The real core of engineering, in addition to technical expertise, is the ability to articulate the inevitable tradeoffs and pros/cons of design alternatives, and then weigh these relative to each other in the context of the application. These tradeoffs are visible and less-visible, encompassing a combination of factors such as basic performance, size, power, impact on overall design, design confidence, reliability, weight, and cost.
I came across yet another basic example of this situation when I was asked a simple technical question. A friend wanted to add an LED to a double-pole single-throw (DPST) electromechanical relay, designated as “2 Form A” in the industry-standard nomenclature (see Reference 1) to indicate that the relay coil was energized. The electrical parameters were pretty clear: the relay source was a 24 Vrms @ 1A AC supply and had a 600-Ω coil, so by Ohm’s law the relay current was 40 mArms (Figure 1).
Figure 1 The basic topology was simple enough: a 24 VAC supply, feeding a 600-Ω, DPST relay.
My first thought was to keep it simple: just take a 40-mA LED and insert it in series with the relay coil, which would function as the current-limiting resistor: one component, quick and cheap (Figure 2). But that solution didn’t sit well with me for several reasons. First, I wasn’t sure if having a diode (yes, the LED is a diode) in series with coil would affect the coil’s operation with the half-wave rectified current causing relay “chatter” at 60 Hz. Second, relays often create an inductive “kick” when they de-energize, and that would likely not be good for the LED’s lifetime; it’s no secret that electrical stresses, along with mechanical and thermal ones, are the most common causes of premature component failure.
Figure 2 The first idea, of putting the LED in series with the relay coil seemed a “no-brainer” one-component solution, at least for a few minutes.
These can be frustrating failures to figure out since the installation worked for “a while.” I could add a clamping diode across the coil, but that means adding another part. Third, I generally try to avoid physically cutting into a properly operating circuit to add something in series, as it is often mechanically difficult to do cleanly and secure properly.
So it was on to plan B: add the LED and a current-limiting resistor across the coil (Figure 3). The 1-A transformer had plenty of capacity to handle the additional 20-30 mA load, so no problem there. I also figured that there was probably no need to filter the rectifier AC output to reduce visible ripple, as the eye’s vision persistence would most likely take care of that. But there still was the issue of inductive kick.
Figure 3 An alternative was to put the LED and resistor across the coil instead of the LED in series, which would cost just one extra resistor beyond the first idea.
Further, some people are especially sensitive to 60-Hz visual ripple; if that was a problem, then I might have to add an external half-wave or full-wave rectifier to power the LED and then filter it with a small capacitor of 10 μF to 100 μF. Since this add-on version of the indicator circuit would be in parallel with the coil, physical installation would be easier (as would be taking it out, if necessary).
But stepping back, I realized there were a lot of “maybes” and “most likelys” in these approaches, which is often worrisome, would consume excess time to fine tune, and even affect long-term reliability. So I took a moment to reflect on a basic question: “What other resources do we have readily available here?” and the answer was obvious: the unused relay-pole contact pair, which is electrically independent but mechanically coupled to the active one. Note that while single-pole relays (SPST — 1 Form A) are widely available, many designers choose double-pole versions for convenience and “just-in-case” opportunities, or even go to a DPDT (2 Form C) relay and leave those extra contacts unused, as the cost difference is often outweighed by the extra future flexibility they offer.
So I suggested he wire the LED and its resistor from the 24 VAC source through that unused relay pole (Figure 4), giving him an indicator circuit that was totally independent of the relay-coil circuit itself, which is good for minimizing both electrical impact and actual installation difficulties. Also, the independent circuit also indicates that the relay armature has actually pulled in, in addition to the coil being energized, so it provides a higher level of confidence in the relay’s operation. If driving the LED from the AC line results unacceptable visual flicker, well, there are easy ways to deal with that and there’s no need to worry about affecting the coil-drive loop.
Figure 4 Making use of the unused relay pole pair was a better, cleaner, flexible, and risk-free solution that also verified not just that the coil was energized but that that the relay armature had actually pulled in.
Have you ever looked at a simple design question and found that despite the KISS (keep it simple, stupid) principle, the obvious solution really wasn’t a smart move after additional thought? Even better, have you ever found extra, unused relay contacts that you could leverage, or upgraded your BOM from an SPST relay to an SPDT or even DPDT relay to solve your problem?
- “Relay Schematics & Forms,” Sensata Technologies.
- HVAC Upgrade Shows Interconnect Complexity of These “Simple” Systems
- Relays Are Great, But There’s No Need to Let Them Waste Power
- Reed relays switch to new designs, applications