LM5143-Q1: Design issue for LM5143

Zhang Liang

Expert 2145 points

Part Number: LM5143-Q1
Other Parts Discussed in Thread: LM5143, , CSD18563Q5A

Hi,

The DCDC LM5143 is used for the main power supply.

The reference design is based on datasheet "8.3.17.2 Single-Output Interleaved Operation",the spec. as below:

1) Input voltage is 48V;

2) Output voltage adjustable, 42V MAX., use the potentiometer for the voltage adjust;

3) Output constant current 25A;

4) We use the 200KHz frequency, and modify the LO1 and LO2 from 0,68uH to 10uH;

4) The MOSFET is Infineon BSC123N08NS3G(80V, 12.3mohm, 5*6 package);

Now, we have below issue for the design:

1) The highside MOSFET broken down with load when the power on;

2) The highsiede MOSFET broken down without load when adjust the potentiometer from 0V to 42V;

Please refer to the schematic diagram and share your comment, thanks.

over 2 years ago

0 Timothy Hegarty over 2 years ago

TI__Guru 51057 points

Hi Zhang,

The MOSFET is not suitable as its Miller plateau is too high (it is not a lgic level FET with Rdson rated for Vgs = 4.5V and thus suitable for a 5V gate drive)

Note the Miller plateau from the gate charge curve is 5V, so there is no overdrive with a 5V gate drive amplitude.

Please send a completed quickstart calculator for this design so I can review further.

Regards,

Tim

+1 Timothy Hegarty over 2 years ago in reply to Timothy Hegarty

TI__Guru 51057 points

Just looking at the schematic - do you need the high value of electrolytic cap? This seems excessive given the high Vout requirement, makes compensation difficult, and requires a lot of current to ramp the output voltage during startup.

0 Zhang Liang over 2 years ago in reply to Timothy Hegarty

Expert 2145 points

Hi,

Our MOSFET BSC123N08NS3G which the Vgs=10V is same as the reference design(Infineon IPC50N04S5L-5R5), so please advise why we need to modify the Vgs to 4.5V logic gate drive, thanks.

0 Timothy Hegarty over 2 years ago in reply to Zhang Liang

TI__Guru 51057 points

Hi Zhang,

This MOSFET is completely different to the 40V part in the EVM. More specifically, the Rdson is not rated for Vgs =4.5V. This is required as the gate drive amplitude from the LM5143-Q1 is only 5V.

Regards,

Tim

0 Zhang Liang over 2 years ago in reply to Timothy Hegarty

Expert 2145 points

Hi Timothy,

Thanks, I think you mean we need to choose the logic level MOSFET for this design.

0 Timothy Hegarty over 2 years ago in reply to Zhang Liang

TI__Guru 51057 points

Yes, correct.

0 Zhang Liang over 2 years ago in reply to Timothy Hegarty

Expert 2145 points

Hi, Timothy,

Thanks for your professional commont, and we will choose the logic level mosfset to test again.

0 Timothy Hegarty over 2 years ago in reply to Zhang Liang

TI__Guru 51057 points

Ok, thank you. Just ensure that Rdson is rated at Vgs = 4.5V in the MOSFET datasheet. The MiIller plateau (flat line in the gate charge (Qg) curve should be less than ~3.5V.

0 Zhang Liang over 2 years ago in reply to Timothy Hegarty

Expert 2145 points

Hi,

We consider to use the MOSFET NCEP0140AG for the high/low side switch, please help to check if this is available, thanks.

The datasheet of NCEP0140AG for your evaluating:

NCEP0140AG.pdf

0 Timothy Hegarty over 2 years ago in reply to Zhang Liang

TI__Guru 51057 points

HiZhang,

This is a 100V FET with high Rdson and hih Qrr. If Vin-max is 48V, thenyou can use a 60V FET. This will give lower power loss. Take a look at TI's NexFET portfolio of 60V FETs. In partiicular, you can use the NexFET calculator tool here: https://www.ti.com/tool/SYNC-BUCK-FET-LOSS-CALC, For example, the CSD18563Q5A is a good general-purpose 60V FET.

Also, I recommend using the LM5143 quickstart calculator for your design to check losses, etc. https://www.ti.com/tool/LM5143DESIGN-CALC

Regards,

Tim

Power management

Power management forum

LM5143-Q1: Design issue for LM5143