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.

TPS27S100: Excessive Current Spike

Part Number: TPS27S100
Other Parts Discussed in Thread: ALLIGATOR

Hello,

We have the EVM for this device. The current limit is set to 500mA using a 5kOhm resistor on ILIM. When the output is shorted to ground, the current reaches peaks of >20A for over 10us before setting down to the current limit.

Looking at the application curves in the datasheet, the current overshoot in this situation should never get that high.

Can you help us identify the problem?

Other facts:
-BW limiting the probe does not change the result.
-The EN pin is driven by 1kHz, 50% duty, PWM with 5V amplitude.
-Short is accomplished via wire with alligator clips on both ends. Vout on board is connected to GND on board.
-Vin/Vout is 24V.

Datasheet:

EVM:

  • Hello,

    We are investegating it and will get get back with you.

    Regards
  •  Hello,

    I run the test according to the same setup as per your description and using TI EVM:

    - Current limit 500mA

    - VS = 24V

    - Short OUT to GND and measure the current flowing into the wire using current probe

    - IN clocking 5V at 1KHz and 50% duty cycle

    There is no current overshoot and attcahed scope picture shows the data:

    - Scope channel CH1 = OUT

    - Scope channel CH2 = VS

    - Scope channel CH3 = IN

    - Scope channel CH4 = short circuit current

    Regards

  • Hi Mahmoud,

    Thank you for getting this tested on an EVM so quickly. This is what I would expect, since 20A spikes are so far outside the datasheet specifications.

    However, I think the problem at hand is how can we get the customer's EVM behaving properly so that the customer can confidently continue to consider the part for their design.

    1. Are there any common pitfalls that could cause this? (Wires too long/thin? Measurement error?)
    2. Any other tests we could recommend?
    3. There doesn't seem to be many ways this board could go so wrong... do we need to order them a replacement?

    Thank you,
    Ryan B.

  • Hello Rayan,
    I understood the customer test is performed on TI EVM. I checked the TI EVM as well for observation and I changed the current limit up to 5A and within 0.5 to 5A limit range I did not see abnormal peak current. I will perform more tests to see if the high current spikes can be captured. The current spike max should be 1.5 x limit set.
    When the FET is turned on and off there is delay and slew rate that help to smooth the peak current limit. Peak current limit should be more severe with short circuit step because the edge is sharper and the fast open loop response tine is 1uS.

    I believe the wire size should not affect the peak current.
    I might be some noise is coupled into the current probe during transition of IN low to high. I suggest to move away the current probe from noisy instrument such as switch mode power supply.

    Regards, Mahmoud
  • Hello Rayan,

    The cuuernt spikes is not if the output is shorted and the FET is enabled. The spikes can be there on hot short means if the FET is enabled and the output is manually shorted or push button is used to short the output. Hot short by hand or push button causes a fast current edge and boucing. The internal sense FET is not reacting fast to fast edge because of bouncing.
    If the device is enabled and hot short occurs with bouncing, the part can reveal high current spike. There is no need to order another board. I believe the fix of this matter is to add an inductance of few 10 of micro henry. The inductance will smooth the egde and the bouncing and the sense FET will react correctly to hot short.

    Regards