AUIR3241S for battery charge/discharge switch

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
xiaona
Level 1
Level 1
First reply posted First question asked Welcome!

Hello, I am  using AUIR3241 for the first time, and I have  some problems  while using it.

A power board circuit for discharge-800A  control was constructed using the AUIR3241S gate driver IC and 65V class MOSFET, as below.

 

power board.png

the "VBAT" is connected to a 28V battery pack through the connector "U1", "U3", "U5"  , "Vin" is connected to the load,  "BATTERY_SWITCH_MOS1" is  connected to control board,  "R16" and "D9" was not used.

 

 

A control board circuit for charge-12A is as below:

control board1.pngcontrol board2.png the nets  "CHARGER-CONTROL" and "CHARGER-CONTROL-MOS-MCU" were connected to STM32.

The issue  happened when discharge began,  two nets "CHARGER-CONTROL" and "CHARGER-CONTROL-MOS-MCU" were set low, so the U12(MCM4020) and U20(AUIR3241) didnot work. First , we turned on the battery , set the load current to 800A, then turned on the load , but the two  diode "D5" and  "D6" on power board, was on fire at once, and we quickly turned off all the power .After check, we found some burn marks on the pin7 of U20(AUIR3241) on control board, and within expectation, this 3241 was damaged ,the pin 6(GATE) was always low.

If we replace the new 3241,and changed the load current down to 5A/10A/20A, the chip can work normally.

So  we want to know  where  is our circuit problem located, so we  can  correct it , and how can we promote the circuit ?

0 Likes
1 Solution
Pablo_EG
Moderator
Moderator
Moderator
First question asked 250 sign-ins 250 replies posted

Hello Xiaona,

Thank you for posting on Infineon Community.

We understand that you are testing a load switch for an 800A 28V battery system.

Let us know if our understanding is correct about the failure condition:
-Battery is connected.
-28V 800A load is turned on.
-Gate signal is turned on.
-D5 and D6 break down.
-Gate driver breaks down and pin 7 gets damaged.
-This does not happen in the 12A version.

Are all of the above correct?

If so, there is a high chance that the VCC pin experienced an overvoltage event.
Due to the high current di/dt, there may have been an induced voltage spike due to the parasitics of the board.
This would explain VCC breaking down and subsequently D5 and D6 breaking down.

We can test this by doing the following:
-Start the load at a low current (such as 1A).
-Slowly increase the load current, all the way to 800A.

If the issue is not seen, it tells us that the di/dt is the source of the problem.

In order to avoid voltage spikes, you could add zenner diodes from drain to source.
Please, refer to the following evaluation board schematic.
https://www.infineon.com/dgdl/Infineon-24V-ADR-Switch-Demonstrator-Board-User-Manual-UserManual-v01_...
Here, four 33V high-power zenner diodes are employed to clamp the voltage.

As an additional note, your schematic has 1k gate resistors for the MOSFETs.
When paralleling any power switch, much care needs to be taken so that the switches turn on/off at the same time.
The higher the gate resistance, the bigger the turn on/off matching will be, and current imbalances increase.
This can lead to overload and breakdown of the MOSFET.
Please be aware that this could become an issue.

Best regards,
Pablo

View solution in original post

0 Likes
4 Replies
Pablo_EG
Moderator
Moderator
Moderator
First question asked 250 sign-ins 250 replies posted

Hello Xiaona,

Thank you for posting on Infineon Community.

We understand that you are testing a load switch for an 800A 28V battery system.

Let us know if our understanding is correct about the failure condition:
-Battery is connected.
-28V 800A load is turned on.
-Gate signal is turned on.
-D5 and D6 break down.
-Gate driver breaks down and pin 7 gets damaged.
-This does not happen in the 12A version.

Are all of the above correct?

If so, there is a high chance that the VCC pin experienced an overvoltage event.
Due to the high current di/dt, there may have been an induced voltage spike due to the parasitics of the board.
This would explain VCC breaking down and subsequently D5 and D6 breaking down.

We can test this by doing the following:
-Start the load at a low current (such as 1A).
-Slowly increase the load current, all the way to 800A.

If the issue is not seen, it tells us that the di/dt is the source of the problem.

In order to avoid voltage spikes, you could add zenner diodes from drain to source.
Please, refer to the following evaluation board schematic.
https://www.infineon.com/dgdl/Infineon-24V-ADR-Switch-Demonstrator-Board-User-Manual-UserManual-v01_...
Here, four 33V high-power zenner diodes are employed to clamp the voltage.

As an additional note, your schematic has 1k gate resistors for the MOSFETs.
When paralleling any power switch, much care needs to be taken so that the switches turn on/off at the same time.
The higher the gate resistance, the bigger the turn on/off matching will be, and current imbalances increase.
This can lead to overload and breakdown of the MOSFET.
Please be aware that this could become an issue.

Best regards,
Pablo

0 Likes

Hi, Pablo, thanks for your reply.

the failure condition was right, I'll try the test as you told.

And at last, you said that "The higher the gate resistance, the bigger the turn on/off matching will be, and current imbalances increase."  honestly, I'm afraid I'm not quite with you, especially what does the "turn on/off matching" mean?

Best regards,

Xiaona

0 Likes
Pablo_EG
Moderator
Moderator
Moderator
First question asked 250 sign-ins 250 replies posted

Hello Xiaona,

Thank you for your reply.

My last comment referred to the time lag from when one switch turns on/off until the second switch turns on/off.
If this time lag is big, one switch can be conducting more current than the other.
This is an undesirable behavior.

However, as the current application is a charge/discharge switch, the switching frequency will be very low.
Therefore the current imbalance will rarely happen so you can disregard my comment.

Best regards,
Pablo

0 Likes

oh, I see, I think I understand what you mean.

thanks for your patience and help.

BR,

Xiaona

0 Likes