View Single Post
Old 02-15-2015, 07:35 PM   #1701 (permalink)
e*clipse
Permanent Apprentice
 
Join Date: Jul 2010
Location: norcal oosae
Posts: 523
Thanks: 351
Thanked 314 Times in 215 Posts
Thank you, Piotrsko! I always welcome rocket-science!

In general, I am concerned about the data line recieving the noise. I figured the motor controller would be THE location of maximum noise. Those poor signal wires would be within inches of the IGBT's that are switching 650V, 150A @ 20kHz.

The differential pair wires are ( sorry - sorta OT ) on the BMS, communicating the info from the data gathering IC's to the central computer. Data is transmitted at about 1Gb/s, and unfortunately I have no control of the speed, just the amplitude.

**EDIT** this is exactly like CAN on the physical level! So, I guess it's ON topic, 'cause I would like to communicate things to/from the motor controller using CAN and Paul has already supplied CAN port hardware.
CAN bus - Wikipedia, the free encyclopedia

In my case, I'm trying to reduce the number of wires and connections. This is why I was so happy to find that intermittent twisted-pair cable. I am dealing with a system where the cable length could approach 10 feet, and the voltage between the low side and high side could approach 650V.

The reason I was confused about the value of a shield is that the design advice for a PCB indicated a ground or power plane was actually not advantageous in a stripline application ( buried signal traces in a 4 layer board ) HOWEVER - don't quote this - I may merely be confused.

I've found twisted pair/shielded wires, but often they were very expensive and I never found a good way to terminated them. (common is Ethernet)

- E*clipse

  Reply With Quote