J1708 5V Supply Protection

Hello Tony,

1. Lets consider the two cases which can occur on the bus: dominant or recessive

Recessive: For the J1708 standard, the recessive state occurs when the driver of the RS-485 transceiver is disabled. The A and B pins go into high impedance, and the bus voltages are biased to a positive differential logic level by R1 and R2. In this case, if the serial data bus is pulled to a higher voltage than +5V, there will be a potential drop from the serial data bus to the output of the buck converter (through 4.7k+47ohms). Since the bus pins are in a high impedance state, the path for current flow will be back into the buck converter (though there is some sizable resistance in the path). In this case you may need to supply the converter with some protection from reverse current. Since I don't know the characteristics of your buck converter I can't give a definite answer on that but the question that should be asked is: can my buck converter survive  around 1.47mA of reverse current?

I got that number from:

V = 12-5 = 7V

R = 4.7k + 47 = 4747 ohms

I = V/R = 1.47mA

Of course this is calculated for a DC short since the caps will block the DC path to ground. For short transients, those caps will provide a low impedance path to shunt energy away from your buck converter.

Dominant: In the dominant state things will be a little more complicated. The driver will be enabled and will pull the bus lines to a negative differential state through a Full H Bridge like this:

In this case, there should be a low impedance path from the A line to ground through the device, so as long as the RS-485 transceiver is rated for this condition, you should be ok. 

Enabling and disabling the receiver will have no effect on this issue. 

2. As for standby mode, there is no reason why this setup would prohibit you from using the feature. 

To sum it all up, there is a possibility of reverse current at the buck converter output which may or may not be a problem based on that part. If the device is sensitive to reverse current, it might be a good idea to look into a protection scheme. 

Also as a side note, I noticed that your scheme matches the node load circuit in this app note: 

Except the capacitors are much larger in your schematic (2.2mF vs. 2.2nF). I'm not sure if that was intended or not. 

Best Regards,

Casey 

Hi Casey,

Thank you for the very good explanation.  Let's consider a system with a 24V battery:

1.  Recessive: According to the equation above, this would require my buck converter to sink 4mA of reverse current.  We plan on using the LMR16006Y but couldn't find anywhere a "Maximum reverse current" spec, is there another term we can look for?

2.  Dominant: Consider the worst case where the 24V Battery is shorted to the A line. If we then try to drive this to ground through the transceiver (dominant state), the only thing limiting current is the 47 Ohm resistor, is that correct?  If so, we then have a steady state current of 511mA flowing through the resistor and the device. The SN65HVD1780 has a 200mA "Driver short-circuit output current", which is much less then the 511mA, will it survive?  This would also require the 47 Ohm resistor to be rated for 12.2W, whoa!  

You are correct, those caps should be 2.2nF.

Best wishes,

Tony

Hello Tony,

1. I wish I could help with this question, but unfortunately I'm not very familiar with the internal topology of the buck converter, so this might be a better question for the team responsible for that device. I don't know how the buck converter will respond to a higher-than-expected potential at it's output. 

2. For the SN65HVD1780, the transceiver is guaranteed to survive a short to at least 70V on the A line while driving a dominant state. The short circuit output current is a parameter of the driver output, while the larger current you are referring to would be sourced from the 24V battery. Now, as for the actual current going through the resistor, you may actually see less since the diagram is simplified and there are other resistances in the path like the on-state resistance of the H-Bridge transistor and there will be a drop across the protection diode too. If you would like, I can set this up in the lab and measure the voltage drop across the 47 ohm resistor during this condition which will give you the current flow too. This would give a feeling for the power rating required for that resistor. 

Best Regards,

Casey 

Hi Casey / Tony,

Here is the link to the forum that support the simple switcher: e2e.ti.com/.../858

Hi Tony,

Yes, I was using the SN65HVD1780 for these tests and the highest DC voltage I tested to was +70V. The highest voltage drops across the resistor are actually around 15V DC before the device starts to significantly limit current. At 70V, the drop across the resistor was only around 1.8V because the transceiver's current limit is lower.

The SN65HVD1780-Q1 will behave the same way.

Best Regards,
Casey