DRV8320R: DRV8320R how much rining is good enough?

Hello, and thanks for your question.

In the case of high current & fast switching, the SHx voltage can exceed VM because the high-side MOSFET body diode takes some time to conduct. This can also happen on the low-side resulting in a negative voltage spike (SHx goes below GND).

If the SHx voltage increases too far beyond VM, then current will flow from SHx into VCP through the gate driver body diodes. This can cause damage if not controlled. As long as this voltage is less than 7V @ 200ns, it is within the abs max rating of the device.

The device should not report a fault unless damage to the charge pump is caused by this operation.

Adding snubbers across the high-side and low-side MOSFET can help improve this, but likely will not eliminate the positive and negative voltage spikes.

Please see this article for further details on parasitic effects on high current motor drive systems: LINK

Thanks,

Matt

Hi Matt,

So, some ringing is normal behaviour? I might reduce the ringing even more by choose a lower gate drive current, but I think with 60/120mA for a 11nC Cgd it is low enough. From the gate voltage pictures below, ou can see, there is some ringing there, that also seems to cause the rining on the output. At the moment, fully turn on needs around 300ns, or am I wrong?

Do you know from another project, what normal ringing would be with a similar setup (11-14nC Cgd, 60/120mA gate current, 48VDC supply, 20A motor current)?

The diode has a reverse recovery time of 40-80ns, might this be too high?

Hi Matt,

Ok, thank you for your reply. After some further research I see, that ringing is some normal behaviour and especially in DC/DC converters, there is in most cases a lot of ringing ongoing through the fast switching.

Now I have a further questions, this +7VDC threshold to prevent the charge pump from damage, is this only a problem for the smart gate drive ICs or do other high-side gate driver also have somehow this threshold (For example the DRV8301 motor driver seems not to have a threshold for the drain to source voltage of the high-side MOSFET)? 

You wrote in your first reply the +7VDC @200ns, what if the duration is much shorter (20ns), but it goes to +10VDC?

Thanks,

Sebastian

Hello Sebastian,

You are correct. The +7V threshold is to prevent current from flowing back into the charge pump. While it may be nominally running at ~10V, it is possible for the VCP voltage to drop to easily 7V during normal operation. Practically, as long as SHx stays less than 2 diode drops below the actual charge pump voltage, it will prevent any current  flowing from SHx to VCP. There is a time-voltage relationship here, but I prefer to avoid pushing current through this path in general.

This is a specific restriction on charge-pump based drivers. Bootstrap-based drivers do not have the same restriction since the entire high-side gate driver is "floating" on SHx due to the bootstrap capacitor. However bootstrap-based drivers have a disadvantage with negative voltage spikes, because it can over-charge the bootstrap capacitor and lead to damage that way.

Thanks,

Matt

Hi Sebastian,

Is there any more help that you need on this thread, or may I close it?

Thanks,

Matt

Hi Sebastian,

I will go ahead and close this thread. Feel free to ask a new question on the forum!

You can click the "Ask a related question" button at the top right of the screen.

Thanks,

Matt