0% found this document useful (0 votes)
20 views2 pages

TI Precision Labs - RS-485 Best Practices For Implementing RS-485 Transmission

Uploaded by

pokornypvl
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
20 views2 pages

TI Precision Labs - RS-485 Best Practices For Implementing RS-485 Transmission

Uploaded by

pokornypvl
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Texas Instruments - IT Dept.

| TI Precision Labs - RS-485: Best practices for implementing RS-485


transmission

[MUSIC PLAYING]

Welcome to Precision Labs. In this series, we'll discuss best practices in implementing RS-485 transmission. These topics include
network topologies, termination, external fail-safe, and grounding. Three common typologies which have disadvantages are the

star network, ring network, and the backbone with star network. The reason why both the star network and backbone with star

network are discouraged is because it purposely extends the length of the stub from the main transmission line.

The ring network is discouraged because reflections from nearby connection points are more likely to degrade the signal integrity
at the intended receiver and may also reflect back into the transmitter. If any of these topologies must be used, the data rate

should be drastically reduced to ensure reliable communication. The RS-485 standards suggest that the nodes be networked in a

Daisy chain. In this topology, the participating drivers, receivers, and transceivers connect to a main cable trunk via short network
stubs.

The interface bus can be designed for full-duplex or half-duplex transmission. An alternative to the Daisy chain topology is the

junction box approach. This should be implemented in a way to minimize stub links and thus reduce reflections. If a no-ideal

network topology must be used, it is best to insert a repeater node at each junction. The stub length in this diagram is defined as

the T junction from where three lines intersect to the device pins.

For the Daisy chain topology, this occurs on the PCB board where the two joining transmission cables and the traces to the RS-485

transceiver. For the junction box approach, this occurs at the main transmission wire and the joining wires into the device spins.
Because the Daisy chain topology allows for much smaller stub links to be generated, this helps to reduce the reflections into the

main transmission line, which is the main reason why this topology is preferred.

The junction box approach can also work well. But ideally, the length of the stub should be minimized to reduce reflection. In cases

where the stub length cannot be reduced to a reasonable length, the data rate may need to be slowed down in order to reliably

communicate. This is a general rule of thumb, not a firm rule. The maximum acceptable stub length can be calculated using the

equation provided. tr is the RS-485 driver's rise time and is found in the datasheet of the device. This is measured from 10% of

VCC to 90% of VCC.

V is the signal velocity, which is a function of the dielectric material of the transmission medium or cable. This information can

typically be found in the datasheet of the transmission cable. This measurement is expressed in terms of percentage of a factor of

the speed of light. c is the constant for the speed of light. Terminate the two furthest points of the bus with resistors equal to the

characteristic impedance, z zero, of the transmission cable.

The termination resistance should match as closely as possible without dropping below 60 ohms, or 54 ohms if the transceiver's

loading is included in the calculations. External fail-safe resistors can increase noise immunity of the receiver under bus idle, short,

and open conditions. An app note on how to calculate the fail-safe resistors and termination resistors is linked below the video. So

how can you tell an RS-485 transceiver has a built-in fail-safe? There are two ways this can be checked.

The first way is to just check the front of the datasheet. This is typically a feature on the front. The second way is the electrical

parameter v input threshold. Plus and minus can be checked. A transceiver with fail-safe biasing typically will not center the

threshold values around 0 volts. The values expected from fail-safe biasing receivers should be negative values. Please note that

some of the newer RS-485 transceivers include circuitry that center its v input thresholds around 0 volts and can still provide fail-
safe biasing. This check method, though, should still work for a majority of RS-485 receivers.
Should I provide external biasing even if my transceivers all have fail-safe biasing? This is not necessary because all the

transceivers should still hold rx in a logic high state in the event of a short idle or open condition on the transmission line. If

external biasing is used, the receivers may see the benefit of additional noise immunity. The downside to the external biasing,

however, is increased power consumption because there is a path for current to flow from the VCC to ground with the fail-safe

biasing resistor at all times.

If one of my transceivers has a fail-safe feature, does that mean my network is fail-safe? If a transceiver features internal fail-safe

biasing, in a majority of cases, it applies only to that receiver. It won't carry forward to the other nodes in a network. This is

different from external biasing, which applies to the entire network even if placed at a single node. There are some exceptions,
however, in a few older transceivers.

Should I place an external fail-safe biasing network at all my receiver nodes? External fail-safe biasing is ideally only placed at one

node. If it is placed at each node, the pull-up and pull-down resistances need to be much higher. The drawback to this is that the

pull-up and pull-down strength is now a function of the number of nodes installed. There are a few different ways grounding can be

accomplished in an RS-485 system.

In example A, the system is vulnerable to high ground potential differences. If the ground potential difference is greater than the

limit of the device, it could stop working or even be damaged. In example B, if a high ground potential difference is present, a
large group of loop currents could form, which can be coupled into the data lines as common mode noise. Example C is

recommended by the RS-485 standard. Adding the series resistors lowers the loop current, but the noise could still exist.

When grounding an RS-485 system, the ground potential difference should not violate the common mode specification for the

transceiver, which the RS-485 standard specs to be at least negative 7 volts to positive 12 volts. Some transceivers support even

higher voltages from negative 20 volts to positive 25 volts. In cases where very large ground potential differences could occur, an

isolated RS-485 transceiver should be used.

You might also like