DC Short Circuit, DCLink Voltage Reversal, and Wolfspeed SiC MOSFET Body Diode Failure

petercs

Do you recognize the problem with SiC MOSFET devices that they can fail from voltage reversal when the DC Link experiences a short circuit?

Please read the attached report.

TD006 DCShort Wolfspeed MOSFET Diode.docx

Regards, Peter Cox-Smith

Cox-Smith Consultants Ltd, UK.

Forum_Moderator

Thank you for your post, it has been approved and we will respond as soon as possible.

petercs

Thanks for looking at this issue. Can you please respond as I have more test results (10 units) that show failure of the body diode in Wolfspeed devices type CPM3 002112 0D ? I need to respond to my client next week and it would be valuable to include Wolfspeed's opinion.

Regards, Peter

Peter Cox-Smith, Cox-Smith Consultants Ltd.

TBhatia

Hello petercs,

Can you please confirm if you are using the bare die CPM3-1200-0021A or the discrete device C3M0021120D? It is unclear from your given part number. Thank you!

petercs

Hi TBhatia

The tests were carried out using the TO247 discrete devices, although there is a plan to use the bare die in the final product.

Peter

FrankWei

Hi Peter,

Thank you for your feedback. How many SiC MOSFET dies for each switch position in this traction inverter? the reverse current should be shared by the dies, right? and the current should go through two switches in serials. on the other hand, how many pcs of C3M0021120D are applied in the test？ a single device C3M0021120D may not be able to survive from the current stress as mentioned in the test report. Thank you

Regards,

Frank Wei

petercs

Thanks Frank, I will reply at length later this week…Peter

Peter Cox-Smith

Cox-Smith Consultants Ltd

petercs

Hi Frank

My client is an aerospace company designing an electrical system for an aircraft, including the main generator, which will have an AC to DC structure with a DC bus. Their concern is that a short circuit of the DC bus will cause voltage reversal, and stress, and failure to the body diodes. Failure is not acceptable in an aircraft. Their plan is to use WolfSpeed SiC MOSFETs in the converter and, as you suggest, there will be multi-chip switches; the components I have tested are single devices with the same characteristic as the individual chips. The client expects a wide range of characteristics of the individual chips and expect one device to carry most of the fault current. For example, I have tested 10 devices to failure. The voltage across the body diode at failure varied from 24V to 72V and the I squared t from 92.7 to 182.1.

The design project must specify a protection circuit that can be qualified (to aerospace requirements) to always protect the body diodes, as there is no guarantee that multi-chip switches will always survive. In tests, I have included a damping circuit that prevents or limits the voltage reversal and all WolfSpeed devices survived the DC Short Circuit test with damping present. However all such circuits increase dissipation and/or weight, both undesirable in aircraft, but possibly essential.

Alternatively we could accept the voltage reversal and add a low volts drop high pulse capacity diode to take the current and limit the voltage on the body diodes. The trick here is to find a suitable diode. Does WolfSpeed offer a suitable diode?

I look forward to your further thoughts, Peter

Peter Cox-Smith

FrankWei

Hi Peter,

Thank you for your sharing! The diode will not carry current during normal operation. So a high current rating Si rectifier should be cost-effective. You may try this part. vs-80apf1-m3series.pdf or you can look for a diode with even higher current rating. Thanks

Regards,

Frank