TPS54200EVM-818: Led does not turn on beyond 66.18% duty cycle

Hi Manan,

I need more info. and please provide the SW waveform when device is abnormal. Thanks!

Robin

Hi Robin,

This is the PWM wave at 66.57% duty cycle around these values as indicated in the Fig 1

Fig 1 : PWM wave which leads to dimming of the 1W white led

Fig 2 : PWM signal (yellow) sorry for changing the color of PWM signal, values of the signal are same as Fig 1 and signal seen on SW / TP2 of EVM (blue) [1W white led turns on]

Fig 3 : SW signal with various measurement when the applied PWM signal is same as Fig 1

Fig 4 : same SW signal as Fig 3, just zoomed out with measurement

Fig 5 : PWM signal same as Fig 1 (yellow)  ; signal seen on Vsense / TP7 pin (pink) ; signal seen on SW / TP2 pin same as Fig 3 and Fig 4 (blue)

Fig 6 : as signals as mentioned in Fig 5, just zoomed in

Fig 7 : PWM signal whose duty cycle is 68% (yellow) as seen, no signal is seen on Vsense (pink) and SW (blue)

SW signal when EVM behaves abnormally

(i have uploaded a YouTube video of scope measurement of signal seen on SW pin of EVM since the video size was exceeding e2e forum stated size, kindly mute the video to avoid background noise and you can view the measurements of SW signal on scope if you slow down video to X0.25)

in the video, PWM signal has duty cycle of 68%, and Vin is ramped from 1V to 5V from power supply, as you will see in the video, there is hardly any change in the SW signal during the change in Vin, but when Vin reaches 4V - 4.1V you will see the there is change in SW signal and momentarily you will be able to read the measurement of the signal, but once the Vin moves beyond 4.1V, SW signal does not change.

SW signal & PWM signal along with measurements when device behaves abnormally

(YouTube Video)

same video as above, in this video measurement for PWM signal are also there.

Conclusion :

PWM signal generated by uC (adafruit feather m0 ) which is passed to voltage divider (both resistors are 10k) and that PWM signal is given to pin TP 9 of EVM, when PWM signal`s duty cycle is below 66% and VIN is above 4.1V the 1W white led turns on and dimming starts.

When the PWM signal is above 66% duty cycle (in the above Fig 7 and both videos, the duty cycle is 68%) the led does not turn on and only flashes for 1 second when the voltage from power supply at VIN reaches 4.1V and once the VIN voltage is even above 4.1V (for example 5V or 5.5V) the led does not turn on.

Hi Robin, could you help Manan in this case.

Hi Manan,

How about the output current, inductor value and inductor saturation current?

Robin

there is no output current when device is in the abnormal mode and since it is EVM, inductor value is as stated in the EVM datasheet.

inductor value is 10uh and its saturation current mentioned in datasheet is 4A.

https://www.ti.com/lit/ug/sluubi2/sluubi2.pdf?ts=1617935151668&ref_url=https%253A%252F%252Fwww.google.com%252F

https://www.mouser.in/datasheet/2/445/744066100-538742.pdf

Hi Manan,

Then there is no relationship with internal OCP. This phenomenon is odd and I'll do bench test based on your test conditions.

Robin

Hi Robin,

any updates on the bench test ?

Hi Manan,

I didn't repeat the phenomenon just like yours. Can you check the circuit again or replace TPS54200 with a new one?

Robin

Hi Robin,
i did replaced the TPS54200 ic on the EVM, but unfortunately, the condition is still same as i have mentioned above in the posts.

Please bench test the EVM and try replicating the conditions i have mentioned.

I have been trying to the get the EVM to working for weeks now, and it is not working properly instead it is behaving abnormally, last option for me, will be to change and move on to different IC which can perform the same function as advertised for TPS54200.

Hi Robin, could you help to keep following this case. thanks

Hi Manan,

Could you share the P/N of LED? Besides, how about the current sense resistor? Did you change it?

At the same, TPS92200 is the latest similar device. If needed, you can refer it.

BR,

Robin

Hello Robin,

what do you mean by P/N of led, do you mean forward voltage of led, it is 2.56V and it is 1W white led.

no, i haven't changed current sense resistor, since it is EVM we are talking about.

Since i have already made the PCB, have attached the picture of schematic above, as you can see, the current sense resistor value is different from one on the EVM, also the values of passive components like inductor and capacitors are different. I had design TPS54200 circuit keeping in mind, my load which is a 1W IR laser diode, and i am using 1W white led as a substitute for load. For reference, i have also attached schematic of TPS54200EVM-818 schematic :

                                                       

Robin, can you please bench test the TPS54200 EVM with 1W white led along with conditions i have mentioned above, and please let me know, because i have invested quite a lot of time and resources and built a laser dimming pcb around TPS54200 and i am facing same problem in the PCB, hence i purchased EVM, assuming there might be issues with my PCB, but to my surprise, EVM also has the same issue. It has been 26 days now, since i raised this issue.

If this issue is unreasonable, please do let me know that too, so i don't waste anymore time and resources on TI`s TPS54200 and start looking at other manufactures solution.

Hi Robin please help to keep following this case. thanks

Hi Manan,

Understood. I'll try again.

Robin

Hello Robin,

Any update ?

Since the laber day is from 5.1 to 5.5, our expert will reply you on 5.6