Choosing an extruder drive
Selecting an appropriate extruder stepper motor and gearing for the best performance is a compromise between weight, torque, resolution and acceleration.
Overview
An extruder drive should have sufficient torque to push the filament through the nozzle at the printing speed you want to use, and sufficient resolution so that individual extruder microsteps are not visible in the print. When there is a Bowden tube between the extruder drive and the hot end, you also need sufficient acceleration to retract the filament fast quickly enough to avoid blobs.
You can calculate the force that an extruder drive can provide before it skips steps if you know its steps/mm and the characteristics of the stepper motor it uses. If using a 1.8deg/step motor, the extruder force in Newtons (N) is:
Extruder_force_at_rated_current = Motor_holding_torque * Extruder_steps_per_mm * 0.0014
For a 0.9deg/step motor:
Extruder_force_at_rated_current = Motor_holding_torque * Extruder_steps_per_mm * 0.0007
where the extruder steps/mm is specified at 16x microstepping, and the motor holding torque is specified in Ncm with both phases energised at the rated current. The steps/mm take into account the gearing (if any) and the diameter of the hobbed shaft.
For 1.75mm PLA filament extruded through a 0.4mm or 0.4mm nozzle, aim for extruder force in the range 10N to 25N at rated motor current, and aim to run at 50% to 85% of rated motor current. For 3mm filament, aim for 3 times the extruder force, so between 30N and 75N at rated current.
If you have too little extruder force, the extruder may not be able to push the filament through the nozzle except at excessively low print speeds or at high print temperatures (which increases stringing). The problem with too much extruder force is that if the nozzle becomes temporarily obstructed, then you want the extruder to skip steps. If instead it grinds an indentation into the filament, then extrusion won't restart when the obstruction is removed.
Un-geared extruders
These typically have steps/mm of about 100 using 1.8/deg motors @ x16 microstepping. To reach 10N force, you need a motor with 71Ncm force. Very few Nema 17 motors have this amount of torque, so you will probably have to make do with 60Ncm, giving a maximum 8.4N force.
The low steps/mm makes it more likely that individual steps will be visible in the extruded filament. With 1.75mm filament and a 0.4mm diameter nozzle, 100 steps/mm going in gives about 5 steps/mm in the extruded filament. You can use a 0.9deg motor to double the steps/mm.
An un-geared extruder design can also be used with a stepper motor with an integral planetary gearbox of about 5:1 ratio.
Geared extruders
As well as having gearing, a geared extruder normally has a hobbed shaft with a smaller diameter than an un-geared extruder, because the hobbed insert doesn't have to fit over the stepper motor shaft. Typical steps/mm around 420 with 3:1 gearing, and 650 with 5:1 gearing. This means that you need only about 17Ncm of motor torque to exceed the target 10N force (for 1.75mm filament) with 3:1 gearing, or 11Ncm using 5:1 gearing.
Beware of using motors that have many times too much torque. You can reduce the motor current to reduce the force, however high torque motors also have high rotor inertia, so by reducing the current you also reduce the available acceleration.
Remote direct drive extruders
These typically have gearing of 30:1 to 40:1 and steps/mm of around 4200.
4 Comments
What is the constant on the end of the formula exactly? How is it derived?
Daniel Callander (Knifa) - Reply
Thank you, I have corrected that formula.
David Crocker - Reply
Is there a typo in the equation? should the 0.9 calculation factor be 0.0007 instead of 0.007?
Christopher Tilley - Reply
Is it possible that there is a typo in the formula for
for 0.9deg/step motors? I assume the multiplier at the end is missing another zero.
When I calculate this for a motor with equal holding torque once for 1.8deg/step and once for 0.9deg/step I get an order of magnitude more force for the 0.9deg/step motor. That seams unlikely.
Manuel Coenen - Reply