Iterating on the "boat" for the XY gantry. Got an E3D Aero.

3D Printing
, , , ,
1

Iterating on the “boat” for the XY gantry.

Got an E3D Aero. Once assembled, the first thing was to print a test mount, to find the center-of-mass … by hanging the assembly from a bit of string. :slight_smile:

The result was a bit simpler than I expected. The center is just within the front-center of the motor. (Are stepper motors more dense in the front?)

Note I have a pancake motor on order from E3D. Is the heavier motor needed for higher print speeds? Is the smaller motor adequate? Would the smaller motor with a bigger gear be adequate? Dunno.

The notion is to as far as practical to align forces. If the belts pulling the print-head are pulling “through” the center-of-mass, the print-head will not want to yaw or pitch - which could mean rigid with lighter structure.

So I turned the print-head “sideways”. :slight_smile:

If the nozzle is also near the X-axis (the E3D nozzle is slightly offset) then torsional load on the XY plane is less. (There is still torsion in Z.)

The problem with this approach is that the “boat” takes up space in X, limits the range of travel, and thus the X dimension of the print area. Almost certainly a bad trade-off for slower printers, but perhaps worthwhile at higher speeds.(Arguably, for the same-sized frame, gain a bit in Y.)

The “boat” is sized for the print-head assembly. A bigger print-head means you need to print a bigger boat. Suspect this is a non-problem.

Ended up splitting the boat into inner and outer frames - not as I expect this to be the final design, but to support iteration. Do not want to reassemble the nice E3D hardware every time I revise the boat. :slight_smile:

As a first iteration, have the boat riding on square tube. Riding on round tube under torsional load, would expect the assembly to want to pitch and yaw. On square tube, perhaps less?

The above is a bunch of theories … that need test.

Need to find some square tube of appropriate size. Something cheap and easy for a possibly-throw-away iteration. Then fabricate the slider/ends to match.

The fit between the print-head and top of frame diagonals is … complex. Could tweak this to gain more print area - but not for this build.

Given that the motor is a pretty solid hunk of metal, running structure around the motor is a waste. Tried to unscrew the back of a stepper … but mine really do not want to come apart. Later.

57c5c23657cc3917c67b1cdbca2a53a9.jpeg
57c5c23657cc3917c67b1cdbca2a53a9.jpeg4048×3036 4.55 MB

8ca05b7aeffff0c1e40b682a58db5762.jpeg
8ca05b7aeffff0c1e40b682a58db5762.jpeg4048×3036 4.07 MB

f2804433e686e0f784701f7bed5d40ad.jpeg
f2804433e686e0f784701f7bed5d40ad.jpeg3036×4048 2.87 MB

b1b70c7e066f29e3bfd1f079ff6e3906.jpeg
b1b70c7e066f29e3bfd1f079ff6e3906.jpeg3036×4048 3.08 MB

44387fda9e566598ca29a0f7ac1cbed6.jpeg
44387fda9e566598ca29a0f7ac1cbed6.jpeg4048×3036 2.83 MB

2

Shouldn’t be any reason you can’t take the back of the stepper off. Although fair warning that the permanent magnets will get a little weaker if you actually remove the rotor from the stator. That will make the motor produce a bit less torque, while shifting the torque curve marginally faster. (Probably won’t cause a noticeable change in performance but it’s a real effect.)

3

That pancake motor on a geared extruder is plenty of power for 1.75mm material at least. If it chokes at higher print speeds, it’s probably more because the plastic isn’t sufficiently melted anyway.

4

I use a pancake on a Titan/Aero direct drive. Works very well.

5

@Matt_Shepherd At what sort of print speeds?

6

@Preston_Bannister Do you mean, what sort of extruder speeds? Print speed doesn’t directly correlate to extruder capability because of different extrusion width and layer height. I can print at 200mm/s if I do 0.4mm wide by 0.1mm layers, or 40mm/s with 0.8mm wide by 0.25mm layers, and that’s the same extruder load.

7

@Ryan_Carlyle Well yes, that would be the follow-up question if higher print speeds are reported.

What you are talking about is the volume of plastic extruded, and I generally agree. Though I suspect the viscosity of the plastic would present a greater load to the extruder with thin layers.

8

@Preston_Bannister thinner layers doesn’t seem to make much difference. The fact that you’re usually printing less volume flow rate means the exit temp is higher and thus there’s less viscosity than a thicker strand at higher flow rate. But wider strands than the nozzle orifice do add a LOT more back-pressure – because the plastic starts cooling as soon as it exits the nozzle and hits the cold previous layer. Pushing the extra plastic out sideways as it cools adds a ton of viscous drag. You can see that if you compare printing strands 2x nozzle size at half speed (= same flow rate) with a Bowden extruder – you get a lot more hysteresis because the extrusion force is higher.

9

@Ryan_Carlyle I believe we are saying basically the same thing.

For a constant volume of plastic and print speed, thinner layers yield more viscous drag, which presents greater load to the extruder.

How much that loads the extruder is going to depend on the viscosity of the plastic at the temperature when it leaves the nozzle, the rate of cooling, and the effect of cooling on viscosity.

I am not wanting to become a materials scientist here. I am looking for effects I might want to measure (at some point). :slight_smile:

10

“For a constant volume of plastic and print speed, thinner layers yield more viscous drag, which presents greater load to the extruder.”

Disagree. I don’t think layer height will make a meaningful difference. This isn’t a boundary layer flow /plate-plate shear process where fluid thickness is critical. At this speed and scale the molten filament is more “plastic” than fluid. All the filament takes an abrupt right turn when it exits the extruder. Same deformation + same rate = same energy.

11

As it turns out, ordering square aluminum or carbon fiber tubing of appropriate size is a bit of a PITA. Easier to order carbon fiber round tube (hijacking the drone market). Not exactly what I wanted.

Settled on one of:

Arris (which I ordered):

Alston (no reviews):

Firgelli Automations (more expense, but better focus?):

Dimensional accuracy is a big concern. There might be better alternatives that I missed.

12

What about using aluminum extrusion like square tube?

13

@Ryan_Carlyle That was what I tried first. :slight_smile:

Always possible I missed something.

14

@Ryan_Carlyle Sorry, missed this comment of yours, prior.

For thinner layer height, the amount of heat transfer between the newly extruded hot plastic, and the older layer of “cold” plastic will be greater. Will it be great enough to make a difference in performance?

Dunno. Certainly some. Whether dominant or minor, is a question.

15

@Preston_Bannister Thinner layers allow the fresh material to cool solid faster with less airflow, yeah. Makes a huge difference to print quality when you have weak part-cooling, but I don’t think it’ll make much difference in nozzle back-pressure or viscous drag or anything like that.

16

This is the sort of stuff that would be great to get some empirical data, but the effects are likely to be so small that it’s difficult to imagine how to measure them.

17

@Ryan_Carlyle We are kind-of saying the same thing. :slight_smile:

As we push this technology to higher levels of performance, we are going to have to understand much more dynamic measures.

You might be right. The back-pressure due to cooling plastic might not be significant. You might also be wrong. The back-pressure might prove a decent measure.

As I am still building the basic physical printer, I would like to delay more advanced discussion. :slight_smile: