2173491_en-US

[#1]

PCA9633TK， The problem of chip output logic being reversed

PCA9633TK， We use an 8-pin package with a 5V power supply for the chip. We use two outputs, LED0 and LED1, while LED2 and LED3 are suspended and not in use.
The input and output of PCA9633TK are currently being directly tested without LED connection. The IIC configuration has a PWM0 value of 240, but the waveform duty cycle of LED0 output is very low. The IIC configuration has a PWM0 value of 15, but the waveform duty cycle of LED0 output is very high. The input and output logic of PCA9633 is reversed. The register configuration is as follows:
Address 0x00: Data 0x 01;
Address 0x01: Data 0x 07; //INVRT=0
Address 0x02: Data 0x 28;
Address 0x03: Data 0x 46;
Address 0x08: Data 0x 0A; //LEDOUT=10
Other registers are not configured.
What are the possible reasons for this phenomenon?

Re: PCA9633TK， The problem of chip output logic being reversed

Dear Tanpo1989,

The PCA9633 datasheet does not use the phrase “low-edge LED driver,” but it clearly describes the default polarity and behavior:

1. LED outputs are current sinks
   The LED output driver can be configured as open-drain (sink only) or totem-pole (sink and limited source). In both cases, the LED is intended to be connected from VDD to the LEDx pin, so the device sinks current when the output is LOW. This is stated as:

   “The LED output driver is programmed to be either open-drain with a 25 mA current sink capability at 5 V or totem pole with a 25 mA sink, 10 mA source capability at 5 V.”

2. Default power-on state = HIGH (LED off)
   After reset, outputs are HIGH, which means LEDs are off. This implies that LOW = LED on.

With Best Regards,

Jozef

Re: PCA9633TK， The problem of chip output logic being reversed

Thank you for your reply.

Where does the PCA9633 manual state that it is a low edge LED driver? Manifesting as high PWM values → more low-level time → displayed on the oscilloscope as low high-level duty cycle。

Re: PCA9633TK， The problem of chip output logic being reversed

Dear Tanpo1989,

you're seeing exactly what the PCA9633 is designed to do when it’s used as a low‑side LED driver and you probe the pin voltage without an LED/load.

• What’s happening: PCA9633 is a low-side LED driver. With INVRT=0, the LED pin goes LOW when the LED should be ON. So if you measure the pin voltage, the duty looks inverted:

  • High PWM value → more LOW time → small HIGH duty on scope.
  • Low PWM value → less LOW time → large HIGH duty.

• Why: You’re measuring voltage, not LED brightness. The chip is working as designed.

• Your config: Correct for sinking LEDs from VDD to LEDn. LEDOUT=10 means PWM control, and INVRT=0 is normal.

• Options:

  • For real LEDs: leave as is.
  • For bench testing without LEDs:
    • Invert the scope channel or measure LOW duty, or
    • Set INVRT=1 temporarily (but then brightness vs PWM flips).
    • Add a resistor to VDD to simulate a load.

Bottom line: The “reversed” effect is just because the outputs sink current, so LOW = ON.

With Best Regards,

Jozef