• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 532 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.

TPS92518: 0.1% duty on PWM dimming

KANJI SHIBATANI
Mastermind 6205 points
Part Number: TPS92518


Hello

 

We see data sheet page1 is 0.01%(10,000:1)   for PWM dimming range.

So, Customer is evaluating TPS92518 EVM.

Customer  applies PWM  signal from their MCU to TPS92518EVM.

Customer's  issue is  ILED =0 mA at PWM=0.1% duty.  ..test result is excel file.

Customer is using CREE's 36V/1A LED. (VIN=48V)

Anyway,

I had tested TPS92518EVM myself using PWM signal in LEDSPIMCUEVM and GUI, using same CREE's LED,VIN=48V.

I had also found that ILED is approx 0mA at PWM=0.1% duty.

My test result is attached PPT.

 

Q1

Customer's and my test result about ILED was approx 0mA when duty = 0.1% ...  but would you please advise about the reason?

We want to know whether  our usage ( register setting ) was wrong or no.

 

Q2

We see TPS92518 EVM user's guide (sluubm0) Fig4 shows Output current 5mA~10mA even  duty=0%.

We want to know how to  add Offset as Fig4 of user's guide.

( I heard from you that Fig4 is based on PWM dimming only ( not shuntFET dimming) , but how do you achieve Fig4 ?)

 

Q2

We think the corrective aciton should be customer's  MCU set higher duty in case small duty ,intentionally.

But concern is linearity curve ( gain/offset) may be changed each VLED ( Customer will use  several kind of VLED in their same PCB).

Do you think linearity curve ( gain/offset) may be changed  by VLED ?

( I should ask customer to use 0x0Ah~ 0x0Fh for VLED reading?...)

 

Thank you for your support!

Thank you for your cooperation

Best Regards

Characteristics_001a 0712_customer_English.xlsx

TPS92518EVM_under_1%.pptx

  • 0
    TI__Mastermind 28655 points
    Hello Kanji,

    You cannot get that ratio of dimming from the PWM pin dimming. The current in the inductor cannot ramp up to full value and back down. To get to 10,000:1 you need to use shunt FET dimming. It is also affected by PWM dimming frequency, the inductor value and Vin-Vout. If you can, look at the inductor current and you will see it starting to rise but not get to full current.

    1) What frequency is your PWM? The scope pictures can't be right at 2 uS/division. I see 975 Hz in the the xlxs files. This would mean that the on-time for 0.1% is 1 uS. The current in the inductor cannot ramp that fast.

    2) Yes, it will vary because Vin-Vout varies. If they want higher resolution PWM dimming they should switch to Shunt FET dimming. They also will need make the design have low ripple since the minimum PWM pulse is 1 uS.

    Regards,
  • 0 in reply to Irwin Nederbragt
    Mastermind 6205 points

    Dear Irwin-san

    Thank you very much for your quick answer !!!

    1) I had tested myself at PWMx=500Hz and ILED=0.08mA.   When I set higrt,   PWMx= 1KHz, ILED became 0.01mA and LED distinguishes.

       To have lower PWMx  frequency is better , but if it were too low ( 400Hz), LED_BOOT_UV found when duty was small.

    2) Thank you for your suggestion, I will ask customer to have additional circuit  to have shunt FET dimming.

     However, we don(t know how to design the combination of PWM dimming and shunt FET dimming actually , but I will have another post.

    Thank you again