DRV8842EVM: Regarding Current Regulation

Hello Kobori-san,

No worries on the late reply, I am happy to help. I think I might have found why the current regulation is not behaving as expected, the current regulation threshold of  -100mA to100mA seems to be too low for this device. More information below and possible fixes at the end of the explanation:

 The way in which the devices regulates current is described in section 8.3.3 of the datasheet:

8.3.3 Current Regulation:

 The maximum current through the load is regulated by a fixed-frequency PWM current regulation (I believe this is around 50kHZ), or current chopping. When the H-bridge is enabled, current rises through the winding at a rate dependent on the DC voltage and inductance of the winding. After the current hits the current chopping threshold, the bridge disables the current until the beginning of the next PWM cycle. (This is from either the internal PWM or the Input PWM)

It is important to note that this information needs to be paired with blanking time:

8.3.4 Blanking Time

After the current is enabled in an H-bridge, the voltage on the ISEN pin is ignored for a fixed period of time before enabling the current sense circuitry. This blanking time is fixed at 3.75 μs. Note that the blanking time also sets the minimum on-time of the PWM.

We can see the combination of these effects happening on the Case 1 Waveform of TEST 2 that you uploaded:

We can see that when given 100% duty cycle the outputs go high consistently for around 3.75 μs, as soon as that blanking time passes it immediately turns off because the current is higher than the current regulation threshold of 100mA. And it remains low until the internal current regulation PWM mentioned in section 8.3.3 (50kHZ) causes current to flow again. Effectively, this is the minimum on time that current regulation can achieve, therefore it can not effectively regulate current below this if driven at 100% PWM.

Now I would like to explain why Case 2 possibly had an even higher current with lower duty cycle than Case 1. Here is the image you uploaded of case 2 at 75% duty cycle:

Here we can see that there is an overall higher on time than in Case 1. If we look at the rising edge of a PWM cycle in the input we will see that the output will rise and stay on precisely for 3.75us (the blanking time) before turning off because the current is above the Itrip threshold. However the Input signal stays high for 20uS (50kHz) which cause the output to rise again due to the internal PWM for the current regulation clock. This is similar to the situation of Case 1.

Where it differs is when the falling edge of the IN PWM signal appears. When it does the Output follows it and goes low and stays low until the next rising edge of the input PWM signal. When the Input PWM hits the rising edge, the output will enable and stay high only for 3.75us of blanking time before going low again and repeating the cycle. It is important to remember that the output only stays high for 3.75us because the current through the motor is still above the current threshold disabling the output whenever it can.

This would explain why you are experiencing higher currents at duty cycles between 70%-99% than at 100%.

Past that in cases 3 and 4 the current regulation seem to be working as intended. Turning off the outputs once the current is reached and only re-enabling it when the next rising edge of the input signal appears.

SUMMARY:

All of these problems stem from the fact that at this input voltage level the current is not decaying fast enough to stay in the sub 100mA threshold that you are targeting. Even the minimum on time for the output set by the blanking time and internal clock is causing too great of a current. Therefore, the main solution should be to find ways to minimize the current increase, and to maximize the current decay. Therefore, these are my suggestions:

• If you haven’t already try changing the device’s decay mode to fast decay. This can be done to setting the decay pin to logic high, and should help increase the current decay.
• If possible lower the input voltage so that current will rise slower through the windings.

Please let me know if the fixes provided were helpful or if you have any questions about the operation of the system.

Best,

Pedro Arango Ramirez.