Most HVAC writing online is either a sales pitch or a repeat of a number nobody can trace. These posts cite the study, name the authors, and link the document. Where good data doesn't exist, we say so instead of inventing it.
Last Thursday we put monitors on thirty air conditioners at one of the largest student housing communities in the country, twenty of them on rooftops, in a single day. Standing on those roofs made something clear about how apartment HVAC gets maintained, and it is not the story most people tell.
Read itAny product watching refrigerant pressure can fire an alarm. The engineering is in the conditions where it stays quiet — and each of these would trip a simple threshold product on a perfectly healthy machine.
Read itLow airflow and low charge produce overlapping symptoms — an iced coil, poor cooling, a system that cannot keep up. The fix for one makes the other worse, and the field data says airflow is present in more than half of systems.
Read itEvery threshold that makes sense on comfort cooling is wrong on refrigeration. A freezer running perfectly sits 56 degrees below the line that means an iced coil on an AC — which is why generic monitoring puts refrigeration in permanent alarm.
Read itR-448A and R-449A do not boil at one temperature. They boil across about ten degrees — which is larger than most of the deviation bands anyone judges a system on. Read the wrong end of that range and the error swamps the signal.
Read itA portfolio dashboard that scores a unit it cannot see is reporting our connectivity problem as your equipment problem. We had that bug. Here is what it looked like and why the fix matters to anyone buying monitoring.
Read itEveryone worries about low refrigerant. The field measurements find almost as many systems with too much — usually because somebody topped one off by pressure instead of weighing it in.
Read itNon-condensables are a real fault with a real mechanism, and they get blamed for a great deal they do not cause. A DOE field and laboratory study puts the effect of air alone at quite minor.
Read itNIST tested air conditioners from 82 to 130 degrees ambient and measured capacity and efficiency falling almost linearly with condensing temperature. A fouled coil does the same thing to a system that a heat wave does — permanently.
Read itFrequent starts are widely blamed for killing compressors. The energy penalty is measurable and modest. The lifetime penalty everyone quotes has never been published at all.
Read itThe metering device decides which number means anything. On a TXV, subcooling reads charge and superheat does not. On a piston, it is the other way round — and normal subcooling there spans an eighteen-degree range.
Read itEvery monitoring product implies it makes equipment last longer. We went looking for the study behind that claim. The Department of Energy has listed it as an open research question since 2011, and it is still open.
Read itThree field surveys of failed equipment, one covering roughly 6,000 service records, agree on something counterintuitive: the fault that generates the most service calls is not the fault that ends the machine.
Read itEleven field studies, more than 2,600 systems measured in real homes. Only about a quarter held the charge the manufacturer specified. NREL has calculated the national energy cost of that, together with airflow faults, at 20.7 billion kilowatt-hours a year.
Read itA fault that removes capacity does not stop a system working. It makes it work longer for the same result — and field data shows operating hours predicting how long equipment survives. One line of arithmetic, with real consequences across a portfolio.
Read itI build refrigerant-side monitors for a living. The one on my own house caught a fault my house had been carrying for more than a year — and then my first estimate of what it had cost me turned out to be wrong by a factor of seven.
Read itThe Department of Energy reviewed decades of field testing on installed residential HVAC. The finding is not that some systems are broken. It is that almost all of them are degraded, and almost nobody knows it.
Read itNIST put numbers on what a refrigerant charge fault actually does. The more useful finding is the one about how differently identical faults behave on different equipment.
Read itA smart thermostat measures the room. The room is downstream of the ducts, which are downstream of the coil, which is downstream of the refrigerant circuit. By the time the room is wrong, everything upstream has been wrong for a while.
Read itCooling is the single largest electrical end use in American homes. That size is exactly what makes a degraded system so hard to spot on a utility bill.
Read itASHRAE publishes a median service life for residential air conditioning. A median is not a promise — it is the middle of a distribution, and half of all equipment lands on the wrong side of it.
Read itIt is not coincidence and it is not bad luck. It is margin. A degraded system keeps up fine until the day the load exceeds what is left of its capacity — and that day is the same day for everyone.
Read itThe CDC has counted heat-related deaths in the United States for two decades. The pattern in that data says something specific about mechanical failures in occupied homes.
Read itScheduled maintenance is genuinely valuable and we are not arguing against it. But a once-a-year inspection samples a continuous process 0.3 % of the time, and faults do not wait for the appointment.
Read itEvery car sold in the United States since 1996 carries standardised onboard diagnostics. The most expensive appliance in your house has nothing of the kind, and the reason is regulatory history rather than engineering difficulty.
Read itFor most single-family rental operators, the HVAC monitoring system is a resident noticing the house is hot. Consider what that instrument actually measures, and how late it fires.
Read itThe premium you pay on an emergency call is not paying for a better repair. It is paying for the fact that you found out too late to choose when.
Read itHVAC replacement is one of the largest recurring capital items in a rental portfolio, and most operators forecast it from installation dates alone. There is a reason that forecast keeps being wrong.
Read itA multifamily property will instrument its access control, its cameras, its leasing funnel and its utility billing. The mechanical equipment that residents actually depend on is monitored by nothing.
Read itThe service agreement is the most valuable product in residential HVAC and the hardest to renew, because what the customer wants is not something a scheduled visit can actually promise.
Read itThermal mass is the enemy of early warning. By the time a box thermometer moves far enough to trigger an alarm, the refrigeration system has been failing for hours and the product has been warming for most of them.
Read it