A somewhat justified concern about damaging a complex system can also intimidate you from a basic investigation into a system problem.
OK, I admit it: I avoided checking for an obvious problem due to my fear of making things worse in a high-risk situation. Fortunately, the only cost, in this case, was to my pride.
Here’s the backstory: about a year ago, we replaced our boiler-based heating system (Figure 1). It was at least 60 years old, still worked well and, surprisingly, efficiency was pretty decent if not great.
Even though the return on investment (ROI) for a new energy-saving unit would be ten or fifteen years, we decided that doing a scheduled, planned replacement was the smart thing to do rather than deal with a winter breakdown and crisis-driven system replacement, as that existing one was unrepairable due to lack of spare parts and needed expertise.
The control for the old system was a textbook example of simplicity: the dumb but reliable thermostat energized a relay, and the closed relay contact turned on the AC power for the motor and fan. The fan mixed gas and air and blew the combination into a combustor which heated water. The only additional control was a safety interlock which prevents the gas-flow valve from opening unless it detected sufficient gas pressure, and an AC switch on the wall to turn the line power on and off (Figure 2).
The new unit is a gorgeous example of modern electronics, LED touch-screen display, copper piping, and small motor-controlled valves (Figure 3). It looks so nice that it could be used for a microbrewery in the summer. The unit purred along nicely for the first cooling season, all eerily quiet and presumably efficient, then went into the quiescent mode of the warmer months.
It re-started this fall with the first cooler days, and a week later I saw a small puddle around the unit —never a good thing. I checked all the exposed pipes, and they were all absolutely dry, so it wasn’t a leaking joint or condensation.
My conclusion: there was a leak in the boiler’s tank or its associated internal piping, so I called for service. I didn’t want to try troubleshooting this sealed system, especially as it was new, complicated, sophisticated, and under warranty. The risk of messing things up on a tightly integrated plumbing, electrical, and heating arrangement—and making things worse—is a chance I didn’t want to take, especially as it seemed to be working fine other than this small leak. Opening the core unit to look for leaks seemed to be an unwise gamble as things might not go back together as they should (I’ve had some experience there).
Long story short: the service person came, saw the water, and looked at the AC circuit for the small pump, which periodically empties the condensate-neutralizer collection bucket into a nearby drain (if you don’t know what this is and why just look it up). This pump has its own ground fault interrupter (GFI) mounted on the boiler, installed by the heating system contractor’s electrician as required by code (Figure 4).
He took a quick look at that GFI and…surprise: it had tripped, thus cutting off the AC to the pump. Without that pump, the small collection bucket was overflowing. He pushed the GFI reset button, and the problem was fixed.
So the leak was not a leak at all. That I missed this simple GFI problem and the trivial fix was especially embarrassing to me as an engineer; it was almost as if I had called an electrician because an easy-to-reach light bulb had burned out.
The real question is: what caused the GFI to trip? Was there some underlying system problem? We can’t be sure, but about a week before the heating system kicked on for this season, we had a brief power failure (under a minute) with the usual droops, surges, and spikes resulting in lights flickering, clocks stopping, and starting, and desktop PCs going dim as their internal power supplies tried to get some “traction.” Most likely, that event tripped the GFI.
The lesson here is the one that applies to all troubleshooting situations: even if you don’t know the system, it’s OK to at least start with the basics. Don’t be so intimidated by a system’s sophistication that you don’t check the obvious possibilities, especially those which do not require a deep dive or disassembly. So there’s Lesson #1: Don’t be afraid—within reason.
This GFI problem is the counterweight to another engineering-debug characteristic: overthinking problems and their possible causes. For example, I have often wasted time trying to troubleshoot erratic, inconsistent operation in small battery-powered devices such as TV remote controls.
I immediately assume it’s something subtle, so I open them (often a real challenge), clean the gunky conductive-elastomer contacts, look for loose contacts, and do some advanced poking around. Only then do I do what I should have done first, as the problem is far simpler: the batteries aren’t dead but are on the low side, just “OK enough” to sort of get the unit going but not enough to make it work properly. So that’s my Lesson #2: Don’t overthink the problem.
You’d think I would have learned my lesson by now, but apparently, I haven’t—sometimes, the engineering joy of “digging in” where the risk of messing up is trivial drives me to ignore and skip past the obvious. That’s a habit I need to learn to overcome. Maybe I will start by periodically re-reading “Debugging: The 9 Indispensable Rules for Finding Even the Most Elusive Software and Hardware Problems”—the best guide I have ever seen to troubleshooting—to keep the lessons fresh (Reference).
WTWH related content
Tech guide details ground fault detection methods
Teardown: Leviton ground fault circuit interrupter
Basics of ground fault interrupters
The basics of specifying ground-fault interruption equipment
What sensors are used in today’s and tomorrow’s HVAC systems? Part 1
What sensors are used in today’s and tomorrow’s HVAC systems? Part 2
What sensors are used in today’s and tomorrow’s HVAC systems? Part 3
How are gas sensors impacting the HVAC market?