ISO1450: iso1450 cable length

Sammi,

The limitation has more to do with the bus capacitance over a long distances rather than power supply. We have a small article that explains this relationship here https://e2e.ti.com/blogs_/b/analogwire/archive/2018/07/26/top-7-design-questions-about-isolated-rs-485-transceivers

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

You will have to determine the characteristics of 35m of the the bus cable for this application in order to determine the max data rate. Most likely 10Mbps data rate will not be possible with 35m bus cable.

Respectfully,

Lucas

Sammi,

In addition to what Lucas quoted:

1. If the customer's application does not expect data rate to go beyond 10Mbps, we will recommend ISO1430 (12Mbps version of same ISO14xx family). Rise/fall times are slightly slower for 12Mbps variant, so signal reflection (and corresponding signal integrity) would be less of an issue.

2. Please see below link of another application note which in Section 9, Page 5 states a conservative rule of thumb: 

Product of line length (in meters) x data rate (bits per sec) < 10^7.

This is just a conservative guidance, but with good quality cable it is possible to exceed this. Ultimately, we will recommend customer to do enough testing on actual setup and check out jitter and signal eye diagram on receive end. Acceptable jitter depends on application and can range anywhere from 5-15%.

https://www.ti.com/lit/an/slla272c/slla272c.pdf

Hi Vikas,

Thanks for your response.

If my customer use 1Mbps data rate, the length could only support 10m refer to the formula?

Is the line in the figure zone 2 is a strictly math relationship?

For 1M data rate, could support 35m?

Thanks.

Best regards,

Sammi

Hi Lucas,

Thanks for your response.

If customer use 1Mbps， could it support 35m?

Thanks.

Best regards,

Sammi

Hi Sammy,

Zone 2 in the figure and rule of thumb equation are just guidance in RS485 standard application guidelines (TIA/EIA- 485). In most cases, they are very conservative and you should be able to exceed that.

My understanding is 1Mbps could be supported to 35m easily. Ensure usage of twisted pair, shielded cable (one such example is Belden 3106A cable).

I will reiterate, please ensure sufficient system testing is done and signal integrity is verified.

Hi Sammi,

An additional application note that may be helpful to your question:

https://www.ti.com/lit/an/slla431/slla431.pdf

This is for non-isolated transceiver THVD1450, and shows practical lab results on cable length vs data rate with jitter performance.