Threading PETG rod plastic (3d-printed)

Hello, I am trying to thread some PETG plastic rod which was 3d printed. I need to end up with an adapter of sorts having a threaded M4 0.75 and M5 0.8 portions on opposite ends. To that end I am 3d-printing approximately 5mm PETG studs and attempting to thread them using some metal dies I had on hand. However, I've already gone through dozens of studs without getting workable results.

I don't have a heavy-duty vise, so I'm using a drill to hold the PETG studs while I attempt to thread them from the opposite end with the die. The studs have a conical taper to make it easier for the die to catch. Eventually it does catch, but whatever speed of the drill I use (I go nowhere near actual drilling speed, that would quickly heat up and deform the plastic), and what pressure I apply, I end up with very mangled threads. Obviously every time the die spins too much without moving forward, it annihilates the freshly cut threads. But I cannot figure out how to prevent that.

The weirdest thing is that the first time I tried threading one of the studs, by hand, it worked pretty well and I assumed it would be easy. I've since not been able to get it right though. I am very new to machining and I'm afraid I don't know how to go about getting this done. Could someone please explain a reasonable method for threading this soft plastic without destroying the threads? Would really appreciate any ideas!

I 3D print a lot of PETG and I can tell you it's not very thread friendly.

As you have noticed it's very soft, and has a low deflection temperature.

All you can do is go slow and put something slick on it, PETG is pretty chemical resistant so WD-40 or mineral oil will help.

The easiest thing to do would be switch to a different plastic; ASA comes to mind, easier to print than ABS but hard enough to thread.

Why not just print the threads? Fusion360 will generate modeled threads, then just clean them back up with a die. Having it generate the .stl will take longer than modeling it.

I print in PETG among other things but I haven't printed threads of that size, done some m8 stuff with zero problems though in petg.

I would suggest a soft wax as a lube and use something slower than a drill so you have more control. Threading dies are meant for metal, for plastic as they say "your results may vary".

My big question is WHY are you 3D printing the studs rather than buying 5mm rod? IMO even if you succeed the threads will be very weak and may fail in service.

Thanks for the replies everyone! I will try putting some lubricant on it to see if it changes things. I don't have ASA filament and it would probably be easier to buy the adapters rather than get a spool of ASA just for this project. I was hoping I could do this with the tools I had to save time and money.

I'm printing the rods horizontally for added shearing strength, so I don't think printing the threads would work. Even if I printed the studs vertically, I'm not sure what the results would be like for such small threads. At any rate it would fail more easily because of how the layers are located.

I basically soft-press the drill trigger to get it to go very slow (don't have anything specialized for going slow). I understand that my results may be inconsistent, but I did get it to work once so was hoping for at least a couple of successful ones.

There are several reasons I'm not buying 5mm rod. One, I don't have easy access to metric rod that would ship fast and be cheap. Two, I believed I'd just do this in an hour or so with the materials I have on hand, which proved to be false. Third, I would have to turn the 5mm rod down to 4mm so I could thread the M4 portion. I do believe there's enough strength in the PETG threads for the intended application, it's not exactly going into a rocket.

Tried putting some cutting fluid on the studs and going quite slowly. Thread appears slightly less mangled than before but unfortunately still unusable. At this point I am stuck. This might just give me an excuse to learn to use my lathe, unless someone has another magical suggestion.

How long of threaded portions and how much load is this thing seeing?

The threaded portions are approximately 15mm each, and the load is a back-and-forth manual cleaning motion inside a bore, with a wire brush, so not much. My guess is occasional assembly-disassembly will eat the threads faster than the load, assuming axial layers. If the layers are radial though, I think it will break fairly quickly from layer delamination, since the thinnest portion will be less than 4mm.

Are you constrained by size? Perhaps you can print an adapter block that holds an M4 and M5 bolt out in opposite directions.

You will have more control if you use the drill chuck simply as a vise, and turn the die using a die stock, by hand.

Don't run the drill.

Strostkovy said:

Are you constrained by size? Perhaps you can print an adapter block that holds an M4 and M5 bolt out in opposite directions.

Click to expand...

The people who sell this cleaning kit (that I need the adapter for) pretty much took care of that idea when they used nonstandard threads. All my metric threads are the standard kind. Otherwise this is good thinking.

awander said:

You will have more control if you use the drill chuck simply as a vise, and turn the die using a die stock, by hand.

Don't run the drill.

Click to expand...

Ok, will try that as soon as my next batch of studs is ready.

Ok, so something clearly worked, at least partially. First, I printed the studs slightly oversized (because being the newbie that I am, I didn't measure the studs initially, forgetting that my printer likes to print ovals in the XZ plane). Second, I used some cutting fluid. Third, I went very slowly, by hand, after the studs have thoroughly cooled post-print. I can now get a consistent M5 thread on there.

The only issue with that part of the thread is that backing the die off, however slowly I do it, still destroys the threads! This means I can get a full or nearly full length M5 thread, but can't get the M4 thread on there, as I'll have to reverse the M4 die at some point to preserve a portion of the M5 length... Do I have any options?

anvoice said:

Tried putting some cutting fluid on the studs and going quite slowly. Thread appears slightly less mangled than before but unfortunately still unusable. At this point I am stuck. This might just give me an excuse to learn to use my lathe, unless someone has another magical suggestion.

Click to expand...

You have a lathe and still think you'll succeed with flimsy 3D printed plastic?

With a lathe and threading dies you have what you need to make it from something like Acetal. Buy some 1/4" rod and turn part down to 4mm. Thread it with the die, turning the spindle by hand.

Slide more rod out in the chuck and turn it to 5mm. Cut it off, reverse it and thread the same way. Done deal.

For a simple job like this you don't need to be a lathe expert. You can "sneak up" on final dimensions with a fine cut file. The same file can chamfer the ends good enough for threading.

IMO you will never get satisfactory results with a printer. They are not made for such stuff and even if you succeed the threads will likely strip in use.

To have a lathe, any lathe, and substitute inferior methods is foolish. You can learn what you need faster than making 3D printed parts work.

At this point, you're probably right. How difficult this has been is not a good indication for the reliability of the final part, if it works.

I was hoping to save myself a trip to the store and learning how to use the lathe from scratch. I literally did nothing but test the motor. However, that might be my best option at this point.

As you have discovered, this is not an easy thing. I suspect one of your problems may be the in-fill. 3D printed parts are NOT solid unless the settings call for that when it is printed. Fill densities range from 10% up to 100% and many parts are printed with a 25% or 50% in-fill so they are between 1/2 and 3/4 AIR on the inside. There will be an outer shell that is solid, but it's thickness also varies according to the settings when the object (your rod) is printed. That is probably not thick enough to allow threads, even a relatively fine pitch like 0.75mm. You are probably either breaking through that outer, solid layer or leaving a very small amount of it, not enough to withstand the stresses of threading with a die.

I would:

1. Use a relatively dense in-fill when 3D printing them, say 75% to 100%.

2. Use a relatively thick outer shell, at least twice the depth of your 0.75mm thread (1 mm).

3. 3D print the thread. You will need to use a very fine nozzle for this. 0.2mm or finer if you have it.

Even with a fine nozzle your 0.75mm thread will probably not be well formed. So you will still need to go over it with a die.

4. DO NOT use power to thread it with the die. Do the threading by hand.

I am not sure what kind of lubricant would be best, but you should use one. You might start with a liquid dish washing detergent. Or perhaps others here may be able to suggest a better one. But DO use something to cut the friction. And back up the die often to allow the LUBRICANT to get where it is needed. It is not being used as a coolant, it is a LUBRICANT.

If and when you get a few good ones under your belt, you may experiment with skipping some of the above.

Good luck! Happy chips!

What is everyone's thoughts on chilling the plastic to make it cut rather than mush? I know brittleness is an issue then, but maybe there is a fine line?

EPAIII said:

1. Use a relatively dense in-fill when 3D printing them, say 75% to 100%.

2. Use a relatively thick outer shell, at least twice the depth of your 0.75mm thread (1 mm).

3. 3D print the thread. You will need to use a very fine nozzle for this. 0.2mm or finer if you have it.

Even with a fine nozzle your 0.75mm thread will probably not be well formed. So you will still need to go over it with a die.

4. DO NOT use power to thread it with the die. Do the threading by hand.

I am not sure what kind of lubricant would be best, but you should use one. You might start with a liquid dish washing detergent. Or perhaps others here may be able to suggest a better one. But DO use something to cut the friction. And back up the die often to allow the LUBRICANT to get where it is needed. It is not being used as a coolant, it is a LUBRICANT.

If and when you get a few good ones under your belt, you may experiment with skipping some of the above.

Click to expand...

Thanks for the thoughtful post! I'm actually not that clueless about 3d-printing: my studs are all 100% infill, with plenty of perimeter thickness. I doubt I can print the threads, even with a fine nozzle, as I am printing the cylinders with the axis horizontal to the bed, so layer de-lamination is not an issue. However, your idea about backing up the die often is something I haven't tried, and probably should. I'd still be out of luck if backing up the die destroys the threads, but it's worth a try.

Strostkovy said:

What is everyone's thoughts on chilling the plastic to make it cut rather than mush? I know brittleness is an issue then, but maybe there is a fine line?

Click to expand...

Clever! I should have thought of trying that.

I did find some 3/8" aluminum rod and will attempt to make the part on my lathe as well. I'm a bit worried now about the strength of a plastic part (less than 4mm diameter is small) but for the sake of science I'll explore the PETG method further to see if I can get it to work. Perhaps it'll be useful to someone in the future.