The other day a couple of us were sitting around contemplating picking up the suction and discharge pressure signals from the 4-20 milliamp transducers serving a controller in a package chiller so we could feed the information into a DDC control system with out having to purchase and install pressure transmitters on the refrigeration circuits. The idea was to add a scaling resistor to the current loop (assuming E=IR verified that we had sufficient power to drive through it in combination with what was already there) and feed the voltage generated into our new control system. The figure below illustrates the concept.
In the course of the conversation, I recalled a problem I had encountered while doing this years ago and also the “trick” I discovered that allowed me to solve the problem. So I thought I would do a post on the subject to supplement the 4-20 ma current loop article in the October 2009 issue of CSE and my previous posts on the topic.
To understand the potential problem, you first have to understand one of the more common terminal clamping arrangements called a tubular screw clamp. The figure below illustrates a typical tubular screw terminal and its basic components.
As you can see from the photo, the contact tube is a more complex shape than the simple hollow square I use in the graphic, and the screw and clamping plate and screw have some hardware associated with holding the plate “captive” to the screw so it doesn’t fall out when you loosen things up to insert a wire. And, the details of how the components are assembled and work can vary a bit from manufacturer to manufacturer.
For instance, when you insert the terminal hardware in the photo into the plastic housing that supports it and turn the screw, the tube actually moves relative housing rather than the plate moving while the contact tube remains stationary. But, the simplifications I used made the graphic easier to draw (remember, I’m an engineer, not an artist) and also will allow me to focus on the fundamentals behind the problem and its solution.
This next figure illustrates what generally happens if you insert a solid conductor into the terminal and tighten it down.
The plate quickly captures the solid, cylindrical shaped wire and the screw stops turning since the solid wire is incompressible. Depending on the details of the terminal construction and where the wire was inserted, the plate may cock a bit and force the wire into a corner. But the shape of the conductor is basically unchanged by the process.
In contrast, a stranded conductor is significantly deformed when it is inserted into the terminal and the screw is tightened as illustrated below.
After initially contacting the outside diameter of the strands making up the conductor, tightening the screw tends to flatten out the strands into more of a rectangular shape until you reach the point where you are trying to compress the solid individual strands against each other, the tube, and the plate, at which time the screw stops turning.
Assuming you fully tighten the screw and the terminal is rated for use with either stranded or solid wire (I think I remember running into one that was stranded only), either situation illustrated above will provide satisfactory service. But, if you insert a solid and a stranded conductor into the same terminal, a problem can arise as illustrated below.
If the solid conductor is the same gauge or a larger gauge than the stranded conductor, there will be a tendency for the clamping mechanism to tighten up on the solid conductor with out really capturing and clamping all of the individual fine wires that make up the stranded conductor. Depending on the specifics, this may not be as obvious as it seems; enough pressure can be generated to make it seem like the wire has been captured if you give it a little tug. But if the connection is not really tight, it will degrade over time, just like any other loose connection.
So, you may be wondering what this has to do with scaling resistors in current loops. The photo and schematic below of a scaling resistor applied in the field illustrates the connection.
Actually it illustrates the potential lack of connection because control system field wire is typically stranded while the leads on resistors are typically solid. As a result, inserting a scaling resistor and a small diameter stranded conductor into the same terminal can create a problem. I stumbled into it in the mid 1980’s when a system I had designed and worked with frequently started having erratic data after performing well for months if not years. The problem was traced to poor connections at the terminals where scaling resistors were inserted with stranded conductors.
You could temporarily (as in for a couple of months or years) solve the problem by loosening and re-tightening the terminal screw, which I hypothesized at the time cleaned things up enough to establish good contact again. But since the connection to the stranded conductor was relatively loose (if you tugged hard, you could pull it out, especially if none of the strands had been trapped under the solid scaling resistor lead) it would again degrade over time.
My first attempt at a solution was to buy a double deck terminal block (an example from Entrelec is illustrated below) and install the terminal between decks on one side and terminate the wires appropriately on the other side using stranded wiring for all conductors.
But, in discussing my application with the sales engineer when I called for pricing, I soon learned that someone had already solved my problem for me. Entrelec along with several other manufacturers like Phoenix Contact, Allen Bradley, and Weidmuller offered a double-deck terminal block with a precision resistor already brazed in place between the decks, as illustrated below.
This meant that you have one terminal for each wire and that the terminal would also act as a sort of boundary between the field and the controller in that the wiring on one side could be the wiring carrying the current loop while the wiring on the other side could be the wiring carrying the signal to the controller as illustrated below.
So there you have it, the relationship between tubular screw type terminal blocks, scaling resistors, and current loops. Discovering the specialized terminal kind of gave me mixed feelings; on the one hand, I felt a little dumb for not having realized there could be a problem in the first place and thus, setting myself up to learn about it the hard way. But on the other hand, I felt a little less alone; enough engineers had encountered the problem and thus needed a solution to generate a market for a product.
So hopefully, the information I’ve shared lets you skip my first step and enjoy a feeling of nerdy comradery; after all, we’re in the engineering thing together and the more we share what we know, the better off we all will be.
Senior Engineer – Facility Dynamics Engineering