Top 7 design questions about isolated RS-485 transceivers


By Anthony Viviano and Vikas Thawani

Looking for more information about RS-485? We’re here to help. Based on TI E2E™ Community feedback, we compiled a list of the most frequently asked questions about isolated RS-485 transceiver design challenges. We hope this list will provide useful insights when isolating signal and power for RS-485.

1. When is isolating an RS-485 bus necessary?

Isolation prevents direct current (DC) and unwanted alternating current (AC) between two parts of a system, while still enabling signal and power transfer between those two parts. Isolation typically protects electrical components or humans from dangerous voltage and current surges; isolation for human safety is called reinforced isolation. Isolation prevents ground loops which are common between nodes communicating at long distances. Isolation also allows communication between nodes with ground potential difference much higher than what the RS-485 standard recommends.

 2. How many nodes can you connect to an RS-485 bus?

To estimate the maximum number of bus loads possible, RS-485 specifies a hypothetical term of a unit load (UL), which represents a load impedance of approximately 12kΩ. The Telecommunications Industry Association/Electronic Industries Alliance (TIA/EIA) RS-485 standard mandates a maximum of 32 ULs that can be added  to an RS-485 bus. The UL of a node is calculated by the worst-case ratio of the input voltage divided by the leakage current, as shown in equation 1:

           

Once you’ve established the UL of the node, equation 2 calculates the maximum number of nodes:

 

Most of TI’s isolated RS-485 transceivers have a UL of 1/8, translating to a maximum of 256 nodes on an RS-485 bus.

For more information about UL and the number of nodes that can connect to an RS-485 bus, see the Analog Applications Journal article, “The RS-485 unit load and maximum number of bus connections.”

3. What is the speed vs. length correlation for isolated RS-485?

There is an inverse relationship between signal rate (speed) and cable length. The exact relationship depends on the resistance and inductance of the cable itself. When building an RS-485 network, the choice of cable can be as important as the transceiver, in order to ensure reliable communication over the necessary distance. Figure 1 is a graphical representation of the signal rate to cable length correlation. Region 3 of the figure shows the maximum cable length regardless of signal rate; in this area of the curve, the DC resistance of the cable causes attenuation in the signal and limits the maximum communication distance. Region 2 illustrates the inverse relationship between signal rate and cable length caused by transmission line losses, which will increase with a longer cable. In Region 1, you can neglect transmission line losses; the driver’s rise and fall times determine the main limitation on the maximum data rate. The RS-485 standard recommends a maximum signal rate of 10Mbps; however, with today’s technology, signal rates as high as 50Mbps are available.

Figure 1: Signal rate and cable length relationship

4. What is fail-safe biasing, and how do you design for it?

In order to comply with RS-485 standards, the receiver output must generate a logic level high  when the differential input (VID) is more than 200mV, and must output a logic level low when VID is less than -200mV. However, you can generate an invalid output under three scenarios:

  • Open bus conditions, such as a broken cable or a disconnected connector.
  • Shorted bus conditions, such as a cable insulation breakdown shorting the twisted pair.
  • Idle bus conditions that occur when no driver on the bus is actively driving.

In any of these scenarios, for a terminated transmission line, the RS-485 receiver VID would be zero and a non-fail-safe receiver output would be indeterminate.

Fail-safe biasing provides a differential voltage to the idle bus in order to maintain the receiver at a logic high level. If you don’t take fail-safe biasing into account, the termination resistors can lower the bus voltage to 0V, leading to an incorrect output or signal oscillation. You can design fail-safe biasing by using a resistor network along with an RS-485 transceiver. TI’s isolated RS-485 transceivers all have integrated fail-safe basing ]  for open, short or idle bus situations, eliminating the need for external circuitry to achieve this functionality.

5. When do you need termination on an RS-485 bus, and what are the advantages and drawbacks to the system?

In most RS-485 applications, termination resistors match the characteristic impedance of the cable to prevent signal reflections. Termination resistors should be placed at each end of an RS-485 cable. There are some short-distance-communication cases where the network will operate normally without termination resistors, but terminating the line is a best practice for all applications. The drawback of termination resistors is the DC loss generated by the resistance, which leads to increased power losses in the system. Even with this drawback, however, termination resistors are still a good idea for the majority of applications.

6. What kind of transient protection do you need with isolated RS-485 devices?

Transient protection on an isolated RS-485 device depends on the type of disturbance - such as electrostatic discharge (ESD), electrical fast transient (EFT), or surge - expected in the end system and the level of protection needed. TI’s portfolio of isolated RS-485 transceivers will have some level of internal transient protection on the integrated transceiver bus terminals with respect to floating isolated ground. In addition to this, with proper system design, you can use the isolation barrier to present high impedance to these transients. If you aren’t expecting differential transients in your system and test all transients with respect to end-equipment earth, then connecting protective earth (PE) to the logic side of isolated transceiver will create a  scenario where  all high-voltage transients occur across the isolation barrier. Connecting PE to logic side can potentially eliminate external components such as transient voltage suppression (TVS) diodes or pulse-proof resistors.

Figure 2 illustrates these techniques for enhanced transient protection using the ISO1410.

Figure 2: Half-duplex-isolated RS-485 transceiver with optional bus-protection components

7. How do you generate isolated power for an isolated RS-485 node?

There are several options to generate isolated power for an isolated RS-485 node; the best solution depends on the specific application needs.

One option is to use a transformer driver like TI’s SN6501, which operates in a push-pull configuration with a transformer and optional rectifying low-dropout regulator (LDO) on the secondary side (Figure 3). The SN6501 is capable of delivering as much as 1.5W to provide isolated power. This device has the flexibility for use in almost all applications, because the transformer and turns ratio can provide the necessary isolation rating and output voltage for the power supply. You can use the SN6505 instead of the SN6501 for as much as 5W of output power if you need isolated power for additional devices. SN6505 has extra protection features such as overload and short circuit, thermal shutdown, soft start, and slew rate control enabling a robust solution.

Figure 3: Isolated power for the ISO1410 using the SN6501

Another option for space-constrained applications is the ISOW78xx family of devices, which provides signal and power isolation in a small-outline integrated circuit (SOIC)-16 package. The ISOW7841 can be combined with a non-isolated RS-485 transceiver, as shown in the Isolated RS-485 with Integrated Signal and Power Reference Design. This combination is compact; doesn’t require a transformer; and makes certifications easy. See all documentation available in the ISOW7841 online product folder for more information.

Figure 4: Isolated RS-485 with integrated signal and power using the ISOW7841

What questions did we miss?

If you’re looking for more RS-485 information or have a question you would like to see added to this list, leave a comment below and help us keep the conversation going.

Additional resources

  • Hello,

    The PPS UHF Metal tag by OPP IOT can withstand Fuel B, mineral oil, petroleum, salt mist, vegetable oil.

    For more information, please view www.oppiot.com.

  • HI

    Very nice article . One question about the termination . You have recommended to have termination at both ends of the cable . What is we are using multi drop with multiple nodes . Should we have terminations in at  all the nodes ?

    also even when the transceivers can handle the fail-safe , is it recommended to still have external fail-safe biasing . For Fail-safe also is it recommended  on all the nodes

  • Hi Eby,

    For multi-drop network with multiple nodes (1 master node, other slave nodes), termination is needed just at the farthest receiving end- since the signal will travel in only 1 direction.

    Having external fail-safe biasing even when all the transceivers in the network have this capability built-in depends on the customer. Some customers like to have more noise margin than what the transceiver can provide and they install fail-safe biasing resistors only at one point(typically master node) in the network. Other customers  just rely on the transceiver. Please note that external fail-safe biasing affects loading on the bus and reduces the maximum number of nodes that can be connected to the network.

  • : Thanks for the reply .  so for Multiple nodes , termination is required only  for the Farthest node. Do we need to have at the master side also ?

    Also fail-safe is required only at master side only right (in case if we are using it ) . I have read some where that the fail-safe should be at two ends . Please confirm whether the termination and Fail-safe are required at two ends of the communication or just at far end for termination and at the master for fail-safe

  • also one more thing , should we have a common ground connection while connecting to two nodes .

    also this  post is related to isolated RS485, but is  the answers for my queries below is applicable for normal rs 485 transceivers also

  • Also how can we decide on the termination resistor value . Is it fixed to be 100-120 Ohms . Is there any method for calculating it

  • Hi Eby,

    Termination is needed at only one farthest receive node if network is multi-drop (one transmitter, remaining all receivers, signal only travelling from in one direction). Termination is needed at both far ends if network is multi-point (multiple transmitters, multiple receivers, so signal can go in both directions).

    Termination resistor value should be equal to characteristic impedance of cable. If you check datasheet of certain cables, you would know. As an example, you can check Belden 3105A or Belden 3106A- both have characteristic impedance of 120ohm.

    External failsafe is only needed at one master node.

    Above answers are applicable to isolated and non isolated RS485 systems.

    For non isolated systems, you should not have ground connection between nodes- to avoid ground loops.

    For Isolated RS485, since isolated side is floating, you can connect all floating GNDs by one wire. Isolation provides very high impedance and avoids any ground loops.

    For more information on this, please refer to www.ti.com/.../slla272c.pdf

  •  Thank you for the valuable information