Shop OBEX P1 Docs P2 Docs Learn Events
twisted pair vs. data cable — Parallax Forums

twisted pair vs. data cable

yarisboyyarisboy Posts: 245
edited 2009-08-11 07:33 in Propeller 1
Hi, I'm putting my Propeller and my IGBT drive board/DC buss in separate adjacent enclosures. The data wires from the Propeller to the drive box can be twisted pair as suggested by some of my chip suppliers but I also have some shielded data cable that has a static wire in it and an aluminum foil shield inside the outer teflon jacket. I've been told that you only ground one end of the static line to avoid making an AM antenna. Both my opto-isolators and my hall-effect current sensors give me isloation but I want to avoid having noise reflected back to the Propeller. Should I consider installing ferrite beads over my wires or should I just step up to coax for each line?

Comments

  • w8anw8an Posts: 176
    edited 2009-08-11 04:32
    Depends on how you're driving your data signal. Using twisted pair, you can utilize common mode rejection to eliminate line noise www.maxim-ic.com/appnotes.cfm/an_pk/2045 and not have to worry so much about different ground potentials.
  • Toby SeckshundToby Seckshund Posts: 2,027
    edited 2009-08-11 07:33
    With the shielding connected at one end it is an electrostatic shield ie it will ground only the electric field, that could induce voltage into high impedance ccts. Magnetics will pass through this non feric material and induce current into the wires. Twisting the pairs is only effective if they are driven differentialy to ballance out common mode interference such as magnetic inductions. That is the only reason CAT5/6 cables work.

    If the shield were to be grounded at both ends then there is the possibility of an "earth/ground" loop being set up. This happens because there is a difference in "earth" potentials, beleive me there can be huge differences over just a few feet. This would set up currents on the shield which could then induce ino the signal ccts. Again ballanced feed reduces this.

    As to the antenna being formed, this will very much dependant on how hostite the surrounding enviroment is. If you are installing it on the tail end of a high power transmitter then caps, coils and ferrite collers will be required. Even the if the cabe lenth equates to 1/4 wavelenth (or multiples) then it will be a fight ( We had a 100 Watt lamp connected to the outer cassings, of the two halves of a 200KHz transmitter, if it was bright then both halves were running OK).

    Keep the interconnection short and low imedance.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Style and grace : Nil point
Sign In or Register to comment.