.

[bartley]

2 hours ago, kurtvd19 said:

Just shear alone will not be an issue with models.  What will break a joint is a sharp rap to the piece - deck house on deck bumped from the side.

Yes, Kurt this "sharp rap" is somehow different.

 

Many years ago my son was sailing in a small boat constructed as a "klegicell sandwich" .  This is foam core with kevlar fiber on each side.  It was like rigid like a sheet of steel with respect to movement against the water but once coming around a mark, the boom of an another out of control sailor hit the side of his boat.  Punched a neat hole straight through.  So, very strong in one sense but very poor impact strength.

 

John

Edited February 10, 2021 by bartley

[Clark Griswold]

1 hour ago, Keithbrad80 said:

Just the opposite actually but your on the right track! Ca glue requires a porous surface to bond to and PVA can be stick to pretty much anything if you let it dry long enough. 
 

Bradley

I think you are confused which one is which.  Build a plastic model with PVA and you have a pile of plastic parts if you look at it wrong.  Nothing likes bonding non-porous materials, CA just hates it significantly less than PVA... particularly when you don't sand the joint to provide it something to grab onto mechanically.

[Keithbrad80]

49 minutes ago, Clark Griswold said:

I think you are confused which one is which.  Build a plastic model with PVA and you have a pile of plastic parts if you look at it wrong.  Nothing likes bonding non-porous materials, CA just hates it significantly less than PVA... particularly when you don't sand the joint to provide it something to grab onto mechanically.

Good point, I had only put wooden models into the equation, I could imagine wood glue would not do great on plastic without sanding or something. 
 

Bradley

Edited February 10, 2021 by Keithbrad80

[Balclutha75]

9 hours ago, Clark Griswold said:

Nothing likes bonding non-porous materials, CA just hates it significantly less than PVA...

Thanks guys, this is what I learned by trial and error. I have to glue some small parts to an already stained surface but will sand the stained piece beforehand.

[Gregory]

12 hours ago, Balclutha75 said:

Apologize if this is covered above, but I didn't see it.

I don't think this was discussed in detail beyond pointing out it works better on  porous material.

 

I have had mixed results on painted and varnished surfaces.  I make it a point to rough up such surfaces, and usually have good results.

 

 

[Keithbrad80]

3 hours ago, Balclutha75 said:

Thanks guys, this is what I learned by trial and error. I have to glue some small parts to an already stained surface but will sand the stained piece beforehand.

I apologize if my previous comment was misleading, I had a specific environment in mind with your question and failed to consider other alternatives. I hope this didn’t confuse you much but your original thought was correct. Hopefully today I plan to post a long write up a quite a few tests I have conducted with CA and how it reacts with certain stuff hopefully you find it useful and it answers some questions you might have! 
 

Bradley

[Balclutha75]

2 hours ago, Keithbrad80 said:

your original thought was correct

I'll admit to being confused for a moment, but then got straightened out. I confuse easy in any case.

 

Thanks for posting this information. Some of it has gone over my head, but it's pretty interesting. Seems like CA is a good tool to have around and use in the proper situation.

[Keithbrad80]

I wanted to present some findings from a few tests I did. A few things about these tests before I get started; the woods I tested were basswood, cherry, mahogany, beech and walnut. The vast majority of these tests were done with basswood, due to its common appearance within kits. Also, I didn’t want to waste a bunch of hardwood for this. All chemicals were mixed or introduced outside, and I wore an N95 mask, a face shield, rubber gloves and a not cotton shirt while conducting all of these tests. If you do not know the chemistry do not try some of these tests at home. Like my dad use to say: don’t be dumb! 

 

Once again my intention with this thread is to help educate new modelers or those who haven’t used CA before. 

 

The tests I conducted obviously were all centered around the glue, but obviously the thing we care about the most is the wood and how it behaves. A lot of these tests you will likely never encounter, but I wanted to see if unconventional methods would work. First I wanted to see if you could remove enough CA from wood to make it look like it wasn’t used, the second goal was to see if something could be applied under CA glue to prevent it from penetrating to deep in the wood, and the third set of tests were to see what I could use to hid CA glue. All basswood strips were cut to 2 inches, and a line of thin CA was applied to either side of the strip of wood after clamping to the board I used. Any other specifications will be given during the direct test. 

Finally on to the tests!

Test #1
Usually the first thing a modeler might try when they get something on wood is to sand it, so can CA be sanded? The simple answer is kind of. When you put CA on wood, you may not notice but there is some on the surface so its no longer flat. Sanding works great to remove large amounts of excess CA glue, but it wont remove the CA from the pores of the wood, it just removes the shiny look the CA leaves behind. 

 

I sanded with 180, 220, 300, 400 and 600 grit. 180 removed all excess CA glue, from there it just polished the wood. 

 

 

Test #2
This second test was water, can you wash it away? I’m sure we are all familiar with the difference between end grain and edge grain, so I wont discuss that here, but the idea is, if CA glue is applied to the edge grain of a piece of wood, you could swell the wood grain enough to sand it away. 

 

This didn’t really work, honestly it just added another step to the process. I saw pretty much the exact same results with just sanding. 

 

 

Test #3
Some glues will soften with heat, so the next logical step is to try heat right? I anticipated that this wouldn’t work but I tried it anyway, honestly I was pretty curious about this one. CA glue is flammable, with a flash point of 75 C, so I had to be pretty careful with this one, I did this one out side with my hose in hand. I used my heat gun to heat the glue at the joint to 70 C while  constantly monitoring the temperature, I didnt leave the glue at this temp for more than a few seconds, and it quickly cooled down. 

 

The heat actually thinned the glue making the problem even worse! 

 

 

Test #4
If I was desperate and didn’t know anything about wood or glue I might try goof off next, who knows maybe it will work. Goof off kind of surprised me. Along the line that I placed the glue it didn’t have any effect, although goof off did a really good job at removing the bleeding of CA glue outside of the line. In some spots it removed all of the bleeding glue.  
 

 

Test #5
I have seen lots of people suggest the use of Isopropyl alcohol, AKA rubbing alcohol, (referred to as IA from now on) to remove already cured CA glue, and I think I have even use this method before. My results weren’t great though, I only had 70% IA which is probably the reason my test didn’t give the results I expected, I’m not sure you can buy 100% unless your a professional but places like The Home Depot, any grocery store and I think hair supply stores sell

up to 91%, maybe even more.

 

I did two tests with IA, the first was rubbing the IA into the joint with a Q-tip to see if I could soften the glue, I replaced the Q-tip every 10 seconds and rubbed the joint for two minutes without separation. The next test I did was a two inch joint held together with CA that was submerged for 2 hours, again no separation. I submerged the joint only halfway, because I was also curious what would happen to the wood around that joint that you didn’t put IA on. My test didnt show very good results, but thats ok!

 

 

Test #6 
We established much earlier that the product of the bond made by CA resembles plastic based on its composition and characteristics, and so if you really want to remove something held together by CA use acetone. Acetone is a powerful solvent that is used mainly in the plastic industry, but it is also powerful enough to completely debond CA glue, which can be great sometimes. 

 

I conducted the same test with acetone as I did with the IA, with two minutes of rubbing the acetone in I had some separation, after 2 hours of submersion it had completely separated and most of the CA glue stain was also gone. I suppose this method works well for a single piece, If I built something and didnt like the CA stain I could submerge it in acetone to remove it, however I would not use it on something I already glued to the model. Acetone works very quickly so be careful not to spill any on anything you dont want to remove it also evaporates very quickly at room temp so don’t worry if you pour a little in a cup and its gone in an hour. Last note about acetone; its very flammable and volatile, store this in a cool dry place that is not exposed to direct sunlight, wear a face mask as well, the odor is really strong. 

 

 

Test #7
At this point I decided to test some stuff that I may have not considered an option before, like WD-40. WD-40 is a great multipurpose metal cleaner and lubricant. Interestingly WD-40 itself actually cant lubricate anything, it works by displacing the water around it, so it pushes it out of the way resulting in a

slippery feel. 

 

I sprayed WD-40 right on the joint and noticed it did not penetrate the wood but instead sat on top of it, it did a really good at camouflage because it stained all the wood around it. I dont recommend WD-40 for CA glue removal. 

 

 

Test #8 
Hair spray was next, and it had similar results of that as WD-40. It actually did remove some of the glue, it didnt penetrate the wood very deeply and the didnt break the joint, but did remove some of the CA glue stain around the joint. I also dont recommend hair spray, as the stain it created was way worse than the CA stain.

 

I did a few more tests afterwards with the following products that pretty much all had the same result. I tried: Makeup remover wipes, gasoline, liquid chlorine meant for a pool, hand lotion, air brush cleaner, and baby wipes. It’s not worth posting pictures of these because they all look the same as WD-40 and hair spray. I don’t recommend any of them.

 

 

Test #9
So from here I wanted test what would happen if you put CA on a surface that wasn’t bare wood, the first test I tried was water. I really wasn’t sure what to expect because water is critical in the curing process of CA glue, personally I thought the CA glue would cure even faster, which isn’t exactly what happened.

 

I did three tests with this, the first was putting CA on a wet piece of wood, the CA glue just sat right on top. I suppose nothing, not even water can inhabit those pores for CA to properly penetrate the wood. The second was if you soaked the wood then let it dry would that have any effect? Surprisingly, kind of. With a dry piece of wood that was wet, the CA penetrated fine but left a really hard and shiny shell over the wood when applied, different than regular CA stain. I have no idea why this happened and I wouldn’t be surprised if my test was the reason for this. The third was what if you just wiped it really quick instead of soaked it? It quickly thinned out to cover the area of the water, I will compare it to a primer of sorts, there was a short pause before it thinned out (a fraction of second) but then covered pretty well. It did not stop or avoid any CA glue stain. 

 

Test #10
Have you ever gone to stain something and gotten splotches or streaks? In regular wood working it’s likely you didn’t apply an even coat or waited to long to wipe it off or for a bunch of reasons, this can an incredibly frustrating result because once the wood stain is on there, there isn’t much you can do. Preventative measures do exist though, a prestain wood conditioner does a great job at evening out the stain. I wanted to test what the results would be if I applied a prestain wood conditioner and put CA glue on top of that.

I conducted two tests with this one, CA glue on wet conditioner and CA on dry conditioner. The CA didnt penetrate at all with wet prestain conditioner. It did penetrate the wood with dry prestain conditioner and with a little sanding actually looked pretty good, there was still some CA stain but it didn’t look to bad. 

 

 

 

Test #11
What if you stained your wood before gluing to the model, would that help at all? The simple answer: no.

 

Wood stain is oil based, so its very thin and as soon as you make contact with the wood it spreads very thin and really penetrates the wood. CA acts in a similar way but if the oil is there, the CA will not want to penetrate that wood, instead it just sits right on top. Once again I applied CA to wet and dry stain and both showed the same result, 0 penetration of the wood and just a spot of CA sitting on the wood. Awhile ago I mentioned that when CA comes in contact with air a “rind” will form on the outside of the drop. I got a picture of it that I hope better illustrates that idea. 
 

 

Test #12
Prestain wood conditioner was allowed to dry for this one, then a wood stain was applied on top of that and CA glue was applied to a wet sample and a dry sample with the same results as the previous test. No penetration. 

 

Test #13
I tried IA as well, who knows! It didn’t stop the CA from penetrating the wood but when CA was applied to wet IA that had not dried it never cured, I wonder if the alcohol changed the glue at all? Interesting, but I don’t think I would try this on my model. 

 

 

Test #14 
I wanted to see what happened when you put CA on top of PVA wood glue. With dry PVA the CA “ate” its way through the wood glue until it reached the wood, I dont think there was any penetration of the CA in the wood below the PVA. When applied on top of wet PVA it created this weird texture but nothing really happened, they didnt mix or react. The CA just kind of disappeared, I’m honestly not sure. 

 

I tested a few other things that were not that interesting, and gave results similar to wood stain. I tried vinegar, cooking oil, a salt solution I made, and pool water. 

 

 

Test #15
Ok, so we used CA glue, we tried to remove what we could and that didn’t work, we tried to better prep the wood and that didnt work. Looks like CA is here to stay, so how do we hide it? That was my next question during these tests. 

 

These next few tests, are me trying to figure out which type of paint or finish will cover CA glue well. First up, an acrylic based primer, it worked great! I used a brush to brush it on, I didn’t care about the neatness of the paint, just how well it stuck. I have used an acrylic based primer with an air brush on surface with CA stain my self with fantastic results. If painting your model after using CA I highly recommended some sort of acrylic based primer. 

 

 

Test #16
Acrylic based paint was the next obvious test, it also worked just as well, I will say that both primer and paint did take a few extra minutes to dry in spots where they sat above CA glue stain. This really was not that big of a deal though, because eventually it evens out and dries nicely. The two working together create an excellent finish that hides glue stain all together. 

 

 

Test #17
Oil based paint was next. I dont use oil based paints very often, spray paint is a pretty good example of an oil based paint, but you can buy them in all sorts of configurations. The oil based paint soaked into the wood very quickly and dried within minutes, except for the spots with CA. The oil based painted almost highlighted the CA in a way because the paint on bare wood dried in a lighter color. I would not recommend oil based paint on CA glue stain. 

 

 

 

Test #18
I have almost never used enamel based paint, the first model I built was a plastic model just a few years ago and used enamel to paint it, thats the extent of my enamel knowledge. I will say it didn’t care for CA, like oil based paint it did great on the wood but dried a different color over the CA, it also took almost an hour for this to dry. 

 

 

Test #19
This next test is my go to if I have bare CA glue stain that I want to hide. The use of a dull coat or sand and seal, or a wipe on poly like I used for this test, some how completely hides CA glue. I’m not sure what the chemistry is there but I know it doesn’t remove any CA glue.

 

For this test (and all other painted tests) I put a big drop of CA on a basswood board then applied the finish on top. For the wipe on poly test, I put a large amount of CA, enough to really soak it, I then gave 15 minutes to fully dry. One fully dried I outlined the stain in black pen and applied the wipe on poly on top. I didnt do any sanding of the CA or poly, I only applied one coat and this is the result. While the wipe on poly took longer to dry, it still does an excellent job of hiding CA glue. I highly recommend this technique. 

 

 

Test #20 
Now I wanted to test the physical capabilities of CA and the environments they could handle. First up is the cold. I had two pieces of wood clamped together had joined with CA, like every other test, and put it in the freezer. The temp before was about 25 C (room temp) and I cooled the joint until it reached 0 C, the glue didn’t debond but became very weak, I was able to pull the two piece apart with a little effort. This is something to consider if you live

in an area that gets very cold, if its freezing every night, your joints could potentially become really weak. 

 

Test #21
I read that mixing CA with certain things like sodium bicarbonate (baking soda) creates incredibly strong bonds. Real quick note: CA glue is exothermic as we have already discussed, meaning it releases heat during the curing process. Baking soda drastically increases the amount of heat CA puts off, and creates a toxic vapor so don’t do this if you are not familiar with these properties. I only mixed a small amount outside while wearing the proper PPE with my hose in hand next to the pool, so I felt safe conducting this experiment. 

 

What’s interesting is I thought if I mix a 50/50 ratio of CA to baking soda it should create a soupy mix, which is not happened. Instead the baking soda absorbed a huge amount fo the CA very quickly formed a very hard and very light disk of CA. Along the long axis (Ill call this the x-axis) the CA was incredibly strong, I hit it two or three times with my 28 ounce hammer, pretty hard. Along the short axis (y-axis) it was very brittle, one hit and it shattered. This test showed me the physical impact CA could take.

 

I did a few more physical tests like this that were already described or were not interesting enough to mention here. 

 

Test #22
This is where I tested how CA would look and react to other wood species. For all of these next few tests, I use a two inch strip of basswood glued to a board of the wood in question and a line of thin CA was applied to both sides, and allowed to fully cure. First up was cherry, my personal favorite. I have yet to build a model out of cherry but have worked with it in other senses.

 

It did ok with the CA glue, unlike basswood the CA took a longer time to dry and presented an even worse CA stain. I had to clamp these pieces together to get them to hold. 

 

 

Test #23
I then tested Mahogany wood, it absorbed the CA better than the cherry but still didn’t do as well as the basswood. I didn’t have to clamp the wood this time and it took hold after a second or two, the stain on this one was the worst I think, it just spread everywhere once it touched the wood. 

 

 

Test #24
Next up was Beech wood, it actually did really nice with the CA. It held immediately, didn’t need clamping and the CA didn’t spread very far. 

 

 

Test #25
The last hardwood I tested was walnut. Walnut didn’t really take the CA at all, I had to work the CA in with the other piece of wood, and clamp it for awhile before I noticed enough strength in the joint to let go. A little push with my finger though knocked it right off.  I dont recommend using CA on walnut. 

 

 

Test #26
This next one is something you do not want to happen. You dont want to get CA on clothing, specifically cotton. Clothing has a very high surface area because of the individual fibers, this allows for CA to become very thin and cure extremely rapidly. As I mentioned a moment ago, CA is exothermic, so when the surface area of CA is increased, and the rate at which it cures is increased, the heat output of the reaction is also increased. This means when you get CA on clothes it gets very hot very quickly, and because its CA it’s likely to stick to your skin as well. This obviously screams danger so try not to get it on your clothes, in fact a recent experiment conducted by scientists found even putting a Q-tip inside a bottle of CA can cause it to catch fire. So always take the proper precautions. 

 

This ends my discussion on CA glue, I’m sure there are several things that I missed, but then again I’m not a chemist, so most of this information was research done on my part. If any one has any questions please feel free to ask, I’m sure there are a lot of other inputs and experiences with this topic, personally I don’t advocate using CA for everything. I think a harmony between PVA glues, CA glues, and epoxy are what is going to give you the strongest model, but hopefully with the information I have provided you can make the determination on your own of when and where to use this glue. Thank you again for taking the time to read this very long post, happy modeling every one!

 

Bradley  

 

Edited February 11, 2021 by Keithbrad80

[Canute]

Well organized tests, Bradley. Great food for thought.

[Gregory]

Thanks for the hard work.  very interesting..

 

It looks like you used some very thin CA.  I mostly use some stuff called medium gap filling from BobSmith industries.

I mentioned in another discussion how I threw away a lot of stuff before I found the stuff I use.

I was trying to fill out an order at ModelExpo to get free shipping, and I ordered their house label CA.

 

Turns out it was re-labeled BobSmith Industries, and I haven't used anything else sense with any satisfaction.

I'm sure there is other good stuff out there, but I have no further need to experiment.

 

It doesn't spread out like your last examples on wood.

 

The only time it leaks out from under the edge is when I use too much.  It took me a long time to get used to how little it takes

I have used it on pear, cherry, boxwood, walnut, mahogany and others.

 

When I get visible residue, I use acetone on q-tip, stroking it several times,  and it virtually disappears, even after it cures.

Edited February 11, 2021 by Gregory

[Keithbrad80]

5 hours ago, Gregory said:

and I haven't used anything else sense with any satisfaction.

Same. I buy my CA from model expo as well, and have tried a few different types but have never been happy with any thing else. I use super thin CA a lot for a lot of different things, it’s a really popular finish for pen turning. Applied in several thin layers CA create a very shiny and hard shell which looks pretty good on pens. Your also totally right about thicker glues, they won’t bleed nearly as much, I just didn’t have any before hand and didn’t want to buy more. It also would have taken me a long time to recreate theses tests with different thicknesses of the glue. Thanks for your input on this!

 

Bradley

[marktiedens]

I use the gap filling CA from Bob Smith Industries also, but I use the use the odorless type with the gold label.  It takes a few seconds longer to "grab", there is no odor or fumes that I can detect & is even safe for use with foam.  Regular CA will melt foam.  I went to that because the fumes from regular CA really bother me.   The only drawback I find is that it`s a bit more expensive than standard CA.

 

 

Mark