Hi All,
I’m going to print (by necessity) some fairly small gears. When I look at the slicer rendering the .4 nozzle is not providing a detailed enough outline of the teeth. I have a range of nozzles from .01 all the up to .1. When I slice it using a .01 nozzle the teeth look perfect. Has anyone tried using a .01 nozzle? I was thinking of giving a .01 thick /.05 layer height a try. Thoughts?
John
No personal experience but cnc kitchen just released a video of printing with a 0.1mm nozzle.
I assume you also mean 0.1mm and not 0.01mm
Great article. Thanks. Could you imagine the detail of a .01 nozzle…
No, I couldn’t. I would have died by a heart attack, just looking at the total print time it would take :-p
2 Likes
I’ve had good luck printing a 0.05mm layer height on my v1, but the forum recommendation is to use a multiple of 0.04mm for your layer height. I’ve done both 0.04 and 0.08, and don’t really see a quality difference between 0.05.
.04 doesn’t make any sense. I thought the maximum axis resolution is .05mm. Which is the smallest movement used when calibrating the print head. If it does have a finer resolution then I would think that it would be available for that procedure.
@Thick8 limits of layer resolution isn’t dependent upon the linear modules alone, it’s also dependent upon nozzle diameter, change the diameter and your capable layer resolution can change as well. It’s also supposed to be in increments of the nozzle diameter as well, so for example a .4mm nozzle would have resolutions .08 .12 .16 .20 .24 and so on.
Why would that be the case? That doesn’t make any sense to me. The usual recommendation is that the minimum layer height is approximately 25% of the nozzle diameter and the maximum layer height 75%. As to guarantee that there is room for the plastic to extrude or to make sure that there is a certain amount of squish to make sure there is sufficient adhesion between layers.
Other than the the capabilities of the stepper motors and/or linear guides I don’t see any reason why there would be any maxic numbers like those depending on the nozzle size.
I think this is the post I read explaining why.
I hadn’t heard about layer height being a multiple of nozzle diameter, other than maximum layer height should be 80% of diameter. Which I had already broken by printing stuff with a 0.4mm layer height using a 0.4mm nozzle. It’s my prototyping quality setting, for when I need to print a model I know isn’t quite right yet.
So I’ve printed both 0.05mm and 0.4mm layer height parts fine on my v1. The 0.05mm did take some extra effort to dial in the calibration.
1 Like
Here’s a brief synopsis of a few things.
General printer recommendations:
Extrusion width range (absolute min to max): 60% to 200% of the nozzle width (generous)
Layer height range: 25% to 75% of the nozzle width.
Origin of the “magic number” and where 0.04mm comes from:
Given: Driver using x16 microstepping, 4-start 2mm-pitch, 8mm lead screw, 1.8 degrees of rotation per full step of the stepper motor.
- 360 degrees in a circle / 1.8 degrees/full step = 200 full steps per rotation, x16 = 3,200 microsteps per rotation
- Steps per mm = 3200microsteps/8mm = 400microsteps/mm
- Minimum distance for a full step = 16(microsteps/full step) / 400(microsteps/mm) = 0.04mm/full step
The reason 0.04mm so called “magic number” is important is because when stepper drivers employ microstepping each microstep will have less available torque than a given full step. So there will be 15 weak steps followed by a strong full step (greatly oversimplified, the torque curve is sinusoidal).
The logical conclusion of this train of thought is that using layer heights of multiples of the magic number will sidestep the torque reductions of microstepping by ensuring at least a full step has occured to generate the full commanded motion.
I’m not sure this is all strictly necessary on the snapmaker. I have measured a minimum travel consistently around 0.015mm. There’s a number of possible explanations, but I believe it’s because of the large mechanical advantage of this particular lead screw system, combined with the particular driver being used. This may not be true on the new TMC drivers.
But the minimum motion is not 1/400mm. 6-ish (micro)steps seems to be around the minimum.
2 Likes
That post indeed explains the link with the step size of the the stepper motors (as @brent113 explained in as well). However, there’s no relation with the nozzle size as @Artezio claimed.
And I too usually use a bigger extrusion width than the default. Especially for prototyping and functional parts. Usually 1mm width for first layer and 0.3mm height and for subsequent layers 0.6 to 0.8mm width with a 0.4mm nozzle (or 150%-200% usually closer to 150%). cnc kitchen has a nice video on that: Extrusion Width - The magic parameter for strong 3D prints? - YouTube. Depends a bit on the size of the part and thickness of the walls that I tweak that parameter. It’s often a huge timesaver. And if looks are important it’s sometimes possible to use a different parameter for the outer wall (or printing the outer wall first) to have a nicer surface finish if that’s important.
(But this is diverging from the original question)
1 Like
@brvdboss well I could be wrong. It’s just what I’ve been told from different people at different times.
1 Like
While a full step is 0.04mm, there is no easy way to know if a particular position is on micro-step 0, or micro-step 15, especially when you start a print. This nullifies the concept behind sticking with a movement that is a multiple of it, unless you can guarantee that the first layer is on micro-step 0. If you happen to be on micro-step 7 or 8, you’ll be at the weakest torque and holding position, but that should not matter all that much when using the 3D Printer Module. The only number of real importance is the minimum movement resolution of 0.0025mm.
This whole discussion about resolution makes me wonder why the arbitrary value of 0.05mm was chosen as the minimum adjustment value when calibrating the printer. There are some times when doing a manual calibration that there is a noticeable difference in the drag of the paper/shim between calibration points. Having access to smaller incremental changes would, IMO, go a long way to correcting some of the bed leveling issues people have.
Just use the terminal Luban to jog, and not the touchscreen for that part, Luban does custom values. I use 0.02
2 Likes
I had no idea. The only time I hook my laptop to the machine is when I’m resetting the extrusion value after a firmware update. I’ll give it a try. Thanks.
I recently made a small replacement gear for a robot vacuum cleaner with a 0.2 nozzle. The gear is used for a rotating brush. The gear consists of a larger cog wheel with 40 something teeth and less than an inch in diameter, combined to a very small cog wheel with eleven teeth. The part printed surprisingly well with very good details and fitted perfectly, but equally important is the material you use. Nylon would probably be best, but it requires heat and can be tricky to print. And I didn’t have any.
So I made the first try with Armadillo, which printed fine but broke almost immediately in use. The layers weren’t strong enough for the relatively strong torque. If Armadillo couldn’t take it, then I could immediately forget PETG too (PLA or ABS I wouldn’t even try for gears).
The next try was with PP of which I had a small sample. It is softer, but very tough and has very strong layer adhesion. This part is still working. I also printed a version from a relatively stiff D60 TPU, that feels like it might work well too. It won’t break at the layers, and the teeth won’t snap, but they could be too soft and skip under force. I will try it if the PP part fails, but have not needed to do it yet.
Back to the size of the nozzle. For this project my 0.2 nozzle was perfect. It gave the required detail for the tiny teeth, while still being manageable to get the settings right. At this size even the slightest over extrusion lose the detail and the slightest under extrusion cause the part to be weak. Any smaller nozzle and extrusion calibration really needs to be microscopically spot on. To get extrusion perfect I sliced several variations: the calibrated profile for the filament + variations with 1% and 2% higher and lower extrusion multipliers. If the first version wasn’t perfect, I did the next (not much filament wasted at this scale).
CNC Kitchen tested a high end 0.1 nozzle, but he only used a well known machine and most suitable filament he knew. At this scale the amount of extruded plastic is really small, with ridiculously low tolerances for extrusion, nozzle temp, cooling, speed, filament diameter tolerances etc. Not only to get it to print, but to also make it strong enough to be usable. And not many filaments can even be extruded through such small nozzle without clogging, which may eliminate those that otherwise would be suitalbe for the part.
Thanks for that.
I bought a roll of Overture nylon Amazon.com: OVERTURE Nylon Filament 1.75mm 3D Printer Consumables, Polyamide ¶ 1kg Spool (2.2lbs), Dimensional Accuracy +/- 0.05 mm, Fit Most FDM Printer (Black) : Industrial & Scientific
Now to find an all metal 0.2mm hotend. I think I saw one here the other day.
Hi again.
I’m not aware of any all metal hotends for the Snapmaker, although I saw one thread where the Biqu one would fit. If you find something else I’d like to know.
On the other hand Snapmaker says the max temp is 275 C. If you google PTFE, most sources say there is no degrading before 260 degrees and more noticeable decomposition starts at around 300 with melting point being 327. I couldn’t find the required printing temp for your nylon, but it shouldn’t be higher than some ABS brands. From personal experience, I’ve printed several ABS jobs (50-150 gram objects) at 260-265 and the PTFE tube is still as new. When it finally degrades I’ll replace the hotend…
As for toxic fumes I don’t know, I’m not spending my time sniffing the snapmaker while it prints for hours… It is in the garage where I do all kind of things with much nastier fumes and thus have a decent ventilation. I would guess a frying pan would be much more of a safety issue, it can be much hotter, has waaay bigger area of teflon in direct contact with heat and is also in direct contact with food (compared to a very tiny part being inside the actual hotend). I guess there are differences in material formulations as well.
1 Like
I went with these. The measurements matched the Smapmaker parts. Supposed to be here Tuesday. I’ll post something later on in the week.
Amazon.com: BALITENSEN MK8 Extruder Thermal Barrier Heatbreak - M6x30mm All-Metal 1.75mm Throat Compatible with CTC Wanhao Makerbot 3D Printer (5 Pack) : Industrial & Scientific
Amazon.com: uxcell 0.2mm 3D Printer Nozzle Head M6 Thread Replacement for MK8 1.75mm Extruder Print, Stainless Steel 5pcs : Industrial & Scientific