SN65HVD1176: Part is Destroyed when Switching from 0V to -7V on Terminal A

Raphael,

The input voltage ranges of RS-485 transceivers show that the communication works with a wide common mode range. For example, if the voltage on A and B toggles between -6.8V to -7.0V (-6.9V common mode voltage), you expect the transceivers generate correct data. I'm not sure why you ramp up the supply after the voltage is applied to the wire. In some applications, an extra node is added to an existing bus. However usually the node supply is ready before the module connects to the main bus (hot plug-in). Please let me know if you have more questions about the spec or application. 

Regards,

Hao

Hi Hao,

It doesn't matter if I ramp up and the wire is already plugged in or if I start the supplies and plug in the wire. In both setup the transceiver is destroyed. Since our product (with your transceiver) has to be compliant with TIA-RS485 it doesn't matter for us if it is not a normal usecase as long as it is specified in the standard.

To clarify my setup, here a sketch:

In further tests I found out that it happens even more often when I apply first Va=12V, Vb=11.8V, then switch off the supplies and then apply Va=-7V and Vb =-6.8V.

Regards,

Raphael

Raphael,

The sketch helps. Is it possible for you to monitor the bus voltage in the tests? I suspect that somehow the voltage doesn't settle down as quickly as you expected. If overshoot or undershoot happened and the pin voltage went beyond the specified abs max, the device could be damaged.

In application, this is how the input range is tested. You start with the normal operation (for example: 5V supply, 100kbps PRBS data), then change the common mode of the bus to -7V or 12V to see if the communication is still successful.  

You can refer to figure 10 on page 11 of this datasheet for the common mode voltage check for the driver. I hope it be helpful.

www.ti.com/.../sllse11h.pdf

Please let me know if you have further questions.

Regards,

Hao

Hi Hao

Here the monitored voltages during the test:

I applied first 12V / 11.8V, then shut the supplies down and applied -7V / -6.8V and the -7V supply triggers the current limiter since I have a short between terminal A and GND. Vb goes up approx. 1s after Va, since the power supply reacts slower. Nevertheless, I'm always within +-12V Vdiff (Va - Vb) and this should not harm the transceiver.

Compared to Figure 10 of the data sheet I don't have the 375Ohm series resistors, but those are not specified in TIA-485-A either. What  voltage steps are applied to the transceiver in Figure 10? Since it is only written -7V < V(test) < 12V.

Thanks for your help

Regards,

Raphael

Raphael,

Thanks for your information. I was asking if you saved the waveforms in the tests that destroyed the parts. For example, apply -7V to terminal A and -6.8V to terminal B, then ramp up Vcc with 70mV/us rate to 5V. Can you also confirm the state of logic pins, like D=0V, DE=0V, etc? If you want, we can discuss the issue directly with the help of TI field application engineer.

Regards,

Hao