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.

TPS92515HVEVM-749: Operating at under 100Hz?

Part Number: TPS92515HVEVM-749
Other Parts Discussed in Thread: TPS92515HV

I have a TPS92515HV evaluation board that I am using to control a strobe light. The strobe is intended to have a 50% duty cycle, requires about 34V output, and 1.3A. I am powering the TPS92515HVEVM with a 48VDC power supply.

For this application, I am simply providing on and off signals at 40Hz to the evaluation board. However, when compared to a photodiode, it seems to take 4-6ms to turn on (but turns off immediately). At lower frequencies (20Hz) and higher frequencies (over 60-70Hz) it seems to turn on sharply with the PWM pulse.

I am confused about why this is happening and if there may be some parts I can add to the board to eliminate the issue. I originally assumed that shutdown circuitry inside was going into idle mode due to the long off time and high required output voltage but I would then have expected a constant turn-on delay. Instead, the turn-on delay goes to nearly zero at both low and high frequencies. For instance, at 40Hz it has about 4ms of delay while at 50Hz it is more like 6ms.

Is there anything I can do? Is this how it is expected to work, or should I look in other places for an issue that could cause a lag on the rising edge only?

  • Hello Brian,

    I have seen similar issues before depending on the frequency. Let me guess, you are still using the output capacitor? If so, what can happen is you can get in BOOT UVLO situations depending on how far the output discharges. When the device is off the inductor current will eventually go to zero and the switch node will settle to VOUT. If it doesn't discharge too far the next on time is great, if it discharges just too far you can turn on during BOOT UVLO and there is a delay. If it goes even longer an internal switch will charge BOOT and the next cycle will be ok.

    You can avoid it completely by having no output capacitor. But you can also make it work in a much wider range of low frequencies simply by reducing the output capacitance so that it discharges enough to charge BOOT. Try with no output capacitor and then tweak it from there if you really need to reduce the LED current ripple. You likely don't need nearly 2.2uF. Let me know if this helps. Thanks.

    Regards,

    Clint

  • Thanks Clint, I removed the capacitor and it now turns on as expected! The ripple is noticeably higher, but it will be fine for me.