SN65HVD3082E: Regarding the cause of damage to SN65HVD3082E

Hi Masanori,

So, the driver output is RS-485 compliant which means that an output short circuit (A or B shorted to output voltage of -7V to 12V) will be limited to no more than +/-250mA. The driver has a current limit on its output in cases of short circuit. 

Damage can arise from a short circuit if the self-heating from the short circuit current causes the internal die to break. This device has max junction temperature of 150C - and the "D" package that is being used has a typical thermal to ambient impedance of 116.7 C/W - so it is possible during short to burn a lot of power that could cause the issue with self heating. 

During the failed test what was the ambient temperature? 

Does you have another unit from the same lot as the failed one? I want to ask because I would like to see a picture of the top markings (I can't really see the marking well on the failed unit you have attached because the burn damage is pretty bad) . This will allow me to trace the part back in our system to check if it is legitimate or not. 

Please let me know so I can dig into this issue a bit more - but essentially it is possible for failure due to self-heating of device die during short circuit, but I want to verify test conditions and be able to trace to the lot code of the devices to check for legitimacy. If device is legitimate then we may need to look into adding heat sinks or other thermal performance improvement techniques to prevent the burn damage seen. 

Best,

Parker Dodson.

I now understand the destruction of the internal die due to self-heating.

During the failed test what was the ambient temperature?
→20-25℃ environment.

Does you have another unit from the same lot as the failed one?
→The devices used on side B of the first attached document are separate units and are equipped with the same lot.

I want to ask because I would like to see a picture of the top markings (I can't really see the marking well on the failed unit you have attached because the burn damage is pretty bad) .
→The third line appears to be stamped with A904.It's not very clear, but I'll attach a photo.
SN65HVD3082E marking photo.pdf
If device is legitimate then we may need to look into adding heat sinks or other thermal performance improvement techniques to prevent the burn damage seen.
→The explanation was insufficient.
It has been in production for many years without any problems, and recently one of these was damaged during a post-manufacturing communication check, so I asked for a technical opinion.

→The device was purchased through an authorized route, and we have manufactured many units of this device in the past.
We suspected a manufacturing defect, so we checked to see if there was a possibility that the device would malfunction if there was a solder ball between pins 7 and 8.

I have an additional question, is this lot compatible with RoHS10?