SN65HVD485E: Fail safe doesn't work

Hi Hiroshi,

If the differential bus in the idle state is lowering the bus voltage (A-B) to -0.8V which is valid logic low - the receiver is working as spec'd. 

In both scope shots (one that worked and one that didn't) there is a dip in the A-B voltage. What is changing in the system at this point - since it seems like something is causing this dip?  How are the enables being controlled in during these tests? What is the transmitting device - only the receiver is labeled - please let me know?

The next two steps are as follows:

1. Identify what is causing the dip in the A/B voltage as this low level is causing the false start bit because the receiver is reading a valid logic low signal. With the information to the questions above it could help shed light to where this dip is coming from and if it is resolvable. That is a large voltage dip so I'd imagine it is probably  not expected. I'd also still add external bias as there could be as little as only 10mv of margin for the receiver at a 0V state. 

2.  (Optional but highly recommended) add fail-safe bias resistors  that set the idle voltage to be around 200mV - https://www.ti.com/lit/an/slyt324/slyt324.pdf (app note for how to size resistors - assume both terminations are equal at 120 for an easier process). This will affect the entire bus - not just the single device - as fail-safe devices just have both thresholds <0V - it doesn't bias the bus - whereas two resistors (pull up on A and pull-down on B) will bias the bus in an idle state that's why we still often recommend external bias on our fail-safe parts as in systems with multiple transceiver IC's not all of them will be fail-safe, the threshold varies, and/or noise margins are low - I still think its important to understand why the dip went to -0.8V - if its from a noise signal coupled in or something with the transmitter - as that is a large transient and that should be the first priority.

Rephrasing my questions from above:

1. What are the enable signals (DE and /RE)  of the receiver look like during these tests - I see R,D, A, and B but the enable signals will help to see if that is changing the bus voltage as well - essentially if I can see DE, /RE, A, and B on the same scope shot so I can see if these enables are affecting the A/B voltage. 

2. What is the transmitter for the system being used - if it isn't the same part that's okay but it could be partially to blame for the issue but I want to verify its specifications to see what impact the device could have.

3. Besides results - where there any differences in the test setup between the good shot and bad shot tests? 

4. Are there only 2 transceivers on the bus or is there more?

5. Is the system terminated? If yes - with what resistances and where? 

6. Is there anything else on the bus besides the transceivers + termination (if present)

If you could please let me know I will be able to dig into your problem more and see if we can find a solution!

Best,

Parker Dodson 

Hi Parker,

Thank you for your detailed advice.
I asked our customer about your question.
I will come back again after their reply.

Best regards,
Hiroshi

Hi Hiroshi,

No problem! Please let me know their response when you receive it!

Thanks!

Parker

Hi Parker,

I received an answer from a customer.
However, the publication of the schematic was refused. So I will answer in the text below.

1. During the test, /RE  = L, DE : At the time of transmission = H, otherwise = L.
2. The transmitter uses the same circuit & same device(HVD485E) as the test target.
3. The test settings are the same.
4..here are only two transceivers on the bus (transmitter and test target).
5. It is not terminated.
6. Nothing is connected. RE = GND, DE = 10k ohm Pull down.

Since the bus line length is short and it is a one-to-one connection, I think that the presence or absence of a terminating resistor does not have much effect.
If you have any other problems, please tell us your opinion.

It's a very simple schematic, but please let us know if you need it.
Please tell us your private email address and we will send it to you.

Best regards,

Hiroshi