SN65176B: transient max value

Hi Emma,

You are going to need to add protection to protect against the transient signal shown. This can cause damage. This device doesn't have a pulse voltage rating (some of our devices do - but more often than not we don't include it) 

(please note that this diagram doesn't show termination - but there should still be 2 terms in the RS-485 system)

Using a TVS + pulse proof resistors (10 to 22 Ohms typically - but 10 is more common)  will protect against the transient  - the above circuit we do have a parts list

But you may not need as robust of a diode - as this transient doesn't look like its 500W - but if you place something like this it will help clamp down on the voltages. 

Best,

Parker Dodson

Transient spikes are allowed between −10 V and 15 V.

There are transceivers with better protection like the SN65HVD21 or THVD2450.

Hi Emma,

So I do have a few notes. 

The actual schematic architecture looks okay (placement of protection and resistors) - but the actual values/components chosen aren't ideal. 

1. The protection circuits are using thyrsitors - and my biggest issue is that the model that you are using that the voltage can become rather larger before they turn on  - which can be close to 25V. I think it would be better to use a surge diode that will clamp at a lower voltage based on the current. The high potential voltage possible over the thyristors is enough to cause damage. The surge diode in the example circuit I showed realistically is going to require quite a large current to push the voltage higher and it will be less than 25V. 

2. The 10 ohm resistors - these need to be thick film or wire wound so they are pulse proof. This helps filter off the peaks of a transient - and standard resistors are not as helpful here - i am not sure what type of resistor is being used here but I would suggest wire wound/thick film for the resistors here - since they have a large package (1206) you should probably be able to fit something similar to what I suggest here if it isn't already. 

3. You are using partial termination in the system - this can result in a large signal at the input of the A and B pins 0 - as well as having the issue of not really giving the benefit of a terminated bus - assuming the transmission line has a characteristic impedance of ~120 Ohms will yield a signal reflection of ~78.5% of the incident signal so its not adding a lot of benefit for the partial networking - if possible using 120 Ohms can also help as the voltage signal won't be as large between A and B (therefore A and B will also most likely not be as large)

Those are my main concerns - with point number 1 being the most critical - I don't think the thyrsistors (from datasheet or images that you have shared) are providing enough protection on the line - and the difference between its peak and working voltages is very large so I am more skeptical of using these in place of typical surge diodes. 

Clemens was correct however, this can largely be avoided if you switch devices to something like the THVD24XX line of devices which just have +/-70V fault tolerant bus inputs.

Ultimately it looks as if you have two main options:

1. Increase the robustness of the protection circuit - the placement of the parts is correct - but there are a few value changes - especially on the diode/thyristor that should be made.

or.

2. Switch transceiver to a higher fault tolerant device.

Please let me know if you have any other questions!

Best,

Parker Dodson