This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TLC6C5716-Q1: ERR Pin being asserted with no errors present.

Part Number: TLC6C5716-Q1

Tool/software:

Hello TI Team, 

We are working on an LED application and are doing some prototyping. We found that when we go to turn on all the LEDs that R0 is significantly dimmer than the rest of them and that the ERR pin asserts as soon as we perform as GS write. When we read the SID register however, it shows that the APS and LOD-LSD tests passed and no error flags are set. We begin the initialization with an error clear and global reset which eliminates the error until a GS write happens. We are driving the serial lines with gpio and running at around 120kbits/sec. We are using a free-running GCLK output on a pin of our microcontroller as suggested here, it's running at 4MHz. The code we are using is based off of the sample code provided here. When I read back the RX buffer after as GS write, I find that the data matches the data sent on the FC-BC-DC write, which suggested that the R0 DC is correctly set to the same as all the others. On Page 32 of the data sheet on table 12, a PWM Error is mentioned. What are the conditions for the PWM error and could it be causing the brightness to not be correct? 

  • Hi Sam,

    Which bit is your err? Bit 204?

    Is it possible that this LED is open or short? What's your GS data value?

    COuld you pls mask the LED ERR first?

    BR, Gracie

  • Hi Gracie, 

    There is no err bit set in the SID register, the only bits set are in the reserved section. When I enable the LED_ERR mask these bits were no longer set either. 

    I tried masking the ERR bit in the FC-BC-DC flags, and the light stopped coming on. I was, however, looking to use this pin as an interrupt source on the uC for asynchronously detecting errors. Does the ERR pin work correctly for this purpose?

    After this change the R0 LED is still dimmer than the rest of the LEDs. The values are the same for all LEDs BC=0x0F, DC=0x1F, GS=0x3FF. 

  • Hi Sam,

    We will give you update next week.

  • Hi Sam,

    Did you change another board or IC to see if this error still occurs? It's quite unusual.

    BR, Gracie

  • Hi Gracie, 

    We did change to another board and this solved the brightness issue. However, the ERR pin issue still persists. Currently it is showing that the LSD-LOD self check and APS self check have passed, and no other ERR bits are set from bit 215-204 in the SID Register. Despite the tests passing, all the LSD bits are high and the ERR LED is on. Is the ERR LED reliable for use as an interrupt to determine an error condition? All LEDs seem to be working fine and definitely are not shorted. 

    Thank you,

    Sam

  • Hi Sam,

    Gracie will give you reply later.

  • Hi Sam,

    If you look into the PWM lookup table, you will find some PWM value will also result in the LOD/LSD and you need to ignore it from your software.

    In some cases, users may need to turn off output channels before the 9th GCLK to disable the output channels, or turn on the output channels at the Nth GCLK to get more brightness. LOD_LSD faults are reported as shown in Table 6. Users can ignore the fault according to the GS register setting value.

    BR, Gracie

  • Hi Gracie, 

    I found that changing the GS "off"  value from 0 to 0xC was sufficient to get the ERR light not to come on. The LSD2 bits are still set in the SID register, but the ERR line is now usable for interrupts. Thank you.