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.

LM5170EVM-BIDIR: Destruction of Current Measurement Path

Part Number: LM5170EVM-BIDIR
Other Parts Discussed in Thread: LM5170

Hello Expert-Team,

we are using the LM5170EVM-BIDIR-board to transfer energy from a 51V-source to a 28V-target and vice verses.

We have modified the regulator circuit in order to deliver such voltages.

At the 51V terminal there is a large inductivity which is only activated in buck mode.

The load at the target has a quite low impedance with 50mOhm.

We have configured  the overvoltage protection in that way, that switches at 55V and 31V.

At our first board it came to a destruction of the central ic. This case happens at an wanted activation of overvoltage protection of the 28V net.

We measured of impedance of 1R2 to ground at the net sw1,sw2,... 

We supposed the following root cause:

* An overvoltage at the target net was detected and caused the breaker unit to open 

* There was a massive current flow (30A) through the outer inductivity just before opening

* The inductivity drives the current despite of the open breaker unit

* This leads to an overvoltage in primary circuit which destroys the central ic.

We using a second board and try to protect it by:

* We build a stabilisation unit which is placed between outer inductivity a 51V input

* the stabilisation unit uses 6x 3900µF-Elko and 6x 22µF- ceramic capercitor

* We protected the stabilisation unit circuit by 2 clamping diode (2*24V)

*We protected the input of the board by a second set of clamping diodes (2x28V)

Than we change severall times between boost and buck. The currents varies (primary  from 30A to 1A and secondary from 50A to 1A.

It is a normal use case that the overvoltage circuits firing from time to time.

Unfortunalety  this board was destroyed within short time too. 

We conduct an investigation and find out that the terminals CSW1 and CSW2 has a 2R2 Resistance against ground,

This lead us to the following idea:

* The firing of the breaker unit leads to a under- ore overvoltage.

* The internal clamping diode at 28V-net is not sufficient that.

I have 3 questions:

Could our suspicion be true?

Would it be a good solution to protect CSWx and CSWx terminal against over- and undervoltage? A  low capacitive diode nearby the the terminals to GND and to 51V?

Is is reasonable to arrange a capacitor (470µF) parallel to clamping diode?

We have here the opportunity to exchange the LM5170 on the board. Unfortunalety the are no LM5170 on the market now. Would you be so kind  to supply us with 2 samples?

Thanks for your help

Kristian

  • Hi Kristian,

    There are some points that I'm not clear:

    • At the 51V terminal there is a large inductivity which is only activated in buck mode.

                Question: Can you draw a schematic to show the position of the inductor? What is the inductance?)

    • An overvoltage at the target net was detected and caused the breaker unit to open 

                Question: BRKG would not go low if OVP is triggered.

    • * We protected the stabilisation unit circuit by 2 clamping diode (2*24V) *We protected the input of the board by a second set of clamping diodes (2x28V) 

      Question: The model of the clamping diodes?

    • We conduct an investigation and find out that the terminals CSW1 and CSW2 has a 2R2 Resistance against ground,

                Question: I suppose you are referring to CSA1 and CSA2? CSA1 and CSA2 is protected by D3 on EVM. IS D3 damaged?

    Best Regards,

    Feng Ji

  • Hi Kristian,

    Since it has been a while since we last communicated, I am going to close this thread. If you have further questions, please feel free to open a new thread.

    Best Regards,

    Feng Ji

  • Hi Feng,

    thank you for your support.

    We have solved the problem thanks to your hints.

    The issue was caused by a volatile voltage at the high-supply input terminal. From time to time it fall below 5V plus the voltage at the output terminal. This leads to a unwanted open breaker.

    Unfortunalety there was no active undervoltage detection to prevent this case.

    Now we have sharpened the undervoltage detection.

    Best regards

    Kristian