Why Spindle driven Delta's are more precise than Belt Driven

Quote
evandene
All,
Traveling from A B with a stepper driven system Works well if no speed or accurate positioning is required.
Please do following test with your delta printer;
1) go from A to B in Z height like 25 mm (1") and measure the real distance. Most probably all it's within a few hundreds of a mm.
2) do this again but now Z= 0.1mm (0.00393"). Asuming you use micro stepping of 1/16 or 1/32 you will notice that sometime the requested positioning is slipped and sometimes double or more executed. This is normal behavior for stepper systems.
See Wikipedia for steppers, 5th paragraph.
I can't have that in my system.

OK, so I just tried that. My delta has 160 steps/mm at x16 microstepping, so the theoretical Z resolution (which is worse than the X and Y resolutions) is 0.00625mm per microstep. If I command the head up and down by 0.00625mm at a time, I can hear it moving each time I send the command. Furthermore, if I set the height so that the IR sensor is on the point of triggering, I can turn it on and off by directing the head up and down 0.00625mm.

I was half expecting to have to increase motor current to get it moving on every step, but that turned out not to be necessary. They are 1.67A motors and I am running them at 1A.

Do you really think that achieving a significantly better resolution than 10um will make any observable difference to print quality? There may well be other applications where higher resolution is helpful, but I don't think 3D printing is one of them - whereas high speed definitely is an advantage, up to a point.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

....and there is filament, I've seen 1.75mm filament with a variation of anywhere between 1.6-1.9mm, this means for a nominal 0.3mm bead at a fixed speed you can have a size variation from 0.274-0.326mm


if you really cant have this inaccuracy in your system why? and more to the point....your doing research on Wikipedia!?!?

---

RepRapPro Mendel 3 Tricolour
RepRapPro Fisher
-Carbon Arms
-Easy adjust Carriage+effector
-axis stiffness mods
HE3D -600 delta
-Duet 0.8.5
-PanelDue
-DC42 Height probe
-RobotDigg metal components
Simplyfy3D
RS Design Spark CAD

My experience indicates that filament diameter on one spool rarely changes more than ±0.02 mm. Most of the time it is somewhere around 1.71 mm. I check filament diameter before each print. I buy filaments in the price range of about 22 € per kilogram. So filament diameter error is not so bad. Maybe the cheaper filaments are worse ... or I was just lucky.

Here's a leadscrew driven large delta.
[www.youtube.com]

Looks nice and they are using a direct drive extruder on the effector. But I am not sure it's going to be very fast.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

I had some talks with people and got inspired to look for servomotors and to keep compatibility high a converter to take the stepper signal and transform in to pwm for the servomotors.

-

Quote
sungod3k
I had some talks with people and got inspired to look for servomotors and to keep compatibility high a converter to take the stepper signal and transform in to pwm for the servomotors.

You cant just convert Stepper pulses to PWM,that's not how servo drives work.

The difference between a motor/stepper motor and servo drive is a motor you just send a speed (voltage, or PWM signal), and a stepper motor you send a pulse that represents a displacement (ie 1 step = 1.8 degrees and fractions thereof) with a servo drive you send target position and speed, and there is a position feedback system that tells the servo drive where the motor currently is (ie a pulse encoder, potentiometer, or linear encoder)

looking at a single axis move from point a-point B....

- a simple motor drive will have to have limit switches to control the move (timers possible,but frankly....NO) the controller will send a PWM signal that represents the speed and when the endstop is hit it stops, without any slowing down.

- a stepper drive looks at the length of the move and the steps per mm factor and pulses steps at a frequency that dictates the speed

- a servo drive looks at the current position of the axis to determine an out of position error, and the current speed of the motor and then outputs a speed to correct the position, this function works constantly so if you 'manually move' the axis then it will automatically correct itself (actually you wont be able to move it).

So, unless you plan on using a controller that is specifically set up for using servo drives, ie one that outputs position and speed values, you will have to have a device that can read the step outputs and integrate them to achieve position data (actually not so hard), but the problem is speed, the converter will only be able to deduce speed from the frequency of the pulses, and that's actually 1 pulse too late, and you probably will have to still work at 1 pulse resolution.

.....what do I mean by the pulse being too late, well consider the axis that has been stood still for a hour and you send it just one pulse to move the smallest increment.... how fast should it do that, since the last pulse was an hour ago does that mean that it should be moving at 1 pulse per hour?

---

RepRapPro Mendel 3 Tricolour
RepRapPro Fisher
-Carbon Arms
-Easy adjust Carriage+effector
-axis stiffness mods
HE3D -600 delta
-Duet 0.8.5
-PanelDue
-DC42 Height probe
-RobotDigg metal components
Simplyfy3D
RS Design Spark CAD

Quote
DjDemonD
Here's a leadscrew driven large delta.
[www.youtube.com]

Looks nice and they are using a direct drive extruder on the effector. But I am not sure it's going to be very fast.

You know looking at the pitch of those screws, then I reckon they can get some good speed out of that delta, however, whats driving those screws? are they driven directly?, is there a gear drive?, are they using stepper motors?, what controllers have they got?, and linking back to the OPs point, what is their maximum resolution?.

---

RepRapPro Mendel 3 Tricolour
RepRapPro Fisher
-Carbon Arms
-Easy adjust Carriage+effector
-axis stiffness mods
HE3D -600 delta
-Duet 0.8.5
-PanelDue
-DC42 Height probe
-RobotDigg metal components
Simplyfy3D
RS Design Spark CAD

Kuhling & Kuhling is using a RADDS controler with 1/128 micro stepper driver in their DELTA printer able to run super high frequencies.
The pitch of the lead screws is about 20 mm (close to an inch), no need for a gearbox. NEMA 23 steppers will do the job.

Quote
evandene
Kuhling & Kuhling is using a RADDS controler with 1/128 micro stepper driver in their DELTA printer able to run super high frequencies.
The pitch of the lead screws is about 20 mm (close to an inch), no need for a gearbox. NEMA 23 steppers will do the job.

Gear drives are not only about torque you can also use them to finesse steps per mm and speed

---

RepRapPro Mendel 3 Tricolour
RepRapPro Fisher
-Carbon Arms
-Easy adjust Carriage+effector
-axis stiffness mods
HE3D -600 delta
-Duet 0.8.5
-PanelDue
-DC42 Height probe
-RobotDigg metal components
Simplyfy3D
RS Design Spark CAD

Kuhling & Kuhling know how to avoid using gearboxes because of cost and the always under-estimated gearbox efficiency and ever increasing backlash.
When someone can play with the lead-screw pitch, there is no reason for adding costs for a gearbox.

Worm-Wormwheel gearboxes have an efficiency from 45% to 65% (expensive ones) because of the high internal friction.
Planetary gearboxes go from 70% to +90% (super expensive) (stamped sheet-metal gear or hardened and grind gears)
Straight gearboxes go from 65% to max 85% (stamped sheet-metal gear or hardened and grind gears)
Serious supplier do mention the true and guaranteed efficiency and backlash in their spec.

Quote
evandene
Kuhling & Kuhling know how to avoid using gearboxes because of cost and the always under-estimated gearbox efficiency and ever increasing backlash.
When someone can play with the lead-screw pitch, there is no reason for adding costs for a gearbox.

Worm-Wormwheel gearboxes have an efficiency from 45% to 65% (expensive ones) because of the high internal friction.
Planetary gearboxes go from 70% to +90% (super expensive) (stamped sheet-metal gear or hardened and grind gears)
Straight gearboxes go from 65% to max 85% (stamped sheet-metal gear or hardened and grind gears)
Serious supplier do mention the true and guaranteed efficiency and backlash in their spec.

I never said 'gear box', and gear boxes are not the only form of gearing, anyway, I said I don't know, and they could just have a slow machine

---

RepRapPro Mendel 3 Tricolour
RepRapPro Fisher
-Carbon Arms
-Easy adjust Carriage+effector
-axis stiffness mods
HE3D -600 delta
-Duet 0.8.5
-PanelDue
-DC42 Height probe
-RobotDigg metal components
Simplyfy3D
RS Design Spark CAD

Gearing, gearbox, whatever drive based on two or more involute teethed disks or similar do have an efficiency and they do have backlash.
The more precise machined and the more precise assembled, the higher the efficiency of a drive system.
For example a 3D printed gearing system like sometime used in extruders, do have an efficiency of maybe 40 to 60% due to geometric tolerances and assembly tolerances. Often we can get away with it and that's okay, nothing wrong with it; it's just good to know.
High efficiency gearboxes or otherwise high efficient gearing's are most of the time to costly for the DIY market; at least for me they are.

So a gearbox or gearing system combined with a Ball-Screw spindle is not logic to me. If you like to travel 10mm /rev you take a lead-screw (Ball-Screw Spindle) with a pitch of 10mm; if you like to have 20mm/rev you take another lead-screw.
Preferable to pick a ball-screw spindle because they can be ordered pre-loaded in order to eliminate backlash between Ball-Screw and Ball-Screw Nut.
For me, Trapezoid lead-screws are not an option for a Delta printer although some printers do have this option.
If more torque is needed, it's better to move to a NEMA 23 size hybrid stepper motor or to a servo system.