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.

TPS92640 - PWM dimming and thermal pull back issue

Other Parts Discussed in Thread: TPS92640

Hello,

please help us to overcome an issue...

We use pin 3 (UDIM) in regards to PWM dimming - it works. Besides, the implemented circuit is identical to application note in datasheet.

Problem:

Additionally we use pin 6 (IADJ) to implement thermal pullback in case of elevated ambient operating temperatures.

The corresponding circuit is:

pin 5 (VREF) - 15k - PTC - pin 6 (IADJ) - 10k - GND

Besides - the PTC is a 110°C component with a resistance of 680Ohm before .

The issue is that this kind of thermal pullback doesn't have any effect on the LED current although the voltage on pin 6 (IADJ) is changing due to the PTCs resistor change at elevated temperatures.

QUESTIONS:

Can anybody advise us please so we can overcome the issue?
What is the switch "PWM_DIM" for ... on the OTA output (pin 6 IADJ) ? (Refer to page 10 of TPS92640 datasheet)

Any known issues regarding thermal pullback implemented in that way?

Thanks

Erwin

  • It should work fine that way. But it's possible it is misbehaving, or even damaged, due to the excessive current draw from VREF and into IADJ. When the PTC is only 680 ohms the voltage between VREF and the internal IADJ clamp is 676uA. The ABS MAX current rating for both pins is 200uA. I would 10x the PTC and resistor value and try it again with a new IC.
  • maybe you have misunderstood my problem....
    The PTC is not connected between Vref and Iadj.

    The corresponding circuit is:

    pin 5 (VREF) --- 15kOhm resistor --- PTC 680Ohm --- pin 6 (IADJ) --- 10kOhm resistor --- GND

    The voltage on pin 6 (IADJ) is changing correspondingly to PTC resistance at higher ambient temperatures but there is no effect on the driving current.

    So I doubt that the internal switch on the OTA output has an effect on this behaviour. What is the switch "PWM_DIM" for ??
    .... Refer to page 10 of TPS92640 datasheet

    Appreciating your further help. Thanks.
  • I understand. Could it have anything to do with the PTC resistance change rate? Does this happen on multiple boards or just one? Can you verify that the IADJ pin is soldered down correctly with no cold solder joint? Have you tried replacing the IC to make sure it isn't just damaged? This behavior was not observed during over temp silicon evaluation. The PWM_DIM switch in the block diagram is there to disconnect the COMP capacitor during the LED off time so that when they are turned back on COMP is still at regulation level so the current rise time is much faster and more accurate.
  • We have found the issue:

    The converter was running at 900kHz switching frequency - the input voltage range is 16V - 32V and the LED forward voltage is about 13V. The ouput current is 1,2A.

    The problem was that the converter was already running with regulators minimum possible ON time (based on datasheet values) and the IADJ had no effect on reducing output current.

    The design works properly and output current is reduced according to PTC characteristics when switching frequency was set to 400-450kHz and regulators minimum ON time is considered.

    --> double check regulators minimum possible ON time in similar cases

  • That's good to hear. I guess I should have asked for more information initially. As soon as I see a switching frequency that high minimum on time is one of the first things I consider. So good catch.