SN65HVD1474: vs. SN65HVD1477 for 20 Mbps, termination for longest cable reach

Hi Ueli,

Apparently 1477 (50Mbps) chip works worse than 1474 (20Mbps) for the same data rate of 20Mbps.

In what way is the customer seeing worse performance with SN65HVD1477 compared to SN65HVD1474? These devices have different slew rates to accommodate different data rates. A higher slew rate (faster edge) is more likely to cause ringing in a system due to its high frequency contents. Could this be the cause of worse performance for this device?

Would this two chips require different termination because difference in differential driver output?

No. The termination scheme can be the same. This is more dependent on the cable and node layout than the driver characteristics. 

What are the best termination methods for transmitter and receiver to active longest cable reach.

Single and Bi-directional parallel termination are both commonly used for single-ended RS-485 systems. Termination value should be selected to match cable impedance. See this application report on different termination methods: 

AN-903 A Comparison of Differential Termination Techniques

Is there any difference in input circuit between 1474 and 1477?

The equivalent circuit the receivers in the SN65HVD147x family are all similar (no added noise filtering, etc. depending on model). Timing differences such as propagation delay, output slew, and enable times exist due to different data-rate receiving capabilities. 

Is there better driver for 20 Mbps application?

Our featured devices currently include THVD2450, a 50Mbps full-duplex transceiver with +-70V fault protection. If you system appears to be sensitive to fast edge rates, I would also like to mention THVD1429 which is designed for 20Mbps. 

Let me know if you have any more questions.

Regards,

Eric