Electromechanical relays: an old-fashioned component solves modern problems

One of the frustrating challenges in integrating systems is interfacing an existing subsystem with a new or different subsystem. In cases where the input/output characteristics of both sides are standard and known, a level-shifter/translator IC easily solves the problem (Figure 1).

Figure 1 When the two interfaces are standard and well-defined, an off-the-shelf level-shifting IC may quickly and easily solve the problem, as it can accommodate different digital-signal levels on its inputs and outputs. Source: Texas Instruments

But there are many scenarios where the interconnection between the older and newer systems—often an older one and a newer one—is not easily resolved. Worse, there are also many situations where the characteristics of one or both sides of the interface are not known or are poorly defined, so the solution is more challenging and has risks.

This was the problem I faced in two separate situations. In each case I was able to easily and neatly solve the problem using a modern version of the nearly 200-year-old electromechanical relay (EMR). The invention, circa 1835, is usually credited to American scientist Joseph Henry.

While relays may seem like holdovers from the days of dinosaurs versus using an optocoupler or solid-state relay (SSR), there are many cases when their “old time” virtues are very beneficial. Hundreds of millions of such relays are sold every year and, while many are for replacement use, a substantial fraction go into new designs.

Relay types fall into three general groups: small-signal devices for contact currents under 2 amps; power relays for much-higher currents; and RF relays which usually handle small signals into the megahertz and gigahertz range. In addition, there are other family siblings such as audio relays and reed relays for automatic test equipment (ATE), to cite just two of these. Further, relays are available with many contact configurations starting with the basic single-pole/single-throw (SPST) version and going up to units with multiple poles and contacts.

I’ll be honest: I really like relays. Among their many virtues, they:

are rugged and reliable;
provide near-perfect galvanic (ohmic) isolation;
have “input” and “output” current/voltage ratings which are largely independent;
have the contacts which are largely agnostic with respect to carrying AC or DC, or signal or power—as long as it is within their ratings;
come in a wide range of ratings for the coil and contacts;
usually have a compatible socket which simplifies wiring and debug;
are offered in many versions with multiple normally open/normally closed (NO/NC) contacts which prove very handy;
provide a visual indication via armature position showing if they are energized;
and, finally, also have a very satisfying, audible “click” when they open or close;
are reliable, so long as a quality relay is used (that is within its specifications), will last for millions of cycles and tens of years.

In short, they are no-hassle, flexible problem solvers, and can be used for two unrelated scenarios where a small, basic EMR provides crisp, clean solutions.

Challenge #1

I “volunteered” to help a friend upgrade a heating-only system control from a basic but still-working old-fashioned thermostat with just a simple temperature-driven on/off function (similar to the one in Figure 2) to a more sophisticated, smart, Wi-Fi enabled unit. I took a quick look at the existing thermostat and saw it had just two wires, so I figured “how hard could this be?” especially as its output action was a simple contact closure.

Figure 2 The classic T87F Honeywell thermostat was introduced in the early 1960s and, though it is no longer offered for sale, million were installed and many are still in use due to an extremely reliable bimetallic thermal strip and a hermetically sealed mercury switch (no longer allowed, of course). Source: New York Historical Society

But that was not the situation at all. The smart-thermostat documentation showed that there were many possible configurations and options for wiring when going from the two-wire unit to a new, much-smarter, three-wire one.

First, the new thermostat needed to be powered via a 24-VAC transformer, so an extra lead was needed between the HVAC controller and the thermostat. Fortunately, there already was a third unused lead in place (thank you, whoever did that!), so that part of the problem was a non-issue.

There were many ways shown to connect the new three-wire system to the existing two-wire contact closure, depending on the vendor of the system controller (which in this case was fairly new), whether it was heat-only versus heat plus AC (heat-only here), and how you wanted to deal with galvanically isolating the new thermostat from the existing controller which needed to see a basic dry-contact switch closure to turn the heat on. Although this is a 24-V signal indication, it’s also at modest-power levels.

One option was an optoisolator or SSR, but I wasn’t sure about voltage and current levels, or if an SSR output that is “on” looks enough like a true contact closure, as there is a small voltage drop at the SSR output. The easier and less worrisome option was to use a relay for isolation a general-purpose DPDT relay, with a 24- VAC coil and 2-A contacts in a basic translation/isolation circuit (Figure 3). Obviously, its switching speed of tens of milliseconds is not a concern in this situation.

Figure 3 The problem of interfacing a modern smart thermostat to a new, simple two-wire heating-system controller was solved with a relay having a 24 VAC coil. Source: Bill Schweber

And the good news is that it all worked fine the first time! You can’t beat that.

Challenge #2

I was asked by another friend to help replace the now-obsolete landline telephone dialer on a home-security system with a modern wireless-cellular unit. As both the old and new units were triggered by a simple two-wire closure, I hoped this would be a drop–in replacement—but it was not.

The previous dialer required a transition from an “open” to “closed” contact-pair to trigger it, while the manual of the new one said it needed a transition from ground to “open circuit” as its trigger. Unfortunately, “open circuit” is one of those ambiguous terms: does that mean “floating” (truly open) or will disconnecting the line but doing so via an open-collector output be sufficient?

Adding to the challenge, the documentation about the output pin of alarm-control unit driving the dialer was unclear as to its electrical nature: it might be an open-collector structure, or maybe not. Thus, I could not even say for sure if the control unit’s output was at least potentially electrically compatible with the dialer’s input needs.

I thought about it for a while and distilled the issue to this summary: what I had was an output that went from high-to-low with unclear structure, and what I wanted instead was for it to look like a DC-signal going from ground to open.

Again, a small general-purpose DPDT relay with 5-VDC coil and 1-A contacts seemed like a flexible, risk-free way to address these issues. I connected the coil between the supply rail and active output of the control unit, then used the unenergized relay contacts in NC mode to connect the input of the dialer to true ground (Figure 4). When the control unit’s output went low (pulled down) it energized the coil, which opened the NC relay contacts, and thus provided a true open circuit to the dialer input.

Figure 4 The translation and signal inversion between the poorly defined output of the alarm-system trigger and the new wireless dialer was also accomplished with a low-voltage DC relay. Source: Bill Schweber

Again, problem solved…the relay functioned as both a level shifter and signal inverter and gave me absolute electrical (galvanic) isolation as well.

Old-technology “revenge”

There’s some irony (or a form of old-technology “revenge”) in using the classic electromechanical relay to solve modern problems such making a Wi-Fi-enabled, firmware-based device compatible with the older interfaces, or addressing inversion and isolation for interfaces with unclear characteristics.

Although the schematic diagrams of these two solutions are simple, they elegantly and effectively solve problems, so what’s not to like? The lesson is not to hesitate to consider old-fashioned components, as many are still very viable and can offer needed functionally and flexibility.

As for me, I think I need to stop volunteering to help friends with their interfacing problems.

Related Content

Reed relays switch to new designs, applications
Adding an LED to a relay illustrates circuit tradeoffs
Relays Are Great, But There’s No Need to Let Them Waste Power
Relay hysteresis provides toggle function
<!–

–><!–

div-gpt-ad-inread

–>

The post Electromechanical relays: an old-fashioned component solves modern problems appeared first on EDN.