DS3695A: Does this transceiver need external biasing on the differential pair outputs?

Hi Daniel,

thanks for reaching out, this is a good question. To determine if a passive fail safe resistor network is needed, a look at the differential Input threshold voltage of the receiver tells you if you need a external biasing.

So lets take a look at the DSA3695A datasheet. At the section Electrical Characteristics on page 3 you'll find:

VTH Differential Input Threshold Voltage for Receiver  −0.2 V to +0.2 V -> this means any voltage between -0.2 V to +0.2 V is undefined and a fail safe biasing is necessary like shown on your picture.

If yo look a more modern device like the THVD15xx you can see in section 7.6 Electrical Characteristics in the datasheet Positive-going input threshold voltage is -20 mV and Negative-going input threshold voltage -200 mV so its always below 0 V which indicates that the device has a internal fail safe biasing.

In order to design the passive fail safe resistor network for a 5 V RS-485 system you can go on with the values shown on your picture if you wan't to make it work quickly. For more details please take a look at this application note:

 /cfs-file/__key/communityserver-discussions-components-files/138/7127.7824.slyt324.pdf  

Additional info also here:

https://e2e.ti.com/support/interface/industrial_interface/w/industrial_interface/2575.the-unit-load-in-rs485?keyMatch=rs-485%20unit%20load&tisearch=Search-EN-Everything

/cfs-file/__key/communityserver-discussions-components-files/138/8228.RS485Infographic.pdf

Please let me know if that helps you out, don't hesitate to ask further questions!

Kind regards 

Dierk

Dierk,

Thanks so much for your input. You made that very clear.

However, adding the external basing didn't help the way I expected it to. I'm still having intermittent communication issues.

Here's an example of what I'm trying to do:

I'm trying to convert the output of the SN65HVD63 to RS485. Everything I'm reading says I'm doing it correctly but when I add the external biasing on the DS3695 communication fails. When I measure the differential pair it looks fine though. 

Thanks again,

Daniel

 

Hi Daniel,

do you operate with an open loop like shown on your schematic, if yes maybe you have forgot about the other 120 Ohm termination resistor at the end of the Bus-Cable.

If you have not connected (nodes) and a cable with 120 Ohm termination at the AB output, you have to change the 120 Ohm resistor between pin 7 and pin 6 to 60 Ohms. Other than that the schematic seems to be good.

If this is nor solving your problems we have to check your signal wave-forms.

Kind regards
Dierk

Dierk,

I'm not operating with an open loop. It's connected to another design using the same DS3695A via a small cable. But I see now that the other design does not have external biasing resistors populated. I believe it should. Is that correct?

However, I was taking measurements with an open loop. When transmitting without bias resistors populated I get the following differential pair:

This is what made me believe I needed the biasing resistors in the first place. 

However, when I populate the bias resistors I don't get a differential pair output. Looking at the logic side of the DS3695A I see the following before any attempt to send data:

R0 - Low (0V)

DI - High (5V)

RE/DE - Low (0V)

This tells me the DS3695A thinks it is seeing data input on A and B even though the connection is open. Can you help me understand why that might be? 

Thanks,

Daniel