UCC28600: UCC28600

Hi Vivek,

The UCC27531 and UCC27532 are nearly identical. The only difference is that the input pins of the UCC27531 is independent of the VDD supply voltage (typical input rise threshold voltage = 2V and typical input fall = 1V), while the input pins of the UCC27532 is dependent on the VDD supply voltage (typical input rise threshold voltage is 55% of VDD and input fall is 45% of VDD).

It looks like the OUT pin of the UCC28600 (assuming you are connecting that pin to the IN pin of UCC27531) switches between GND and VDD of UCC28600. In that case, I would recommend using the UCC27531.

Thanks,
Rubas

Hi Vivek,

I believe the only difference between the UCC27531 and UCC27537 is the split output (OUTH and OUTL vs. only OUT). This FAQ explains the advantage of split output: https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1387611/faq-what-is-split-output-and-how-to-use-it

Thanks,
Rubas

Hi Vivek,

An advantage of split output is that you can individually control the source and sink current strength. This advantage shines more with gate drivers that have different source and sink current strengths, which is the case for the UCC2753x family of devices (2.5A source and 5A sink).

While using a diode in a single-output gate driver can ideally replicate the benefits of a split output gate driver, the problem starts to appear when taking account of the diode parasitics. The parasitic inductances from the diode can lead to false pulses when the gate driver sinks current fast (V = L * di/dt). The false pulses can be small, but large enough such that it reaches the turn-on threshold of the SiC FET. Repeated pulses that barely turn on the SiC FET can damage the SiC FET overtime due to conduction losses. Of course, using the diode method with a higher gate resistance can mitigate the false pulses, but then the switching times will be slower, which could cause switching losses. The diode will also increase the bill of materials cost.

There's many pros and cons to using a driver with split output vs. single output, and it all comes down to your project requirements. Adding a split output functionality would typically mean it replaced another functionality; an example is the UCC57108, where the split output variant removes the VEE pin. However, in the case of UCC27531 vs. UCC27537, there's no other functionality or difference other than split output. Because of that, I would recommend you choose the UCC27531.

I also recommend taking a look at the UCC57108/UCC57102 family of devices, which is our latest family of gate drivers designed for SiC FETs and IGBTs and has desaturation protection for overcurrent events.

Thanks,
Rubas

Hi Vivek,

Please ensure that the UCC27532 will fit your design by referring to my first reply in this thread.

1. We always recommend an RC filter for our IN pins. A resistor value between 0-10ohms and a capacitor value between 10-100pF will suffice. It's not recommended to slow down the rise/fall time of input PWM signals, as it can cause issues with ground bouncing.

2. The UCC27532 will reference the PWM signal that is entering the IN pin to whatever the UCC27532 GND is referenced to. You need to ensure that all signals referenced to UCC27532 GND is within the gate driver ratings. For example, let's say that +Ve = 12V and UCC28600 GND = DC- = -6V. The PWM signal produced by UCC28600 will be -6V to 12V, and if the UCC27532 GND = 0V, then the UCC27532 IN pin will process -6V to 12V. Since the absolute max specs of the UCC27532 rates the IN negative voltage handling at -5V, this PWM signal could potentially break the UCC27532. If the UCC27532 GND = -3V, then the UCC27532 IN pin will process -3V to 9V, which is within the UCC27532 ratings.

3. The UCC27532 does not have a status pin. If you mean the UCC28600, then I would recommend creating a separate thread for UCC28600 questions, as their team can better answer your questions.

Thanks,
Rubas