My TMC236 does not work properly at higher current settings. What can I do?

This often is a problem with the TMC236 detecting a short circuit condition. Try to monitor this in the SPI serial word coming from the TMC236, e.g. using an oscilloscope on the MC236's SDO line.

The typical causes for wrong detection of a short condition are:

  1. The shunt resistor on the high side has got to long/thin PCB traces. These traces can easily raise the value by several 100mOhms.
    • Use thick, short and straight traces.
    • Make sure, that your sense resistor traces add no substantial resistance to the high side sense resistor.
    • Quick workaround: Use lower value for high side shunt or add a voltage divider for VT.
  2. During switching of the coil outputs of the TMC236 voltage spikes occur on the sense resistor. This can be caused by long/thin PCB traces from GND to the sense resistor or from the sense resistor to BRA/BRB. There especially should be no parasitic inductivities in these traces.
    • Use thick, short and straight traces
    • a massive grounding is preferable.
    • Make sure, that your sense resistor traces add no substantial resistance to the sense resistors.
    • Quick workaround: Increase blank time or add RC-filtering for SRA and SRB.

What can be the problem when I observe reduced motor reliability at lower supply voltages?
This can very well be coupled to the above points. Generally the motor current in fact is independent of the supply voltage, and the TMC236 with its low internal resistance is a very good driver for lower voltages. But one has to consider, that a lower voltage means that the current needs a longer time to reach the target level. This means, that at lower voltage level the maximum motor velocity may have to be reduced. You can very well monitor this condition watching the open load flags of the TMC236: If they flicker during the motion, the motor velocity might be too high with regard to the actual supply voltage.

My circuit uses a TMC236 driver. Do I have to make any changes to use a TMC246?
If you use the ENN pin for switching off the driver, make sure that the ENN pin is pulled up externally to at least 2.8V for circuits supplied with 5V. Do not use an open collector output to drive it. Allow for a small increase in standby current on the VCC supply pins, because VCC feeds the ENN pin overvoltage protection comparator.

My application needs very high motor velocities. How can I get the maximum RPM?

Full stepping is a good solution for high RPM applications! Why? In general, RPM is just limited by the inductivity of the motor and the available supply voltage. So the answer is, use a motor with low inductivity, i.e. high coil current, and maximize supply voltage. Microstepping does not have a major influence on motor performance at high RPM. As compared to microstepping, fullstepping control allows to get maximum current into the coils. So, when exceeding some velocity, you might reprogram your microstepping table and change into fullstepping. When you use the TMC428, you can realize this on-the-fly, by modifying the driver configuration table to all phase bits constant, or by modifying the sine wave table to a fullstepping table. An important DO NOT for reaching high RPM: Do not set the motor current too high. To keep motor resonance low, it is important that the programmed maximum current can be reached by the application. So you will typically reach a higher velocity, with a lower current setting, when you are at the limits of the motor / supply voltage. To bes sure: Check with the oscilloscope at the sense resistor, that the motor driver starts chopping for at least a small part of each (full)step.

The motor makes noise during stand still. What can I do to get it silent?

These are some hints for conditions, which can (but don't need to) lead to audible chopper noise, even during stand-still of the motor:

  • Low chopper frequency
  • Use of mixed decay mode
  • Low motor supply voltage
  • High motor coil resistance
  • Some microstep positions with very low motor currents on one coil
  • Filter elements on SRA and SRB pin required, but not used ? Blank time too low / higher than necessary
  • Bad PCB layout: long / thin / bend traces between GND and sense resistors or to BRA/BRB

So, for a given application, the following main points should be checked:

  • Increase chopper frequency to 36kHz.
  • If resonable for your motor, it is advised to switch off mixed decay mode during stand-still. (don't do this, if the exact microstep position is very important).

What are possible causes for a driver IC failing?

These are some hints for failure conditions, which can (but don't need to) damage the driver ICs:

  • Motor Voltage above 34V (for non-A-types) / above 40V (A-types) or logic supply voltage above 8V
  • Pulling motor connector during operation
  • Short to GND without high side sense resistor
  • Soldering problems (unconnected or shorted pins)
  • Motor current set to more than 3A for extended periods of time
  • Missing or far too small oscillator capacitor
  • High ESD voltages applied to circuit before soldering it to board. We have never seen this happening, but please remark, that the drivers are sensitive to electrostatic (dis)charge, especially before soldering the IC onto the application board.
  • Missing GND connection when powering the unit with a laboratory supply
  • Missing sense resistor (or too high value)
  • Missing short to GND resistor (or too high value)

All of these are severe violations of the drivers specifications. In fact the devices are very rugged and may be operated well beyond spec without problem.

Q: Can the TMC236 drivers operate at voltages > 28.5V?

For the TMC236 and TMC246 operation is specified up to 28.5V Motor voltage. Typically the drivers can work at up to 33V for unlimited time, but this is not specified, as the power MOS transistors inside the driver are 30V types. The transistors generate excess heat due to zener diode effects, when their maximum drain-source voltage is exceeded. The new A-types use 40V transistors and are qualified for 34V operation.