• State
  • Locked Locked
  • Replies 15 replies
  • Subscribers 64 subscribers
  • Views 562 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TPS92641: Issue with Non-Linearity and Flickering in Constant Current LED Driver Using Shunt FET Control

Rahul Yadav
Prodigy 30 points
Part Number: TPS92641

Tool/software:

I have designed CC driver controlling through Shunt Fet at 25Khz PWM frequency 
I am not getting proper linear out and the LEDs are also flickering at particular values like 52% intensity

Vled = 27V
Iout = 2.5A
Vin = 36V

L = 47uh
Fsw = 458khz
D = 0.85
Ron = 24k
Con = 1nf
Rvout1 = 100k
Rvout2 = 10k

Cout Cap = 0.1uf
VIadj = 0.2V
Rcs = 0.08E 

I am attaching all reading from my DSO at 40% Duty Cycle, at this value there was no flickering

VIN


Vref


RON


COMP


SDRV


CS


SW

LG

HG

VOUT


Output current vs Duty Cycle graph


Current Readings

Duty Cycle OUTPUT CURRENT
1 0.009
2 0.015
3 0.02
4 0.13
5 0.126
6 0.137
7 0.157
8 0.158
9 0.171
10 0.182
11 0.219
12 0.246
13 0.274
14 0.288
15 0.306
16 0.316
17 0.343
18 0.362
19 0.385
20 0.416
21 0.442
22 0.464
23 0.474
24 0.495
25 0.513
26 0.55
27 0.571
28 0.6
29 0.616
30 0.635
31 0.657
32 0.698
33 0.708
34 0.738
35 0.767
36 0.792
37 0.82
38 0.846
39 0.863
40 0.893
41 0.914
42 0.957
43 0.966
44 0.986
45 1.014
46 1.02
47 1.066
48 1.099
49 1.114
50 1.193
51 1.149
52 1.19
53 1.238
54 1.255
55 1.275
56 1.284
57 1.325
58 1.355
59 1.38
60 1.414
61 1.425
62 1.439
63 1.464
64 1.504
65 1.536
66 1.534
67 1.571
68 1.584
69 1.613
70 1.642
71 1.666
72 1.69
73 1.713
74 1.716
75 1.792
76 1.81
77 1.824
78 1.836
79 1.884
80 1.915
81 1.934
82 1.961
83 1.965
84 1.9866
85 2.022
86 2.06
87 2.09
88 2.106
89 2.12
90 2.14
91 2.18
92 2.2
93 2.24
94 2.26
95 2.27
96 2.29
97 2.33
98 2.37
99 2.38
100 2.46
  • 0
    Prodigy 30 points

    Also attaching Schematic for reference

  • 0 in reply to Rahul Yadav
    TI__Mastermind 27296 points

    Hi Rahul

    The schematic is not clear, can you please re-attach the file? And can you please share what's the end equipment ?

    Thanks and best regards.

  • 0 in reply to Gui He
    Prodigy 30 points

    I am using TPS92641 for LED panel controller 
    Panel is 9 in series and 18 in parallel
    I selected TPS92541 for high resolution pwm dimming as it supports 20000:1 Shunt FET PWM Dimming
    Also, does it support this range at any frequency or is there any pwm frequency limit?

    For reference i have attached both png and pdf of schematic

     TPS_SCHv5.pdf

  • 0 in reply to Rahul Yadav
    TI__Mastermind 27296 points

    Hi Rahul

    I will check this on our evm and reply to you later.

    Thanks and best regards.

  • 0 in reply to Gui He
    Prodigy 30 points

    Hi Gui He

    Have you checked?
    Waiting for your response.

  • 0 in reply to Rahul Yadav
    TI__Mastermind 27296 points

    Hi Rahul

    Sorry for late reply,  my computer got some issue last week. 

    You are performing PWM dimming using SDIM &SDRV , and the  UDIM -PWMB/PWMC/PWMW... signal is not used, right?

    Can you please set the WPM dimming frequency as lower frequency like ~200Hz  and see if the linearity is improved?

    Thanks and best regards.

  • 0 in reply to Gui He
    Prodigy 30 points

    I have tested both UDIM and SDIM to evaluate their effects. I observed that reducing the PWM frequency improves linearity. The shared readings and graph were recorded at 25 kHz. At 10 kHz, the linearity showed some improvement. I will conduct further testing at 200 Hz as suggested. However, my application requires a frequency above 20 kHz.

  • 0 in reply to Rahul Yadav
    TI__Mastermind 27296 points

    Hi Rahul

    Thank you. May I ask what's the system consideration to set f_dim >20kHz? 

    Best regards.

  • 0 in reply to Gui He
    Prodigy 30 points

    What do you mean by system consideration?

    I chose above 20khz to avoid any audible noise, to avoid any led flickering at any fps and even for slow motion cameras.
    And also I tested at 1khz and the linearity is pretty good 

  • 0 in reply to Rahul Yadav
    TI__Mastermind 27296 points

    Understand. So although under 1khz the linearity is ok, we need to find other ways to deal with this, right?

  • 0 in reply to Gui He
    Prodigy 30 points

    Yes, I want better linearity at 20khz or above in shunt dimming

  • 0 in reply to Rahul Yadav
    Prodigy 30 points

    Have you found any issues in my schematic or the DSO waveforms I shared of all the pins of IC?

  • 0 in reply to Rahul Yadav
    TI__Mastermind 27296 points

    Hi Rahul

    The schematic looks ok to me. 

    The RON waveform, the voltage amplitude seems not stable, can you please measure the RON when there is no shunt dimming?

  • 0 in reply to Gui He
    Prodigy 30 points



    Hi Gui,

    I have attached a video file of RON signal in multiple grids for better understanding, please check and let me know how can i improve linearity.

  • 0 in reply to Rahul Yadav
    TI__Mastermind 27296 points

    Hi Rahul

    My thought is that you can make lower dimming frequency (but there is requirement we need >20kHz here) or higher switching frequency. Can you use higher fsw? I think reducing the output current ripple might also help, this can be achieved by larger inductance/higher fsw/larger Cout.

    Thanks and best regards.