RS485 Termination Resistor

rs485 termination

Hello,

I have a half duplex RS485 communication between 1 slave and master but was wondering what is the effect of the termination resistor? My length of CAT5 cable is 5 m and the baud rate is 19200. When I add a termination resistor based on the RS485 IC datasheet, the communication is very unstable but when removed it works fine. Can anyone explains it? I have calculated the propogation time and it is faster than the data bit rates so I shouldn't need it.

Cheers
P

rs485 termination resistor

Termination not only changes biasing requirements but also adds heavy DC loading to a system.
If you want to keep it, you may consider adding bias resistors (2 x 1.0k - 1.5k), one between A(+) and GND and one between B(-) and +5V.

However, I wouldn’t worry about termination on short transmission lines such as yours.

Rgds,
IanP

termination resistor

If my data rate speed increase to the MHz region which surpass the propagation delay of 0.6 ns for a 5 meter cable, I guess I would need a termination resistor.

rs-485 termination

the propagation delay of 0.6 ns for a 5 meter cable

Click to expand...

Faster than light! Where did you get the cable?

rs485 terminal resistor

I have frequently used multi drop systems with up to 100 receivers and one transmitter. Termination resistors would put too much total load on. Sometimes use end termination for single runs or star wiring. Either way never had problems with several hundred meter runs on standard telephone type cable, should be better with cat5. Usually lower rates than yours though 300-9600B. Hope this helps.

rs 485 termination

FvM said:

the propagation delay of 0.6 ns for a 5 meter cable

Click to expand...

Faster than light! Where did you get the cable?

Click to expand...

Sorry miscalculated the value It should be propagation velocity = 0.66 x c = 197863022.28. For a 5m cable the propagation delay should be 76 ns assuming it damps out in 3 round trip.

rs485 resistor

There's nothing against using termination resistors also in low speed, short range RS485 connections, if you don't care about the involved waste of energy. It's generally done in industrial applications. Standards as PROFIBUS don't have an option to omit the termination below a certain baudrate * distance product.

Clearly, the termination has to be applied at the end points of a multi drop party line only and the said bias resistors are needed for correct operation.

rs485 terminating resistor

That is what our hardware engineer said that the termination resistor should not affect the communication when running at low speeds but it seems to cause 20% of our units to fail. By removing the termination resistor, it works fine with no problem on the failed units.

Could it be the low voltage RS3483 we are using and the termination resistor is causing DC loading which affect our communication or do we need proper biasing resistors? The termination resistor works fine in our other products but the other products uses the 5V RS485 device instead of 3.3V device.

When units failed, the communication at the 485 output is an active low line compare to the units that pass which is an active high line. When active low, it seems to affect the start bit and possibly the stop bit.

Any ideas?

rs485 resistors

Does new fail-safe RS485 transceiver coming really work without bias resistance ?? does any body has any experience with that

Just want to close this topic off. The solution was I remove the termination resistor as it was not needed and has work perfectly without any problems.

Hi... Can you please explain the calculation for getting 76ns...
as i know the speed of light = 299 792 458 m / s...

agg_mayur said:

Hi... Can you please explain the calculation for getting 76ns...
as i know the speed of light = 299 792 458 m / s...

Click to expand...

Propagation velocity = 0.66 x speed of light = 197863022.28 m/s.
Propagation delay = (1/Propagation velocity) * cable Length
Propagation delay = (1/197863022.28)* 5 = 25.33 ns

Assuming it damps out in 3 round trip

Propagation delay = 25.33 *3 = 76 ns

Hope It will help