This one blindsided me today when my boss called me in for consult...

A customer wants a bid for installing an "RS-232 signal line" - single pair shielded twisted pair, between two panels. (I had to look three of four times at the specs, it does indeed say "RS-232." Go figure.)

Here's a reproduction of the drawing we were given:


My questions are:

1) Would the connection be some pre-fabricated D-shell like the venerable old RS-232 I once knew and hated? Or would it be PLC-esque screw terminals? Or something else?

2) Would the caps and resistors be some commercially available assembly, or would we need to fab something up?

3) What is the purpose of the caps and resistors? If it were just caps, I'd say it was for de-coupling noise to ground. But resistors? Is this for impedance matching or something?

If anyone has experience withthis kind of termination, I look forward to your reply!

 

I wonder if maybe the labeling of the cable is just a reference to the type of cable that is recommended to be used. Some cable manufacturers use labels like RS-232 cable as a category of cable. The fact that the two ends are labeled with a "+" and a "-" tends to make me think that it is a RS-485 serial link. Most likely a half-duplex communication link.

Are the compressor and starter panels located in a noisy environment and possibly a significant distance apart? That would explain the resort to differential communications link.

hgmjr

 

If it's low-impedance differential, terminal strips and spade lugs will do quite well. Or some nice ugly old Molex connector. You just can't do 232 on one pair of wires. When GM used 6800's in their robots, all the signal inputs came in through 1000 ohms with a .1 uF cap to ground.

 

It is indeed a noisy environment. Lots of very big motors turning on and off.

Based on your RS-465 serial link idea, I've been doing a little Googling... looks like we might be dealing with either a "Modbus" or "Mosaic UART Wildcard" using "AC termination configuration."

http://www.national.com/an/AN/AN-903.pdf

The resistor matches impedance, and the cap eliminates DC loading. I've come across other websites giving similar comment.

I just hope I'm not barking up the wrong tree.