Bias enquiry

Hi Colbee,

Thank you for your request - I would be happy to help.

Unlike a standard RS-485 transceiver, the pull-up and pull-down on the A and B terminals of the MAX13487E are to help set the pin-state when the driver goes to a high impedance state.  This happens, for example, when DI goes high.  When DI goes high, the device drives the bus for a shirt time and then disables the driver and allows the external resistors to hold the bus in the high state.

So, do you need a pair of pull-up/-down resistors for each transceiver on the bus?  The short answer is maybe.  If the distance between nodes is fairly short, then you probably don't need these resistors at every node. If, however, you have transceivers that are large distances apart - so that the cable parameters become an issue - then you likely want a resistor pair at each node to ensure the state of the bus as soon as possible after DI goes high.

Thank you,

Shasta

Thank you for your reply. I am operating at the extents of the line length for the spec at around 1km. I did find my bias values to be in the indeterminate state at the farthest node. I find the low bias values suggested problematic (220r) with the farthest driver struggling to pull down the line. With around 1K at each node I am sitting around 280mV I think my best strategy is to bias each node with the 1k locally.

Hi Colbee,

This sounds completely reasonable.  Lower value resistors can result in a slightly faster pull-up time for the bits, but if your data rate can sustain the slew with 1kohm - and you still meet the >200mV difference when the bus is idle - then this is a sound solution.

Thank you,

Shasta