SN65HVD3082E: SN65HVD3082E_EMI_issue

Hello,

Can you please tell us a little more about your set-up? There are several things that would be useful to know:

- How is EMI being measured? Is it a conducted emissions measurement or radiated emissions? If conducted, is it measured on the RS-485 lines or power lines and what sort of coupling network is used from the DUT to the RF receiver?

- What components are placed on the RS-485 bus, and how many nodes are populated? If we could see a schematic of the RS-485 implementation it would be very helpful.

- What kind of cable is used? (Is it shielded/unshielded, twisted/untwisted, etc.?)

- What is the data rate on the RS-485 bus?

Once we have a better understanding of the measurement set-up we can help recommend different techniques for improving the performance. In general, though, common techniques for improving EMI on signal lines are to use "split" termination (see this blog for an example: e2e.ti.com/.../the-importance-of-termination-networks-in-can-transceivers ), use filtering capacitors on the signal lines (assuming the EMI problems are in frequencies much higher than the RS-485 signals), or using common-mode chokes on the signal lines. Improving EMI on power lines can often be accomplished via choice of decoupling capacitances and other filter devices like ferrite beads.

Note that PCB layout can also have an impact on EMI performance, especially when higher-speed signaling is used. You can refer to this application note for further details: www.ti.com/.../szza009.pdf. If you would like, we could review the layout in this case and see if we can recommend any improvements.

Note also that TI does have newer RS-485 transceivers that are p2p with SN65HVD3082E but have been designed with improved EMI perfomance in mind. THVD1550, THVD1500, and THVD1450 are examples of these. They may be worth evaluating in this case, especially if EMI cannot be improved through simple schematic modifications.

Regards,
Max