PT100 in 4-wire configuration
-
I'm having a strange problem switching my PT100 from a 2-wire to 4-wire configuration. In 4-wire mode I am getting a 2000c reading and "sensor hardware error", but ONLY when either the hotend fan or part cooling fan is running. If the fans aren't running, temperatures are reading perfectly and everything looks great.
All I did was run two more wires to my sensor, connect them to the daughterboard, and remove the solder bridges from the pads near the connector. I have checked and rechecked my wiring, and measured voltages to confirm that everything is connected properly.
If I simply add the solder bridges back to the pads by the connector, it goes back to working right 100% of the time whether or not the fans are running. I'm at a little bit of a loss here given that it reads properly in 4-wire mode if the fans are off, and works fine in 2-wire mode regardless of fans (without changing any of the wiring). Has anyone else successfully used a 4-wire PT100? What am I missing?
-
That is very odd. Obviously looks like an interference issue but 4 wires should reduce not exacerbate the interference.
I will run a test later today to see in I can replicate it.
-
It does sound like an interference issue. I use a 4-wire connection, see my blog for details. I have a separate unshielded 4-core cable for the PT100. These cables have a gentle twist along their length, which helps avoid inductive interference pickup, although not as well as twisted pair does.
-
DC42, yep twists are the way to go.
I'm running PT100's with twisted pair 2 wire and have the wires running next to and/or very near the stepper motor wires for the extruders. No interference at all. Twisted pairs work really well for this sort of thing.
Jeff
-
Thanks all, this makes perfect sense. Thinking back, I believe I probably did run the two additional (untwisted/unshielded) wires in the same channel as my fan wiring. I'll re-run it with twisted pair cable and see what difference that makes.
-
ahh that may well bit. For clarification the PT100 chip cancels out the signals it sees on both wires on each side, if they are following different paths they are going to have different noise on them which will mess with the results.
