TCAN1042HV: Bus Fault protection ±70 V condition

Hi Klaus,

I can confirm that the units pictured here are 'H' units and are rated for ±70V bus faults. 

Can you share your test procedure where this unit was damaged?

Was the power supply active before the connection to the CAN bus made? We have seen in some tests that the transient from connecting an active voltage line to the bus with no series impedance can lead to much higher voltage transients at the device pin that can lead to damage. The recommended test condition is to apply the fault before the voltage source is active so that the bus can be slowly brought up to the fault voltage to avoid these transients. You may also choose to include series impedance on the short to simulate a cable fault which will similarly slow any transient caused by the hardware connection.

Is the transceiver set up the transmit on the CAN bus during the fault? If the CAN controller continuously drive data instead of going bus off as expected, the transceiver may overheat from the fault current and lead to damage. Make sure that the CAN controller is capable of going into the bus off state when the CAN fault is applied. 

Regards, 
Eric Schott 

Hi Eric,

thanks for your answer.

Where on the component itself can I see whether it is an "H" device?

I repeated the test a third time with the same test setup. I used a device from a PCB which was assembled about 2 years ago. I assembled it on the same PCB. With this device it worked. Please see following photo:

I can see that the device marking looks a little bit different.

In the following plot you can see the voltage on the CAN lines:

Are theses the same devices?

Yes the power supply was active before the connection to the CAN bus made.

The transceiver is assembled on a BMS for battery protection. The BMS is connected to the battery. The BMS is in sleep mode, so the transceiver is not supplied. Then I use a voltage supply which is switch on and connect it in short times to on of the CAN Lines with respect to GND. First it was the CAN Low line. After the first connection the transeiver was defect.

Please see the Osziplot:

The yellow line is CAN Low and the green line is CAN H.

Thanks for helping.

Klaus

Hi Eric,

thanks for your answer.

Ok. Am I also able to track the lot trace code or do I have to ask for it here in future?

I am going to repeat the test like you have written and come here again.
Also I will try to plot possible transients.

Thanks.
Klaus

Hi Klaus,

Unfortunately lot tracking is only an internal tool. This is important for us to limit access to some information that could be used for counterfeit activity so we are not able to make this information public. You can post here on E2E in the future if you need us to check some information for you but note this should only be used to identify lots that you are unsure of or currently experiencing an issue with. We cannot simply validate lot codes for the same reason we have to keep this information internal. 

Please keep us updated on the results of your testing. I hope this round goes smoothly.

Regards, 
Eric Schott 

Hi Eric,

Thanks for your answer.

I think you were right.

I repeated the test with two different power supply. And I used the OutputON Switch of the Power Supplies.

Here is a plot of the first one:

chanel1: CAN_H with releation to GND
channel2: CAN_L

And here the other one:

chanel1: CAN_H with releation to GND
channel2: CAN_L

The second one has a slower ramp. But it worked with both power supplies.

If I first switch on the power supply an than connect it to CAN_H ist looks like this:

chanel1: CAN_H with releation to GND
channel2: CAN_L

I madea second try on CAN_L and the TCAN was defective:

chanel1: CAN_H with releation to GND
channel2: CAN_L

You can see voltage transients with 100V. I think that is the reason for the defect.