TLC5940: Regarding the issue of LED not lighting up properly

Hi Kaji,

Please refer my comments below:

1. The chip doesn't power up properly.

2. The IREF pin is not connected correctly.

3. The communication timing is wrong.

BR, Jared

Thank you for your reply.
I received the following questions from a customer in response to your answer.

>1. The chip doesn't power up properly.
How can I check if it is powering up properly?

>2. The IREF pin is not connected properly.
This is the pin that is input to the REF of the OPAMP.
Is it "correct" if a constant voltage is applied from when the power is turned on to when it is turned off?

>3. The communication timing is wrong.
Is this communication timing serial communication?
I understand that serial communication is setting the GS register and DC register.
Assuming yes, I understand that if an abnormal value is set incorrectly, it will not light up at the desired brightness.
However, there is also a symptom of flashing. Is this behavior possible if the communication timing is wrong?

Best regards,

Hi Kaji,

1. How can I check if it is powering up properly?

- I think each device will have a input capacitor at VIN pin, so you can directly probe the VIN voltage.

- Or you can probe the VREF voltage, if the VIN power supply is normal, the VREF will be about 1.24V.

2. This is the pin that is input to the REF of the OPAMP.
Is it "correct" if a constant voltage is applied from when the power is turned on to when it is turned off?

- I think VREF pin doesn't need to apply on a constant voltage. It's from IC itself.

3. Is this communication timing serial communication?
I understand that serial communication is setting the GS register and DC register.
Assuming yes, I understand that if an abnormal value is set incorrectly, it will not light up at the desired brightness.
However, there is also a symptom of flashing. Is this behavior possible if the communication timing is wrong?

- Communication can cause a lot of problems, depending on the format of the commands the MCU sent, and of course may cause the lights to blink or turn off. Btw, if it's really a communication problem, then I think it's probably because the software engineers are inexperienced and the code is not robust enough.

BR, Jared

Thank you for your reply.
I appreciate your quick response.
However, I have received additional questions from the customer, so would it be possible for you to answer the following?

>Communication can cause a lot of problems, depending on the format of the commands the MCU sent, and of course may cause the lights to blink or turn off.
The customer said that this is a possibility, but that the symptoms do not change even if he sends the command again.
If the device goes into an abnormal state (flashing or off), will the symptoms improve even if he resends the command?
Does the power need to be turned on and off to return it to a normal state?

Best regards,

Hi Kaji,

I am wondering that what do you mean about the abnormal state.

Are you sure the chip is not working properly? What I mean here is that your communication, power, connection and other things are all correct, but only TLC5940 is abnormal.

For your doubt, I don't think it's TLC5940 problem since you said that the device goes into an abnormal state sometimes and other is normal. I think it means your communication is abnormal sometimes and other is normal.

For me, I think the most important is to fix this problem, as I said before and in another e2e you post. My suggestion is to follow the timing sequence on datasheet and you have verified it. So is the problem we discuss about another problem?

Finally, I want to clarify that the behavior of the LED driver is only rendered according to your communication commands, so if you see strange phenomena, I think the most possible reason is communication problem.

BR, Jared

Thank you for your reply. I appreciate your prompt response.
My client is dutifully considering your advice to resolve this issue.
I think this will be my last question, so please let me know your thoughts.

I would like your guarantee that the current problem of the LED turning off and blinking will be completely improved by observing the constraints of th4 and tsu4.
To that end, I asked about possible causes of the current problem of the LED turning off and blinking, and received the following four answers.
① When the constraints of th4 and tsu4 are not observed.
② It is not started up correctly.
③ The IREF pin is not connected correctly.
④ The communication timing is incorrect.
I have verified these four causes,
and for ② there is no problem with a VIN voltage of 3.3V.
For ③, the voltage of the VREF pin was measured and there is no problem at 1.24V.
For ④, the sent command has been confirmed, but since it is impossible to determine whether the LED driver is receiving it correctly, I tried to determine whether resending it would work correctly or not by seeing if resending it would improve the symptoms.
If the problem of the LED turning off and blinking is caused by incorrect communication timing, I believe that it will work correctly if I resend it after the problem occurs.
Currently, if a problem occurs, resending the command does not improve the symptoms, so if we can get an answer to ④, we can determine that problem ① is the only cause and that taking measure ① will prevent the current problem from occurring.

Based on the above, could you please answer the following questions?
(Question)
If the device goes into an abnormal state (blinking or off), will the symptoms improve after sending the command again?

Best regards,