Glass bed temperatures

If you use a piece of glass on the heated bed of your 3D printer, how does the temperature measured by the thermistor compare to the temperature on the surface of the bed? The thermistor is underneath the bed, close to or in contact with the heater, and glass is a poor conductor of heat. So how do they compare?

I set up an experiment where I put a glass plate on a MK3 heatbed, and measured the temperature using the thermistor in the middle of the MK3 and a second thermistor taped to the top of the glass. This is not a very scientific test, but it gives a rough idea. It's a mistake to treat the values from the two thermistors as strictly comparable, as they they are not very precisely calibrated, but looking at the difference between them provides some insights. For the experiment, I started with the heater off and the bed cool and then raised the temperature 10C at a time from 30C to 80C, waited for the values to stabilize, read the temperatures and took their difference. I did the experiment on three different kinds of glass: window glass, thickness 2,2mm; mirror glass with a silvered backing, thickness 2.9mm; and borosilicate glass, thickness 3.1mm. The values are stable to within about +/-0.2C. In addition, for one reading on each material, I estimated how long it takes for the temperature on top to settle after the temperature on the bottom has done.

Here's the results. A delta of 1.9 means the reading on top was 1.9C below the reading underneath, so for example 28.1 if the reading underneath was 30.2.

Set temp	Delta (window glass)	Delta (mirror glass)	Delta (borosilicate)
30	2.1	2.5	2.3
40	3.7	5.3	4.0
50	6.2	8.4	6.5
60	7.8	11	8.1
70	11	13.7	11
80	13	17	13

I estimated the settling time at the 40C reading, and it was around 30s in each test, perhaps slightly more for borosilicate.

Some things to note. The delta gets larger at higher temperatures. The mirror glass delta gets larger at a greater rate than the window glass delta. This is important, as the deltas are not strictly comparable due to exactly how the top thermistor was attached to the glass. This wasn't apparent for borosilicate (and I don't know what happened with the last reading). To make the point, here are the deltas for mirror glass and borosilicate minus the delta for window glass:

Set temp	Delta (mirror glass-window glass)	Delta (borosilicate-window glass)
30	0.4	0.2
40	1.6	0.3
50	2.2	0.3
60	2.2	0.4
70	2.7	0.2
13	3.3	-0.6

What does this mean? First, don't trust absolute temperatures. For example, if someone says print on a bed at 60C, figure out what this means for your printer, as you might have a different bed configuration. This shouldn't be news to most people.

Second, if you are switching between different bed types, you probably don't need to worry about changing the temperature if you are printing at 50C (my usual value). It's only a couple of degrees different between the different types. But at higher temperatures, such as 80C and above, you might need to do some adjustment.

Thirdly, consider inserting a delay in your start gcode after the bed reached temperature and before you start printing. If you are heating the bed first and then the hotend, you probably have enough delay already.

You should treat all this with a healthy dose of skepticism. I was really interested in just one question: what should I expect when switching between the piece of glass that I have? It's not a detailed study or guidance on the kind of glass you should choose.

The extrusion multiplier

There has been some discussion on the Folger thread at http://forums.reprap.org about setting the extrusion multiplier. I've noticed in the past that my printer tends to over-extrude. You see it especially on holes which often print smaller than they should be, and on mechanical parts such as gears. Recently I've been trying to print this pieces for http://www.thingiverse.com/thing:1249221 and it's very apparent.

I did an experiment to get some data on this. When I was still fairly new to 3D printing, I was advised to set the extrusion multiplier to less than 1.0 in cases like this, but I've not previously collected any data on it.

I have already calibrated the E-steps for my extruder, so that if I manually extrude 100mm of filament, I really get that much to within 1 mm. To conduct the test, I created a 10mm cube and then extruded it in vase mode. Vase mode guarantees that there is a single strand of filament, so by measuring the thickness of the (vertical) walls you can see how much filament is being extruded. I did this for four filaments, at extrusion multipliers of 1.0, 0.9 and 0.8 using a 0.4mm nozzle. I also measured the diameter of the filament, by picking three places and at each measuring the diameter twice, with the measurements at right angles to each other. The diameter measurement are a little imprecise, while the wall measurements are probably more accurate. I use a E3D hot end, printing onto blue tape.

The results look like this:

Filament               Wall thickness (mm)          Ave filament diameter
                       x1.0    x0.9    x0.8         and range (mm)
MakerGeeks PETG        0.49    0.44    0.38         1.74 (1.71 to 1.78)
JustPLA PLA            0.50    0.46    0.38         1.78 (1.76 to 1.81)
Colorfabb PLA/PHA      0.48    0.43    0.39         1.75 (1.72 to 1.78)
Meltink3d PLA          0.44    0.42    0.36         1.69 (1.63 to 1.75)

It isn't clear to me why the value are so much larger than 0.40. The numbers looked so close to 0.5 that I even wondered whether I have the right nozzle. However, I use a E3D and the markings on it (dimples in the sides) indicate it is 0.4mm. Another hypothesis is that the weight of the next layer causes each one to squash out a bit; I can't really test this well, but trying to measure just the top layer suggests it isn't the case.

Whatever the reason, I'm going to try setting the multiplier to somewhere around 0.85 for a while and see if I like the results. It may be make PETG printing a little tricky. Reducing the multiplier makes the first layer stick less well, and with this particular brand of PETG (and maybe others, I don't know), that's already a problem. Making the first layer a bit thicker should help.

Printing mechanical parts, part 3: clockwork

I previously wrote about printing static and dynamic mechanical components. Here I take it a step further, with a simple spring driven motor of the sort you might find in a clockwork wind-up toy. I took this design (PLA Spring Motor Demonstrator) as a starting point, and modified in a few ways:

• made the shaft diameters smaller and some of the holes bigger to make the tolerances less tight.
• replaced a shaft that was build into the drive spring with one built into the frame.
• changed how the spring, pawl and winding knob attach.
• mounted the spring outside the frame, so I can easily swap variants of it in and out.

The modified parts are on Thingiverse. Here's a couple of pictures:

And here is the mechanism running:

(If the video doesn't work, try this link instead.)

I lightly lubricated it with 3-in-1 oil. With the standard spring, it will run for about 10 seconds. I also tried a spring made from PETG which runs for a couple of seconds more. Springs made from PCTPE and Semiflex did not work at all. I assume they just don't store enough energy. I made a thicker spring with a 4mm x 6mm cross section instead of the original 2mm x 6mm one. It runs for less time, but also runs faster. The 2mm one sometimes did not start without a slight nudge, presumably because it doesn't have quite enough force to overcome the static friction. The 4mm one started every time.

Printing mechanical parts, part 2

A few weeks ago, I wrote about printing mechanical parts, focussing on the static aspects such as making screw holes large enough. The conclusion was not very satisfying: basically, that you have to allow clearance, but neither a fixed amount or a fixed ratio of the size seems to work consistently. One thing I did find was that there wasn't much (though some) variation with the materials and printing conditions.

Since then, I've put a little time into printing dynamic parts, particularly gears. I didn't approach this in the same systematic way, and instead just have a few examples and general comments. All of the models I used for this come from sources on the web, sometimes with a little modification.

The NASA space wrench
NASA famously designed a small wrench and emailed the file to the International Space Station, where the astronauts printed it. I don't know what kind of printer they have, though I suspect it's a bit better than my Folger 2020. The wrench prints as a single part, and then applying a little force splits off the ratchet while still leaving it captured within the overall shell of the wrench. Here's mine, printed in PLA using my standard settings (30% infill and, I think, 0.2mm layers):

It doesn't move smoothly, but it does work, without any modification to the model or tweaking of the printer settings.

Planetary gears 1 (with base, without base)

I tried this set of planetary gears twice, once on PLA/PHA (red) and once in PETG (white). It will just about turn but tends to jam. I think this is for a couple of reasons. First, the printing leaves a raised seam. It might be possible to improve this by using the Slic3r setting to randomize the seam position. Secondly, the triangular piece which holds the large gears has holes with only eight sides, and the same goes for the shafts on the gears. More sides might make them smoother. However, I think the overall reason is just that the tolerances are too tight. The white one runs better than the red one, and this is probably because it doesn't have a rigid base, allowing the outer ring to deform slightly and adjust to the gears as they rotate.

Planetary gears 2

Thingiverse contributor emmett has created some nice designs, one of which is a planetary gear bearing in which the gears have angled teeth. It prints as a single piece. You can customize it in many ways: diameter, thickness, number of gears, number of teeth and so on. One parameter is the tolerance, or clearance between the elements. I printed one of these at the recommended tolerance of 0.15 and it came out as completely solid and unusable. With tolerances of 0.6 and 0.7, it works. It's quite loose at these tolerances, so 0.5 or even 0.4 might work better. These are 10mm tall, instead of the recommended 15mm, to reduce the printing time a little. After a bit of working in, they move quite smoothly, though the large tolerance and reduced height means the gears tend to slip out of alignment and jam.

Cube gears
Emmett is also the originator of the cube gear. There's various customizations possible. I made one with a 5:6 ratio, and one with 1:2 ratio. I shrank the second one to half size in Slic3r. Here they are:

I was using these for experimenting with different filaments, hence the range of colours. The large one worked well. It can be a little stiff, but once it starts turning, it goes smoothly. Some people recommend lubricating it with graphite or oil, and this would probably help. The smaller one does turn but is loose. The problem here is that it is held together with small plastic pins, and at this size, the reaction force from turning the gears pushes some of the pins out of their sockets. The more salient point is that both worked without changes to the clearance.

I don't have an overall conclusion, except to say something similar about the static parts experiment. You need clearance, but may need to experiment to find the right value.

Finally, here's a video. (Sorry for the poor quality. I don't have a good way of holding my phone steady. Now if only I could print something to do that....)

NinjaFlex

NinjaFlex is a very flexible, rubbery filament. I have tried to print with it in the past and either failed completely or got very poor print quality. When it's failed, it's been because it wouldn't go through the extruder at all. Here's an example of one of the poor quality prints:

(Sorry for thelack of definition in this and the later pictures. Photographing black objects isn't easy.)

I've recently managed to get some decent prints with NinjaFlex. There are a couple of changes that seems to have made the difference. First, I've replaced the original MK7 extruder that came with my printer with a MK9. I'm not sure if these are standard terms for the types of extruder, so here are before and after pictures to explain what I mean:

The essential difference is that the MK7 has a fixed mount for the grooved bearing, while the MK9 has a spring loaded one.

The other change I made was to turn off retraction altogether in the Slic3r setting. I suspect this made more difference than changing the extruder. I print quite slowly (20 to 30 mm/s), and I've found that sometimes it helps to print the nozzle by bring it up to temperature and manually extruding about 50mm of filament before starting the print. This might be needed because it looks like I get a lot of retraction right at the end of a print - probably there is a Slic3r setting I've missed. I use 225C on a 45C bed. Here are some example prints:

The item sitting next to Cthulhu is either a 10mm cube or an odious oblong box of disturbing size. Here's a thin-walled cube and a part of a cube gear.

NinjaFlex is not an ideal wrong material for printing gears, and I just wanted to see what would happen on an object with some simple overhangs. Everything was printed at 0.2mm apart from the gear, which is at 0.1mm. Overall, these are nice results, and tells me that if I have a need to use NinjaFlex in future, I can get it to work.

(Well.... mostly. After writing this, my next two attempts to print with NinjaFlex ended with the filament failing to extrude, as it was buckling in the E3D. This might mean I need to lower the extruder torque as well.)

Filamentary 2

A while back I wrote about some different filaments I had been trying out. Since then I've continued to test them and others, and I have several more lined up to try soon. I've created a document (here) which gives a brief description of them: where I got the filament, the temperatures that worked, some subjective impressions, and a few pictures. It's my intention to update this as I try more samples, so check back from time to time.

Filaments included so far:

• 3dom Buzzed
• 3dom Wound up
• Biofila linen
• Biome3D
• ColorFabb PLA/PHA
• Fillamentum Timberfill
• GizmoDorks PETG
• Ingeo PLA
• JET PLA
• Justpla PLA
• Laywoo-D3 flex
• Meltink PLA
• Porolay gel-lay
• Porolay lay-felt
• ProtoPasta stainless steel PLA
• Quantum 3D PLA
• Ninjaflex
• Ninjaflex Semiflex
• Reprapper Tech colour change PLA
• Sainsmart wood
• Taulman PCTPE
• Taulman t-glase

Check back from time to time, as I'll add more. Last update: 22 Nov 2015.

Printing mechanical parts, part 1

Many of the mechanical and constructional parts that I have printed have not come out right. In particular, holes for screws and the like have been too tight, and I've needed to change the model or the printer settings to get them right. In this post, I'll describe some experiments on how different parameters affect this. This is all to do with static parts - can you get a bolt on a hole or onto a thread, for example. I aim to return to the subject in future and look at moving parts such as gears. I have been somewhat systematic, though this is far from being a scientific exploration.

Warning: this post is boring. Sorry, that's just how it is.

The areas I want to explore are:

• given a bolt of some diameter, how large does a circular hole have to be for it?
• given a hex nut of some exterior size, how large does a hole need to be for it?
• given a nut designed to fit on a bolt of some diameter, what size thread will it fit on?

I did most of the tests using a M3 bolt and nut. I made an cuboid of about 3mm thickness, and put several circular holes and hexagonal holes in it. To get the base sizes right, I downloaded CAD models of bolts and hex nuts from McMaster-Carr's web site. I then scaled these and used Boolean modifiers in blender to either cut holes in the cuboid or merge them with it. The result looks like this:

The four extra cylindrical shafts are for another test that I decided not to follow through with.

The base size for the circular holes was M3 (3mm diameter), represented in blender as a cylinder with 32 sides. The other three holes used this scaled up by a 1.1, 1.2 and 1.3. The hex holes were made by using a 6 sided cylinder in blender, set to the same size as the CAD model of a hex nut. It was scaled by 1.05, 1.1 and 1.15. Note that I didn't expect anything to fit in the base size (scale 1) holes. They are just for reference. The bolts were scaled down by 0.95, 0.9 and 0.85. I picked all of these scale factors after some initial prototyping.

After printing, I tested the circular holes by inserting a M3 bolt into them.

Circular hole ratings:

• Rating A means it won't fit.
• Rating B means it fits but only if you actively screw it in.
• Rating C means it fits without needing to screw it in.
• Rating D means it fits easily.

The difference between C and D is a bit subjective. For most things that I make, I would want a C or D.

I tested the hex holes by fitting a hex nut into them.

Hex nut hole ratings:

• Rating I means it won't fit.
• Rating J means it fits tightly. The nut wedges firmly into the hole.
• Rating K means it fits loosely. The nut does not stay in the hole.

I would want J or K, depending on the application.

For testing the screw thread, I tried threading a hex nut onto it.

Thread ratings:

• Rating P means it doesn't fit.
• Rating Q means the nut fits on the thread, but jams so you can't turn it far.
• Rating R means the nut engages with the thread.
• Rating S means the nut is loose on the thread.

Rating R is what we want.

Most of the tests were done with PLA from justpla. It's a cheap brand but has worked OK.

In the test results, I will quote the results of the experiments as the rating for each of the four circular holes, the four hex holes and the four threads, with the desirable values in bold.

Experiment 1: layer thickness

The layer thickness could make a difference. I would not expect to see much effect from this except possibly on the thread test, where the vertical resolution matters. The prints were at 100% infill.

Layer thickness	Circular hole	Hex hole	Thread
0.2mm	ABCC	IIJK	PPRS
0.3mm	ABCC	IIIK	PQQR
0.1mm	ABCC	IIKK	PQRS

There is some variation with layer thickness, but not very much.

Experiment 2: infill

For all the remaining tests, I used 0.2mm 100% infill, the first line on the table for experiment 1. I did one test with less infill. It's conceivable that the pressure from the 100% filled body could press in and shrink the holes. So here are the results with 30% infill.

	Circular hole	Hex hole	Thread
30% infill	ACCD	IIJK	PPRS

Infill makes a difference to the first test, with no effect on the others.

Experiment 3: extrusion multiplier

I have read on the RepRap forums that changing the extrusion multiplier in the slicer settings is a common way of overcoming holes that are too small. For this test, I reduced it to 0.85.

	Circular hole	Hex hole	Thread
0.85 ext.mult.	ACCD	IJJK	PRSS

This makes a small difference for both hole types. The "S" ratings for the thread means that the nut didn't engage with the thread in the slightest, more extreme than any of the other tests.

Experiment 4: slicer

I usually use Slic3r, and for one test I tried Cura. The results of this test shouldn't be taken too seriously, as I have not tuned the Cura settings. There are no results for the thread test, as the print had no usable results.

	Circular hole	Hex hole	Thread
Cura	ACDD	IIJJ

Experiment 5: other materials

Lastly, I tried some other materials I had lying around: a PLA/PHA hybrid from Colorfabb, PETG, ABS and a different supplier of PLA. The ABS results are partial. I rarely print with it, and the piece curled during printing. There are no thread results from the ABS test.

Material	Circular hole	Hex hole	Thread
PLA/PHA	ABCD	IIKK	PQRS
PETG	ABCD	IIJK	PPRS
Quantum 3D PLA	ABCD	IIKK	PPRS
ABS	ABCD	IIJK

Experiments I did not run: print speed, nozzle diameter, temperature. All of these would be interesting to explore. Print speed, especially - the other two are likely to be constrained for other reasons.

It's hard to conclude much from this, and I think that is interesting in itself. I expected more variation with material, as different plastics allegedly shrink by different amounts. The infill and extrusion multiplier are the most important ones, as you have more opportunity to control them.

Recommendations when you are designing an object

Suppose you are designing an object. What should you do? These results suggest for interior holes you should make them a bit bigger than the nominal size. A good general factor is 1.2 for circular holes, and 1.1 for tight fitting hex nut holes or 1.15 for loose ones. For threads, scale them down to 0.9 of the nominal size.

Recommendations when you are printing someone else's designs

If you are printing someone else's design and the holes are too tight, you can reduce the infill or the extrusion multiplier. The second is better and gives you more control, but it may make threads unusable.

Do the results generalize to other sizes?

All of the tests so far were based on M3 parts. What about M4, as an example of another size? There are two possibilities here: you could scale pieces by the same factor I used above, or you could change their size by the same absolute amount. Here are results (with the M3 result repeated for convenience). Due to an error, I am missing the thread results for the absolute case.

	Circular hole	Hex hole	Thread
M3	ABCC	IIJK	PPRS
M4 scaled	ABDD	IJKK	PPRS
M4 absolute	ABDD	IJKK

The "S" on the M4 scaled line was very, very loose. So this implies both the scaling and the absolute change have a much larger effect than with M3. This is annoying, as it means if you have a design, you might need to make different adjustments for different size holes.