Written by Graham

Spindle motors.

This is a comparison of what spindle motors are available for small routers, based on the SainSmart Genmitsu ‘3018’ family which for my purposes includes the 3020-PRO MAX.

I have tested the stock motors for my routers (3018-PRO, 3018-PROVer, and 3020-PRO MAX) plus the 20K GS-775M motor and the 44mm 24V brushless motor. All motors are from SainSmart.

Physical differences

The motor size must match the motor mount on your router. You can’t put a 52mm diameter motor in a 44mm diameter hole and you also cannot secure a 44mm diameter motor in a 52mm diameter hole! Height also matters, well it’s more about where you can place the motor in the mounts without obscuring any cooling slots. This will affect how close to the bed the tip of the collet and so the tip of the bit can be placed.

Weight can be important, the more power the heavier the motor and so the more mass that has to be pulled and pushed by your stepper motors. The rigidity of the router has been designed for a specific motor mass, add in a much heavier motor and it may not support it. This can also affect the acceleration values you can safely use in Grbl.

Voltage and Control

The motor you use will be rated to take a specific voltage, the standard 775 motors are normally 24V, the more powerful motors such as the one on the 3020PROMAX run at 48V.

The speed control can be a PWM signal or Manual where the motor speed is set by something other than the Grbl controller such as a knob or slider. This means you will have to set the motor speed independently, the speed you set in your CAM software will be ignored.

All motors in this comparison have the speed controlled by the Grbl Motherboard on the router.

Measurements

The testing device is a motor monitor of my own design, basically, it measures the Motor RPM, supply voltage and current, and the Target RPM (the requested spindle speed from the Sxxxx value in the Gcode). It will either display the current readings or log the data to the serial port. A 25ms logging interval has been used here (50 readings per second) 

• RPM is measured by a sensor reading the time taken for each revolution of the spindle.
• Target RPM is derived by reading and decoding the PWM signal from the router, a low target RPM with a 980Hz PWM signal are at the limits of what is possible for the motor monitor to measure. It can only time pulses to a resolution of 4µ Grbl boards have difficulty producing a low PWM pulse, the internal PWM timer is also used for other purposes which seem to limit it to a minimum of a ~16µs pulse rather than 4µs.
• Volts are read by reading the relative voltage between the Motor + and – connections.
• Amps are read by a sensor on the motor supply.

(For the brushless motor the current and voltage is the supply to the ESC Control board).

While efforts have been made to remove spikes and variations in the readings there is a balance between removing spurious readings and removing actual ones.

Noise measurements in dB are made by an app on my phone, it’s not calibrated so only take the values as a comparison.

Tests

The following tests have been run on each motor. All are under no load for the simple reason that they are easier to make and don’t introduce so many variables such as Bit, Feed Rate, Depth of Cut, material……

For consistency, each motor (apart from the 3020PROMAX motor which takes 48V rather than the 24V on the Pro and Prover) has been tested using a standard SainSmart 3018 Pro motherboard and a 24V power supply which can handle the 20K GS775M Motor. (I believe these are now included as standard with the SainSmart 3018Pro and PROVer)

All the results are presented in the form of charts, these are designed to show the motor characteristics as well as allow a comparison between them.

NOTES:

• The charts are not identical! The axis scales can be different, there is no point in making the voltage scale 48V for a 24V motor.  
• The G-codes used are slightly different for each motor, apart from the MaxRPM test, if nothing else the Max speed ($30) in all the other tests has been changed, delays may have been changed, and for example, in the Min RPM test the step will be different to fully explore the characteristics of each motor
• Before running any test on a new board the motor monitor was re-calibrated.

Max Speed

The Gcode starts with the spindle stopped, sets the speed to the maximum, waits for the spindle to reach maximum speed then sets the speed to zero. This shows the maximum speed the motor can achieve, The maximum voltage and current draw as the motor starts up, and The way the motor speed returns to zero.

After running this test, the approximate maximum motor speed measured is used in the following tests.

Ramp Start

The Gcode starts with the spindle stopped, ramps up the spindle speed to the maximum using a number of steps with a delay in between each RPM step. Then set the speed to zero and the chart stops, the motor slowdown is not shown.

Some motors when started from rest to maximum speed can exceed the current draw that the power supply and router motherboard can safely supply, this shows the difference in voltage and current the motor will draw when a ramped start-up is used.

Linearity

This is designed to show the correspondence between the Sxxxx value in the Gcode and the actual speed of the motor. The motors are not linear in their characteristics. It starts with the motor at rest and increases the speed in a number of steps with a delay to let the motor reach the speed up to its maximum.

Min RPM

This is designed to determine the minimum speed that the motor will rotate. The Gcode starts the motor from rest in small steps with a delay in between each step to see when it starts rotating.

Volume

This is a very rough measure of the noise level of the motor, this is measured during the max speed test and only the approximate maximum value is shown, this is only in the form of a table

Test Results

The lines on the charts can include:

• RPM The measured RPM of the motor
• Target RPM The requested RPM of the motor decoded from the PWM signal
• Target RPM (by Time) The requested RPM based on the Gcode and timings
• Volts The voltage measured across the motor terminals
• Amps The current measured flowing through the motor

These may also be an average of some readings to remove noise (Avg ..) or the maximum value of some readings (Max ..)

Max Start

Two sets of charts are shown, the first set shows the RPM and Target RPM with the motor Current, the second set is the same but the motor Voltage is shown. This is only due to the different scales required to show more detail.

Ramp Start

Only the RPM and Target RPM with the motor current are shown, the voltage does not add any real information.

Minimum RPM

Shows the minimum RPM the motor will operate at. Low RPMs can be useful for engraving things like plastics which have a tendency to melt if the friction from the bit is too high.

NOTE: The Target PWM output is noisy at low speeds, this is due two factors; the low speeds are at the limit of what my motor monitor can measure and the PWM output of the board can be affected by the use of the PWM timer being shared with some other functions of the Grbl board. The Target RPM (by time) line is determined by the Gcode, set the speed to increase by a 100RPM step with a delay of 0.25 seconds between changes means that 250ms will elapse between speed changes.

Linearity

Shows how the Sxx value in the Gcode (target RPM) translates into the actual motor RPMs.

Volume

All values are very approximate in dB.

Summary

^* Including ER11 mount, collet and nut.

^** Motor body.

^*** The Brushless Motor can be slow to start. It will rotate slower but there is a significant delay before it will start to rotate.

By Graham Bland. I can be found on https://www.facebook.com/groups/SainSmart.GenmitsuCNC/