When is a hot glue stick not a hot glue stick? When its PLA, of course! A glue gun that dispenses molten PLA instead of hot glue turned out to be a handy tool for joining 3D-printed objects together, once I had figured out how to print my own glue sticks out of PLA. The result is a bit like a plus-sized 3D-printing pen, but much simpler and capable of much heavier extrusion. But it wasntas simple as shoving scrap PLA into a hot glue gun and mashing the trigger; a few glitches needed to be ironed out.
Some solutions come from no more than looking at two dissimilar things while in the right mindset, and realizing they can be mashed together. In this case I had recently segmented a large, hollow, 3D model into smaller 3D-printer-sized pieces and printed them all out, but found myself with a problem. I now had a large number of curved, thin-walled pieces that needed to be connected flush with one another. These were essentially butt joints on all sides the weakest kind of joint offering very little surface for gluing. On top of it all, the curved surfaces meant clamping was impractical, and any movement of the pieces while gluing would result in other pieces not lining up.
An advantage was that only the outside of my hollow model was a presentation surface; the inside could be ugly. A hot glue gun is worth considering for a job like this. The idea would be to hold two pieces with the presentation sides lined up properly with each other, then anchor the seams together by applying melted glue on the inside (non-presentation) side of the joint. Let the hot glue cool and harden, and repeat. Its a workable process, but I felt that hot glue just wasnt the right thing to use in this case. Hot glue can be slow to cool completely, and will always have a bit of flexibility to it. I wanted to work fast, and I wanted the joints to be hard and stiff. What I really wanted was melted PLA instead of glue, but I had no way to do it.Friction welding the 3D-printed pieceswas a possibility but I doubted how maneuverable my rotary tool would be in awkward orientations. I was considering ordering a 3D-printing pen to use as a small PLA spot welder when I laid eyes on my cheap desktop glue gun.
A glue gun had everything I needed: good ergonomics, good tip visibility and tactical feedback, and simple mechanical operation. If it could be made to extrude melted PLA instead of hot glue, it would be the ideal tool for the job. After some initial tests and a discussion with colleagues, it was clear that trying to make this happen was worth possibly wrecking a cheap glue gun.
According to theRepRap Wiki entry for PLA, it softens around 60 C to 65 C and melts around 180 C to 220 C. Will a glue gun do the job? To find the answer to this question, I manually pushed a bundle of scrap PLA filament through a small desktop glue gun that I didnt mind ruining. The glue gun was a hobby unit made for lower-temperature glue sticks. The small hobby unit eventually melted the PLA but only barely; the PLA came out more like softened putty. Based onthis teardown of a cheap hobby glue gunthe operating temperature is expected to be around 150 C, which isnt enough to really melt PLA properly.
Another thing that became clear during this process was that the glue gun had special needs for extrusion and feeding. To feed properly, the trigger mechanism needed to be able to grip and push on a solid cylinder, not a bundle of filaments. In addition, proper extrusion required a solid shape that filled the opening of the melting chamber completely to prevent backflow. Otherwise, molten plastic prefers to spill out the back instead of being forced through the nozzle. In other words, I needed:
After a bit of research, I purchased an economical high-temperature glue gun that claimed 80 W and an operating temperature of up to 208 C.
Testing PLA Glue Sticks
Test with straight-walled plain cylinder of PLA. The back of the stick was printed with black so I could get an idea of where the stick ended when extruding.
To feed my new glue gun I needed a cylinder 11 mm in diameter and at least 5 in long. Happily, 3D printers exist for the sole purpose of turning 1.75 mm filament into other shapes and sizes. It felt a little strange to use a 3D printer simply to turn 1.75 mm diameter plastic into 11 mm diameter plastic, but in about an hour I had printed a high (75%) infill 11 mm x 150 mm cylinder of PLA for testing.
That first stick of PLA was enough to show that the 80 W glue gun was able to melt and extrude PLA acceptably; the only hitch was a ten to fifteen minute warmup in my cool workshop, compared to only five minutes or less for hot glue.
However, a problem was revealed. The feed mechanism for the glue gun has a small levered arm that bites into a glue stick and pushes it forward when the trigger is pulled. However, the PLA glue stick was smooth and hard, and the feed system could not properly bite into it. In fact, the ridged surface of the PLA cylinder quickly wore away the teeth in the little arm as it tried in vain to find something to grab onto.
The solution was a small change to the 3D model for the glue stick. Adding a series of angled notches to the cylinder model allowed the raised arm to latch and push perfectly.
This little lever rises when the trigger is pulled.
Showing wear on arm that rises to grip the glue stick. Teeth are notched from rubbing on PLA.
Notched PLA glue stick engages properly to the feed system.
With an 80 W glue gun doing the melting, and the notches in the cylinder of PLA allowing the feed system do its job when the trigger was pulled, molten PLA flowed easily and with excellent control. I performed some simple tests:
Using PLA like glue by depositing a molten blob, then mashing a part down on top of it. Results were good, but the PLA blob adds some bulk because it doesnt smear out fully before it starts to cool and harden.
Joining seams: the molten PLA melts only slightly to the surface being joined, but still ends up surprisingly strong. I was able to pull the seam apart but in the process always broke one or the other of the surfaces, or broke the seam itself in two (leaving the remains stuck to the parts I had joined).
Left and center: using PLA as if it were a blob of glue. Right: applying along a seam between parts.
The bottom surface is a scrap raft. Slight melting visible around seam on the right.
Results of pulling the pieces apart until they failed. The two on the left had lots of leverage. The small piece on the right could not be broken off without destroying it.
This part has been cut in two to see how much actual melting together between the parts has occurred. Not much, and superficial only.
Pulling the pieces apart until failure still took effort and resulted in the bead splitting in two instead of simply popping apart.
Is There Really Any Welding Happening?
A true weld results when the material of the joint and both pieces meld together to become one. That clearly isnt the case. There issomemelting together happening, but its superficial at best. Nevertheless, the result easily passes a tug test. The cut-away test joint shows the red PLA has flowed into and filled every nook and cranny, which probably accounts for most of the strength.
A small hobby glue gun reliably softened but did not melt PLA. A glue gun made for higher-temperature hot glue melted PLA acceptably.
The test glue gun was an economical dual-heater 80 W unit that did the job, but takes a good ten minutes or more to warm up and melt the plastic filling the chamber (compared to five minutes or less for hot glue.)
Aside from heat, the biggest issue keeping PLA from being used in a glue gun is that glue gun feed mechanisms are not made to grip and push on smooth, hard PLA. They expect a much softer hot glue stick. The test gun functioned well with PLA sticks that had notches in the side, but this will vary depending on the design of the specific glue gun being used.
The STL file for my simple notched glue stick isavailable on GitHub. It may or may not be compatible with other glue gun feed systems.
Its not much of a weld since its superficial only, but it still takes more than casual effort to pull pieces apart.
My large multi-piece 3D print was a big hollow object with thin pieces fitting together like a jigsaw puzzle, and using the PLA glue gun to quickly and efficiently stick those 3D-printed pieces together with no need for clamping was better than expected.
For situations where a PLA glue gun may not do the job, check outthis method to ease printing and assembly of multi-part modelswith a bit of help from OpenSCAD. But for my big hollow model, manually aligning the surfaces and bonding the inside seams was quick, easy, sturdy, and absolutely worth repurposing a $30 glue gun and giving it a place in the tool drawer.
Posted in3d Printer hacksFeaturedHackaday ColumnsSkillsTool Hacks
Tagged3d printingdiydiy toolextruderglueglue gunglue stickPLAplasticplastic weldingtoolswelding
Repairs You Can Print: A Little Love for the Glove Box
3D Printering: Printing Sticks for a PLA Hot Glue Gun
3D printing is the bitcoin of CNC world.
is the LED strips of the electronics world
Would be cool to make a 2-piece mold of the notched stick, then put scrap PLA into it & heat it up to make the sticks. This would result in recycling scrap pieces into a usable item.Jim
Excellent idea! If this mold were made from porcelain or high temp plasticmaybe siliconeone could simply use the microwave a recycle all scrap material into usable sticks. Good thinking.
***EDIT*** was supposed to be and recycle all scrap, not a recycle all scrap. ***EDIT***
note: I accidentally clicked Report instead of Reply. Mods, please dont remove the parent post.
When my hot glue gun had used up most of the glue stick, I wrapped a regular graphite pencil with electrical tape and inserted the pencil in the gun, so it pushed the remaining glue inside. No printed sticks necessary 😀
2-part silicone will do the trick. Safe up to something around 300 C, far above PLA mt. Does PLA melt in a microwave? Id be concerned about thermal degradation though, as the temperature will be totally uneven in a microwave.
Interesting, but I think regular hot glue would still be superior. The softness helps in keeping parts from breaking apart.
Maybe hot melt is stickier but Ive found in the past its easier to peel glues off when theyre more or less flexible than the attached material. Id go hot-glue for flexible stuff, epoxy for hard stuff. Also it takes more surface prep to bond dissimilar materials well.
Ever since I switched to Steinel hot glue I dont have this problem / handy feature of hot glue being detachable anymore. That stuff in the extra sticky version (I think Bosch sells the same stuff for their cordless glue guns) will never come of anything (except glass and cold surfaces) ever again. Especially not from PLA.
I got my MP 3d printing pen for $15, has heat & flow control, and I dont have to make sticks. If you plan ahead, you can leave a little void in each piece that accepts the nozzle and allows you to inject the PLA between the parts. Ive welded some parts together that are like one piece afterward.
Leaving a void to fill up with molten PLA is a very cool idea. Have you written any of that up? ()
I mentioned elsewhere that it might be possible to drag the glue gun tip along the joint to make a void after printing and, this will heat up the surfaces to be joined before hitting them with the melted pla. Just an idea i am going to try. This might make for more of a weld.
I found this worked well with PVC conduit that I was gluing together but was initially too smooth for the hot glue to sufficiently adhere to. My Bosch glue gun nozzle is certainly hot enough to melt plastic it comes into contact with. Let us know how it goes for you with PLA.
Why not use a solvent to bond the parts?
Because its not using a 3D printer and that train of thought simply wont do..
They have their uses but PLA solvents can be pretty nasty, can spread out to where you dont want it, and dont work as well when you try to join thin pieces edge-to-edge.
Maybe you didnt apply it correctly?
For ABS Id definitely use a solvent but solvents for PLA are not as easy to find.
(My go-to glues for prints are either E-6000 or IPS-16 but neither is a solvent for PLA.)
Are they restricted in your country?
PLA is surprisingly chemically inert stuff, so much that its kind of a pain to glue securely because you only get surface bonds.
The only reasonably common solvents that will readily dissolve PLA are Chloroform and Toluene, neither of which compare favorably in ease/safety to ~200C heat.
I havent personally, from what Ive read and my understanding of the chemistry it should work fine, with the caveat that DCM isnt the best thing health-wise, and is inconvenient to protect against because both Latex and Nitrile are soluble.
I did try a nominally-DCM-bearing pipe cement with some PLA samples and the bonds were very weak, but I believe the particular glue was really mostly Cyclohexanone.
Nitrile is the recommended protection for handling small quantities but if you want to soak your hand in it you need Viton neoprene butyl etc
Can you use a heat gun to heat up the target surface enough so that it is melty and run your PLA glue gun over that?
Im either seeing a starter step where you flash the area about to be glued with a heat gun to make it surface molten, then use the PLA glue gun, or you attach a well insulated metal hose to blow the hot air from a heat gun onto the surface (similar to the way welding equipment might blow shield gasses over an area being welded) so that the PLA is being put onto a surface that is already slightly molten.
I did hit some PLA prints with a heat gun out of curiosity once, but it always deformed the print badly because the PLA softens waaayyyy before it actually melts. Maybe if the hot air were hotter and more focused?
Nice! I like that I can use a tool I already have instead of getting a new hand-held 3D printing pen.
What Ive been doing is plastic welding my 3D prints just like they do for repairing car bumpers except Im using printer filament and an hot-air SMD soldering pencil instead of a dedicated plastic-welding gun. My joints also look smoother and I think stronger because the filament and the target pieces all get heated. It does require two hands unlike the glue gun.
I saw Bill Doran (Punished Props) trying a 3d printer pen as a welder but the his results werent strong enough. He ended up super gluing the bits and then using the pen for weld line aesthetics. I wonder if your thicker PLA stick carries more heat with its additional mass, that could be improving the bond vs a regular printer pen.
Good article. I think with just a bit more tweaking and prep, this could be really very useful indeed. When I get home, I am going to measure the temp of my hot glue gun. Thanks.
Also, I am going to try slowly dragging the glue gun tip across the joint to heat it up, possibly an inch or so, then hit it with the melted glue just to see if that helps get an actual weld.
I never really came up with a good way to measure how much the glue gun was able to heat the PLA. Taking temperature at the tip with a non-contact thermometer was unreliable. If you have a good solution, let us know!
I have a digital thermometer with a remote thermal probe I should be able to use on the tip. If the tip gets anywhere near 200-210 C it might work to melt a small channel to be filled with the melted pla. I bought a decent glue gun a year or so ago but it may not be the type that gets hot enough for this. Ill see what I can do.
Did you have a link to the gun you used so that the printer grooves will match up? Great idea btw. I use glue guns for this all the time but a little more heat could make a difference
The one in the header image for the article is an 80W High Temperature Surebonder I bought from Amazon. I saw a lot of glue guns of similar design with different names, and just picked one.
BTW ever since this entry about a repair for the trigger arm in a glue gun ( I suspect that the same lever-arm arrangement is probably actually quite common for glue guns.
I would imagine hot glue used as a place holder, just a drop/blob to hold the pieces together and then CA glue (and baking soda) to set the pieces in place after which you can remove the hot glue if you want. I guess another way to put my thinking is spot weld with hot glue and then weld with CA glue. If you are dead set on welding with PLA and a rotary tool, why not use an extension that allows you to hold the tip like a pen?
As Punished Props was mentioned, they did a video with 3d Printing Nerd related to different ways to adhere two prints together. Superglue(CA glue) turned out to be about the best if not the best way to stick two prints together for strength.
That is a great video indeed. I love super glue and after learning the baking soda trick have had even more love for it. That being said I also love hot glue and this idea totally ticks all the right boxes with myself but I think you are right, that super glue is still the best way to connect two plastic prints at least in my experience.
Have you tried getting different hot melt adhesives? The standard (mostly wax) hot melt is pretty useless but there are a lot of glues of varying stiffness, strength and intended substrate that work incredibly well. 3M has a whole line, but just google and find what you need:
Just dont get it on the skin, you cant get the molten blobs off!
should have added, the pla through a glue gun is a neat trick, Ill keep that one in my back pocket.
Whoa. As usual, a visit to 3M shows how much wider the adhesives world is beyond what one sees on the shelf at the hardware store!
How about pre-heating the joint with the tip of the too-cold-for-PLA hot glue gun? Safer than using a soldering iron.
Also, perhaps after using the PLA glue gun to weld the pieces together, put down a layer of low-temp hot glue on top of the more critical PLA welds. At that point speed wouldnt be an issue.
Thats true. In fact the process I settled on when assembling my pieces was to anchor one corner while being very careful to align it first. Then with that corner anchored (it didnt take long because the PLA cools and hardens fast) do the opposite corner and work my way around. It was a little like hand soldering fine-pitch SMT parts. The fast cooling of the PLA really shines here. Its hard to overstate just how much faster and easier the whole process becomes as a result.
Once the piece was spot-anchored at the corners, I simply ran a PLA bead down the seams on the inside but could have used something else. I liked how this meant the presentation side of the seams didnt have glue or anything in them mucking them up & making filling and sanding harder, and again it was very fast to do.
Like you say, once the corners are anchored you could use anything you like to reinforce and also as you say, speed wouldnt be an issue any more.
What about printing 1/2 of a stick (lengthwise) and cutting a glue stick in half and then feeding the two together. Im curious how PLA/glue would set.
FWIW, if you are creating the model, rabbeting the edges with the occasional rivet hole, melting plastic rivets in the holes, then smoothing the seam with a soldering iron hot knife works *VERY* well. Now that I have some Polymorph, Im considering using it for /semi/permanent rivets, as it /melts/ at 60C, allowing me to take things apart with only hot water.
I think you mean tactile feedback. Tactical is a different word.
Fill in your details below or click an icon to log in:
You are commenting using your account.(Log OutChange)
You are commenting using your Twitter account.(Log OutChange)
You are commenting using your Facebook account.(Log OutChange)
You are commenting using your Google+ account.(Log OutChange)
Notify me of new comments via email.
3D Printering: Printing Sticks for a PLA Hot Glue Gun
Black Holes and the Elusive Mystery That Lies Within an Equation
What Are Those Hieroglyphics on Your Laptop Charger?
The Hard-Learned Lessons of the Columbia Disaster
3D Printering: Printing Sticks for a PLA Hot Glue Gun
Making the Case for Open Source Medical Devices
New Part Day: I2C In, Charlieplexed LEDs Out
l_szabionCircuit Bent Casio SK-1 gets an Arduino Brain
wossatfor on3D Printering: Printing Sticks for a PLA Hot Glue Gun
hackadave onBlack Holes and the Elusive Mystery That Lies Within an Equation
Ostracus onBlack Holes and the Elusive Mystery That Lies Within an Equation
Michael onBlack Holes and the Elusive Mystery That Lies Within an Equation
dexdrako onHackaday Links: February 4th, 2018
steve on3D Printering: Printing Sticks for a PLA Hot Glue Gun
steve on3D Printering: Printing Sticks for a PLA Hot Glue Gun
dexdrako onHackaday Links: February 4th, 2018
Tommy Kjær Mikkelsen onMaking Smoke That Really Performs
Dr. Cockroach wrote a comment on Plywood Excavator.
Dr. Cockroach liked Plywood Excavator.
Craig Hissett liked Plywood Excavator.
Mike Szczys wrote a comment on Plywood Excavator.
designlab has updated the project titled Inside the DesignLab Podcast.
Ted Yapo wrote a comment on project log nGame Revisited.
Mike Szczys liked Humanoid robotic arm with EEG control.
designlab has updated the log for Inside the DesignLab Podcast.
Tophness liked Onero – Sign Language Translation Device.
Hackaday, Hack A Day, and the Skull and Wrenches Logo are Trademarks of Hackaday.com
Powered byWordPress.com VIPPost toCancel>