# Is it OK to use RS485 with separate wires for each device?

I know common use of RS485 is top wiring diagram. But in my project we need to use RS485 for extend range and reliability connection but connection will be bottom wiring diagram.

Can I use a single RS485 transceiver IC? Are there any problems with this approach? I'm not sure.

Assuming every device has the same baudrate.

• Each end of the RS485 cable needs to be properly terminated. In your second diagram, that means Device 1 and Device 2. Masters typically are located at the end of a cable, and their line drivers typically terminate a line by providing a fixed impedance path to ground. This may be problematic, but without knowing the specifics of the master, particularly its line driver circuitry, one cannot tell. Do you have a schematic for the line driver section of the master? – Math Keeps Me Busy Feb 28 at 14:28
• There are RS485 multiplexers (hubs) that allow one to connect multiple slaves (through the multiplexer) to a single master in a star topology, but they tend to be rather pricey. They may be something to fall back upon if other techniques don't work. – Math Keeps Me Busy Feb 28 at 14:47
• It would be helpful to have more information. What length and what type of cable do you plan to use? How many devices do you plan to connect? – 10ppb Feb 28 at 16:18
• @MathKeepsMeBusy Master is common RS485 transcevier IC (Non-isolated) eg. SN75176 – user282341 Mar 1 at 1:28
• Length from master to device can vary from 10 to 50 meters and each device maybe difference lenght. Max of device is 4. – user282341 Mar 1 at 1:30

Both of these will work providing you apply the modifications in red: -

Both distant ends of the cable need a suitable termination resistor to protect data reflections causing corruption of the data messages.

Can I use single RS485 transceiver IC ?

The same RS-485 interface IC can be used for master and slaves alike. You can even take the cable interface onto your PCB like this: -

It makes no difference where the master is (middle or end) providing it's using an RS-485 line interface chip. The same interface chip can be used for interfacing the master or the slaves. It makes no difference because RS-485 outputs a current drive signal to the line.

• OK. I will add PCB jumper for terminated resistor. But I 'm still worry if lenght for each device very difference eg. Device 1 is 10 meter, device 2 is 100 meter. – user282341 Mar 1 at 1:35
• @user282341 In most systems I've used the terminator is simply a resistor inside a plastic box (kind of like a dongle but can be DIY) that we add to the end of the network. It does not need to be part of any device on the network – slebetman Mar 1 at 3:26
• @user282341 it should not be a problem with the different distances providing the cable is chosen to have the correct characteristic impedance and both far ends of the cable are terminated correctly in a matching resistor. – Andy aka Mar 1 at 9:03
• @user282341 instead of a jumper add a dip switch, that way you can change it during installation if your device is not at the end of the line. – Jan Dorniak Mar 1 at 13:34
• Thank you for recommended – user282341 Mar 1 at 14:35

What you propose is the simplest star topology in existence.

RS-485 specifically advises against going for a star topology (mostly due to basically making it ambiguous where termination should be, and there's no "good" answer to that question); but at three full devices on the bus, things might be OK. What you need to realize is that there's 100m of cable between Device 1 and Device 2.

Your devices will have to be able to drive 100m of cable, and you'll need to find the sweet spot between having too little load at each node, and too much load overall in termination to avoid problematic echoes. You're building a frequency-selective channel here. Assuming the E-field change travels at $$\\frac23 c_0=2\cdot10^8\frac{\text{m}}{\text{s}}\$$, it needs 0.5 µs to travel the from Device 1 to Device 2 (100m), and 0.75 µs to travel back to the Master in the middle if it's reflected at the end. Now, if these 0.75 µs happen to be an odd multiple of half a signal period, you'll have destructive interference there. Uh oh. That means you'll lose signal integrity for some signalling frequencies (including subharmonics introduced through repeating bit patterns), and get amplifications for others. That's why termination is important!

Also: If you have some "collision avoidance" scheme where access to the shared bus can only be asserted when there's been silence for a specific amount of time, then you'll need to account for the travel time when calculating the minimum amount of wait-before-send.

• But that is a linear bus; master in the center and two slaves at ends. If the master has means to turn termination off, and terminations are added to ends at both slaves, it should be okay. Unless there is something we don't know. – Justme Feb 28 at 18:54
• Exactly! With terminations, we're practically in clean, as that defines a "linear" structure. – Marcus Müller Feb 28 at 19:15

If we are talking about just 2 devices it shouldn't be a problem if you terminate both ends, however, if you use a star architecture (which is essentially what you are doing although with just 2 slaves) with more devices, you will need more terminations, and that will excessively load the driver, the other option is not to put terminations which can screw up the signal due to reflections.

As Marcus says, star architectures in rs485 are specifically advised against, yet in this particular case it should work, whenever you need multiple terminations be sure to calculate the load to not exceed 32 ul, which is the load any rs485 driver should be able to handle

• It is not a star if there are only three devices. Each device must also be able to reply back so it makes no differenence where on the bus each device is; however the terminations must be at the ends of the buses so master at the center must be unterminated. – Justme Feb 28 at 18:58