The Toshiba TB67S109AFTG is a 47 volt, 3 amp, up to 32 microstep driver chip that is fairly inexpensive. Individual drivers can be purchased on eBay for about $7.  The driver output current waveform looks better than the TB6600. I think it runs stepper motors better than the TB6600 however it is limited to only 3 amp output current.  Many eBay listing for the TB6600 are actually TB67S109 chips inside the driver.  If the driver has switch settings for 32 microstep, chances are it is a TB67S109 instead. The TB6600 is only a 16 microstep driver.

TB67S109AFTG current setting test. 3 amps was the maximum this driver can do.

Test Conditions
24Volt Power Supply
Rigol DS1054Z Oscilloscope
Hantek CC-65 Current Probe

Stepper motor used for testing
59Ncm 83.6oz-in

Settings (amps)
RMS    Peak    Measured RMS     Measured Peak
0.5      0.7        0.24                         0.29
1.0     1.2         0.64                         0.86
1.5     1.7         1.0                           1.38
2.0     2.2         1.22                         1.70
2.5     2.7         1.96                         2.77
2.8     2.9         2.02                         2.91
3.0     3.2         2.09                         2.9
3.5     4.0         2.13                         3.04

Toshiba TB67S109AFTG microstep sine wave output current waveform captured with a oscilloscope using a current probe. The output traces look better than a TB6600 driver and should run stepper motors smoother. The TB67S109AFTG does not have any anti-resonance or mid-band compensation. Expect some motor vibrations in the midband 300-500 rpm range. This resonance can lead to motor stalling or lost step positioning.

The board is labeled with 40 volts max recommended motor supply voltage. I did test the driver with 36 volts and no magic smoke came out. At 36 volts, I was able to spin a stepper motor over 2000rpm.

The step, direction and enable inputs do have 4n25 and PC817 optocouplers for safety.

Heatsink temperature testing.
I ran a stepper motor test run at 500rpm. Stepper driver set at 2.2 amps with 36 volt supply. After about 20 minutes, the heat sink temperature leveled off at 45C. I kept running the motor for a hour without any extra rise in heat sink temperature. Ambient temperature was 24C and no fan was used to cool the driver. Temperature measured with a K-type thermocouple.

I ran the test again with the stepper driver set at its maximum 4 amp peak setting. After a few minutes the heat sink quickly reached 55C and the driver shutdown. The TB67S109AFTG has internal over temperature thermal safety shutdown.

To fix this problem, I took the driver apart and removed the silicon pad attached to the small square aluminum block that is suppose to transfer the heat. These pads don’t really work very well. I then sanded down the two plastic stand-offs about 1 mm, the thickness of the silicon pad. I added thermal grease on both sides of the block and carefully screwed the board back together. Running the 4 amp test again, the heat sink leveled off at 76C and ran for over an hour. If your driver goes into thermal shutdown, this may fix the problem. Add a fan to help cool down the heat sink better, especially if the drivers are in a enclosed case.

Step pulse input testing. 
I set the microstep to 32, the maximum for this driver. Using a HP8116A Pulse generator, the stepper driver was able to handle 105Khz step pulse rate. It works best with at least a 3 microsecond pulse width.  Power supply set to 36 volts.   The low step pulse rate is probably due to the slow 4n25 optocoupler used in the step input path.

Just like the TB6600, the TB67S109 is a very basic stepper driver. Not recommended to buy when better stepper drivers are available such as the Trinamic TMC2160.

TB67S109AFTG 8 microstep 100rpm 12voltsTB67S109AFTG 8 microstep 100rpm 12volts

TB67S109AFTG 8 microstep 100rpm 24voltsTB67S109AFTG 8 microstep 100rpm 24volts


Update 3/21/2021
Anti Reiljan sent me two images of a newer revision of the TB67S109 driver. The board has a cleaner layout and soldering job. The silicon heat pad is replaced with a thicker aluminum block with thermal heatsink compound on both sides. This should fix the thermal shutdown problem with the older design.