Advertisement

Analog Angle Blog

HVAC Upgrade Shows Interconnect Complexity of These “Simple” Systems

I was recently asked to help a friend upgrade a heating-only system from a basic thermostat with just a simple, self-contained time-of-day programmability a more sophisticated, Wi-Fi enabled unit. These are sometimes referred to as “Nest-type” thermostats after the first mass-market device of its type, although the one we were installing was not a Nest-brand 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. This type of naïve, simplistic thinking comes when you assume that because you know one thing –basic electrical-wiring principles – you’ll have little trouble with the slightly similar discipline, in this case the wiring of HVAC (heating, ventilation, air conditioning) systems.

We soon found out that was not the situation at all. In going through the documentation, there were many possible configuration and options for wiring when going from the two-wire unit to the new, much-smarter one. First, the new unit needed power via a 24-VAC transformer, so an extra lead was needed between the HVAC controller and the thermostat. While that is no big deal (it’s just safe 24 VAC), it can be a challenge in many houses. (And if you are running one extra wire, why not run another with it, as there was also a four-wire interconnect option).

The real issues began with evaluating the “best” interconnect choice via the three wires (two existing ones plus a new one) between the HVAC controller and the new thermostat. There were many ways shown to connect it, depending on the vendor of the system controller, whether it was heat-only versus heat plus AC (not an issue here), and how you wanted to deal with galvanically isolating the new thermostat from the existing controller.

Those old-fashioned, non-programmable controllers are not as “simple” as you may think, as they have various timing functions, lock-outs, alarms, and thresholds (“if water level is too low, then don’t start boiler even if the thermostat calls for heat”); in some ways, they are really sophisticated programmable logic controllers (PLCs) in an application-specific package (see Reference ).

Not surprisingly, there are different vendor configurations and standards for wiring of these controllers, as well as different alphanumeric designations for their terminals, such as R, RC, RH, C, W, W1, W2, Y, Y2, G, and O/B, among others. This is most likely a result of historical reasons, as different controllers were developed by different vendors with differing perspectives.

The result is that the vendors of the new, much smarter thermostats must explain to users how to identify and then connect these advanced units to the differing controllers in use. Since the person doing the installation – an average homeowner – is not an HVAC-controller expert, nor has electrical expertise, it becomes a difficult documentation assignment for the thermostat vendor which just adding more internal software can’t solve.

There are also decisions which the end-users must make, such as whether to go with the three-wire configuration or use four wires. The former requires a relay to isolate the new thermostat’s output from the existing controller’s “call for heat” input; the latter requires an add-on called a “common maker” which provides isolation and power routing, Figure 1 , (although I am not clear as to exactly what it does and how it does it); see the Nordic Technology Fast-Stat site for some of the possible wiring options using the common maker.

Figure 1

Devices such as this Fast-Stat from Nordic Technology are designed to ease upgrading a two-wire, unpowered thermostat interconnect to a more complicated, loop-powered, four-wire connection but without physically adding wires.  (Image source: Nordic Technology)

Devices such as this Fast-Stat from Nordic Technology are designed to ease upgrading a two-wire, unpowered thermostat interconnect to a more complicated, loop-powered, four-wire connection but without physically adding wires. (Image source: Nordic Technology)

Long story short: we already had an available unused third wire in place, so it made sense to try that option first, as there would be no need to run a new wire. That allowed us to use the venerable electrotechnical relay for isolation, Figure 2 , which is ironic (or a form of old-technology “revenge”!) as we were making a Wi-Fi-enabled, firmware-based device compatible with the older world by using this most-ancient device.

Figure 2

In the end, an old-fashioned electromechanical relay was the perfect interface, offering isolation, reliability, simplicity, and clarity of functions and features. (Image source: Ecobee)

In the end, an old-fashioned electromechanical relay was the perfect interface, offering isolation, reliability, simplicity, and clarity of functions and features. (Image source: Ecobee)

I was very comfortable with that approach, as I really like relays. They are rugged and reliable; provide total isolation; the “input” and “output” current/voltage ratings are largely independent; they come in a wide range of ratings for the coil and contacts; usually have a compatible socket available which simplifies wiring and debug; often come with multiple NO/NC contacts which prove very handy; provide a visual indication via armature position showing if they are energized; and, finally, they also have a very satisfying, audible “click” when they open or close.

It’s easy for “electronics” people to criticize the HVAC market for not having made things simpler and more standardized, but doing so would be the result of ignorance and a misplaced sense of “superiority.” Let’s be honest: the HVAC industry has developed some very good, long-lasting products, and the electronics industry in is no position to complain about having so many overlapping or confusing standards.

After all, it’s easy to assume you know enough to tackle a project, when instead knowing a little can lead to somewhat unrealistic overconfidence. Still, as long as you are willing to think differently, you can learn enough to make things work.

Have you ever tackled projects where you had just enough understanding of the basics to make you assume it will be “no problem,” or where you criticized how “they” did what they did, yet it was really you venting frustration due to your own lack of expertise?

Related Content

How an electromechanical relay literally wrestled my software to the ground

Signal Isolation Reveals True Analog Competition

Keeping it simple still a good idea, for both problem and solution

Long-living products defy conventional wisdom

When software controls your cooling tactics

Reference

How the PLC revolutionized industrial process control

1 comment on “HVAC Upgrade Shows Interconnect Complexity of These “Simple” Systems

  1. eteonline
    December 27, 2017

    I am all with you on relays. If their duty cycle is anything reasonable as it is in HVAC, they last a long time and the isolation is complete.

    I service industrial equipment and much of it is old, It is old because it does the job and the cost is already fully paid.  The cost of replacement is often extreme, in hundreds of thousands of dollars, and there are training, interface and operational differences to be overcome.

    Out in my world and that of my customers, the latest is usually overpriced, overfeatured and different enouigh to make things difficult for my customers.  Service for these brand new darlings is a several hundred dollar service call to say, “It's broke, buy an entire new unit as they are not repairable, $5000 please.”

    In terms of assumptions, a line of CNC machines made in Italy uses green which means frame ground in the rest of the world for +24 volts. And all the services docs are in the German (which I studied 40 years ago) and Italian (which I never did study.)

    In another product line, remote controls, the mounting frame is ground, sort of, so long as the mounting posts that hold the circuit board up in the case are tight, otherwise it is not. Thke unit is dingy and borderline to begin with, so it doesn't need any hellp from intermittent operation.  This particular company started in Sothern California, was bouight out by another company in Pennsylvania that tried to kill the line and replace it with their own shoddy line. It is turn was bought out by an English comgomerate. The docs on the old unit and all the parts went in the dumpster when they moved out of California.

    Remember young engineer that your product will go out in the cold cruel world that will appreciate not your cleverness or saving 2 cents on that borderline resistor, but that their assembly line doesn't shut down costing them $1000/hour so you could save the 2 cents per unit on your $50,000 product.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.