OPEN-CIRCUIT FAIL-SAFE

Hi Julian,

Open failsafe means no connection between A and B inputs, no termination resistor, not shorted, or connections to other notes. It is not a true failsafe. Your waveforms look like the A and B were shorted together. The output is undefined when Va-Vb=0V falls in the +/-200mV range. You can add external failsafe biasing resistors to product true failsafe.

Best,

Jane

Hi YaoJ,

Update: Our A/B bus uses two 120 ohm terminal connections.

If Va-Vb=0V is in the undefined region, why does the false pulse only happen when the A/B bus goes from logical “high” to “undefined”, but not when the A/B bus goes from logical “low” to “undefined”?

Also, why does the false pulse that occurs when moving to the “undefined” area automatically revert to logic “high” after 500ns? We have measured this many times with the same result.

If the main reason for the false pulse is that it's in the “undefined” zone, then the RO state should change randomly?

Thanks.

Alan

Hi Alan,

My apologies for taking so long to get back to you. This is an old part, and the failsafe function could have been updated. Please allow few more days and let us look up for the files. Could you please provide me the complete part number and the top mark?

Best,

Jane

Hi YaoJ,

Thank you for your reply. The part number is ADM3485EARZ-REEL7. Top mark is ADM3485EARZ#228

Alan

Hi Alan,

Our team has done an extensive check with the date code your provided along with our design files. The final confirmation is the ADM3485 datasheet is update-to-day. The receiver has an open-circuit fail-safe, meaning pins A and B should have no connections (floating pin). Output RO can be high or low when the differential input Va -Vb = 0V, “undefined” input. The output can randomly change when Va -Vb = 0V, like the “500ns” pulldown glitch you observed. Since the part is with an undefined input, we cannot explain nor predict the signal on RO.

Best,

Jane

Hi Alan,

Our team has done an extensive check with the date code your provided along with our design files. The final confirmation is the ADM3485 datasheet is update-to-day. The receiver has an open-circuit fail-safe, meaning pins A and B should have no connections (floating pin). Output RO can be high or low when the differential input Va -Vb = 0V, “undefined” input. The output can randomly change when Va -Vb = 0V, like the “500ns” pulldown glitch you observed. Since the part is with an undefined input, we cannot explain nor predict the signal on RO.

Best,

Jane