Bearing Failure Leads to Cooked Winding

I ran across a failed capacitor start motor on an air compressor recently. It is obvious that the start winding got barbecued (see picture).

A student asked me how I knew right off that it was the start winding.  Notice that the winding which is burned has smaller gauge wire than the winding that appears OK. The start winding in single phase motors is constructed of smaller gauge wire than the run winding and has fewer turns. It is tempting to call this an electrical failure after seeing the cooked winding. However, most motor failures can be traced back to bearing failures.

Disassembling the motor we saw that the rotor had been dragging – a sure sign of bearing failure. (see picture)

The lead end bearing was to blame in this case. You can see that the dragging all took place on the lead end of the motor. Taking a close look at the stator you can see where the rotor has been rubbing the stator. (see picture)

This can cause two types of winding failures: one where the rotor knocks some of the metal layers into the slots where the windings are, and another where the rotor stays magnetically locked down at startup, which is what I believe happened here. Even though the rotor turns easily by hand and there is no play that can be casually observed by hand, it is obvious the bearing was allowing the rotor to touch the stator. If this happens on startup, the reaction will be like two magnets with opposite polarity pulling together. The motor will lock down, draw high current, and heat up.

Tracing a Communications Problem

One bad apple don’t spoil the whole bunch girl …
You may know the Jackson Five hit that line comes from.
(The first time I heard it was NOT on an oldies radio station.)  Unfortunately, when it comes to communicating controls, one bad communicating board CAN take down the whole network. If the communicating portion of a board is sufficiently damaged, it can prevent communication to ANY of the components on the communicating network. This is because they are all tied together in parallel. A sufficiently low resistance can pull down all communication signals on the whole bus. We saw this first hand this week in our lab. A system that had been operating correctly started having problems communicating. A lab instructor had put in a problem and the student tried to troubleshoot by swapping wires around, resulting in a dead board. (In my experience, swapping around wires just to see what will happen almost never ends well!) Now, the ENTIRE SYSTEM stopped communicating and the control could not find the furnace, which did not have any problems. When the communicating wires to the condensing unit were disconnected, the thermostat found the furnace and was able to operate it. The bad condensing unit board managed to pull down the entire communication bus. When you have a communicating control system that will not communicate, the first step should always be to check your connections. Make sure they are correct and making good contact. Then, try to isolate different parts of the system by connecting one part of the system at a time to the communication bus and seeing which specific component will not communicate. Unlike standard 24 volt controls, you cannot read a voltage to determine if a board is communicating. Isolating the components and checking them individually usually identifies the component which is causing the problem.

Reading Voltage Across Switches

Voltmeters read the difference in potential from one point to another. When a voltmeter indicates a reading of 120 volts, this means that one lead is 120 volts higher than the other lead. We often refer to this as voltage drop. I find that people often misinterpret voltage readings across switches. A switch is designed to either allow current flow, or stop it by opening the circuit. A voltage reading across a switch indicates that the switch is open (off). Let’s take a circuit with a light and a standard toggle switch controlling the light. With the switch off, you will read 0 volts at the light. If you read voltage across the light switch, you will read 120 volts with the switch off. The voltage is dropping across the switch. Now if you turn the switch on, the voltage reading across the switch will change to 0 volts. Checking the light, you now read 120 volts at the light. When current travels through a switch, there should be NO drop in voltage. When current travels through a load, there SHOULD be a voltage drop. A closed switch should have a resistance of close to 0 ohms, while a load should have a measurable resistance. Using Ohm’s law to calculate the expected voltage drop across a switch you would get 0 volts because 0 ohms times any amount of current would still be 0 volts. I have often seen students looking for the break in a circuit read voltage across a switch and declare that switch is OK because they got a reading. In fact, that switch is open because they got a reading. One point to keep in mind is that the voltage across all devices in the circuit should add up to the source voltage. If you are reading 120 volts across a switch in a circuit wit a source voltage of 120 volts, there is no voltage left for the load.

Random Puzzle Assembly

Many students trying to troubleshoot HVACR systems have difficulty navigating the sheer amount of data one can collect. If you go about just collecting a lot of data, it does not take very long to confuse yourself. I often see people taking resistance, voltage, and current readings on just about every conceivable location on the system without having a clear idea why they are taking the readings. This is like trying to build a 1000 piece jigsaw puzzle by just randomly trying to fit pieces together without looking at the shapes or color patterns on the pieces. When building puzzles, it helps to have a plan. There are different plans – outside edges first, organizing pieces by color, or organizing pieces by shape, but having a plan improves your chances of success. Troubleshooting is the same way. You should have a plan to organize your data collection into something meaningful. You should know why you are taking a voltage, resistance, or current reading before taking the reading. Ideally, each measurement you make should eliminate an area of inquiry. A simple example is checking the power supply to the unit. If you read the correct incoming voltage, you can eliminate that as a source of the problem. But if you don’t read the correct voltage at the power supply, then there is no need to check anything inside the unit until you have solved the power supply problem. There are probably as many systems as there are technicians. It is not too important what your system is, so long as it is based on an understanding of how the equipment functions and proceeds logically. For a non-functioning component, I generally want to know if that component is receiving the correct voltage. If it is, then I need to take a closer look at the component. Otherwise, I need to check the circuit that supplies voltage to the component. Many technicians check the line voltage and control voltage first. This does not take long and covers a large number of problems. Another technique is to start at the thermostat and work towards the non-functioning component until you find a break in the circuit, or eventually reach the component. The key point is that each piece of data you collect should tell you something. This will allow you to take far fewer readings and isolate the problem much faster.

Furnace Ignition Sequence

One of the best ways to prepare for troubleshooting any piece of equipment is to learn it’s normal operating sequence. If you understand what is supposed to happen and the order in which nit is supposed to occur, you can narrow the probable causes by observing the unit operation. By carefully observing the operation of a gas furnace, you can often narrow the possible causes of trouble to just a few items based on how far into the sequence the furnace gets. Most furnaces today use many similar components. This includes hot surface igniters to light the burners, induced draft fans to pull the combustion gasses through the heat exchanger, and draft proving switches to insure the draft.

The first step is to energize the induced draft fan. This is the little fan on the front of the furnace, not the larger indoor blower. The induced draft fan runs for a minute or so to purge the combustion chamber. The draft safety switch should close after the induced draft blower starts. If it does not, the process stops there – often with the daft blower continuing to run indefinitely. If the draft switch senses a proper draft, it closes and the sequence continues.

Next, the hot surface igniter is energized. If it is working, you will see a bright glow.

After a warm up period of 30 seconds to a minute, the gas valve is energized. The gas should ignite when it flows over the hot surface igniter.

A flame rod must sense the presence of flames within a few seconds or the system will close the gas valve and start the process again. After several aborted attempts, the system will lock out and quit trying to ignite.

After the flames are ignited, the control turns off the hot surface igniter and waits for the furnace to warm up before operating the indoor blower.

When faced with a furnace that does not operate properly, pay attention to how far into the sequence it gets before the process stops. That can help you decide where to start your troubleshooting.

Simple Troubleshooting

One of the problems with knowing a lot about possible complex issues is that we tend to look for complex solutions to simple problems. After learning how to measure superheat and adjust expansion valves, we send to see every problem as a problem with the TEV. This is similar to buying a car that you swore you never saw on the road before, and then by next week you notice every other car seems to look like yours. Our brains have a kind of data pre-fetch routine that tries to speed up our recognition of our surroundings that leads us to jump to conclusions. It is just how we are wired. Recognizing this, we need a system to prevent us from confusing ourselves by making incorrect assumptions based on very limited input. This is one of the reasons that systematic troubleshooting saves time in the long run. Having a system helps us avoid the temptation to solve the problem by guessing. Although there are many ways to approach a problem, I try to remember to check easy things first. If you are going to run down a list of things to check first, at least make sure the list includes simple things that don’t take long to check and should normally be checked anyway. For example, it is never wrong to check the air filter. You really should always do this anyway, and trying to check the system operation with a dirty air filter just wastes time. Dirty air filters reduce the airflow, causing a host of other problems such as low superheat, floodback to the compressor, furnaces cycling on the limit, burned heat strips, and generally poor heating and cooling performance. Similarly, it never hurts to take a look at the condenser coil to make sure it is clean. Dirty condenser coils can lead to units tripping on high pressure switches and compressor internal overloads opening. Always check to see that the thermostat is actually set to bring the unit on. Don’t assume that the thermostat is set correctly. Customers often don’t know how to set their thermostats, especially with the newer electronic thermostats. If the thermostat uses batteries, a fresh set of batteries will often cure a thermostat that is acting erratically. While you are at the thermostat, set the fan switch to on. If the fan comes on, you know that you have power to the indoor unit and also control voltage. With digital thermostats that have batteries, seeing a display on the thermostat does not necessarily mean you have control voltage because the display can operate from the batteries alone. If a unit will not operate, you need to check power to the unit first, and then check control voltage. The problem will often become apparent during these initial checks. Even if the problem is not discovered during these preliminary observations, you have eliminated many common problems in a relatively short period of time.

Problem Solving

HVAC/R Service is about practical problem solving. All the tools, technology, training, and literature are just there to help us identify and solve problems. The primary skill that any service tech needs is problem solving. The most important tool at your disposal is your mind. I am frequently asked why we make students do BTU calculations of ice turning to steam, series parallel ohms law problems, gas law calculations, or any host of other primarily mental exercises that nearly all HVAC/R students must suffer through. Usually, the students asking are doing the most suffering. Although I can justify all of the above as an endeavor to garner a deeper understanding of the principles which make HVAC/R systems work, I usually tell them that you can’t become a champion weight lifter by lifting marshmallows. Service techs are not paid to connect gauges or take voltage readings, they are paid to solve problems. Obviously techs need to be familiar with all the tools at their disposal, but we should never overlook the fact that their primary tool sits on their shoulders. I am afraid the current focus on standardized testing throughout our educational system has not prepared our students for practical problem solving. Rather, they are used to selecting the best solution from a very limited set of answers. My prescription? Lots of work that requires students to recognize and define the problem, systematically find the cause, and offer a solution. Assign work that requires students to provide written answers. Ask questions whose answers have not been explicitly stated, but require students to put two pieces of information together. If you get an answer that is way off base, try and ask leading questions to help the student work their way through a solution to the problem. The exact answer is not as important as the process. Encouraging students to use their minds to solve problems is crucial. The shop is a great place to work on problem solving because real life problems are never as simple as a, b, c or d. But don’t accept “the part is bad” as an answer. The student should be able to explain what the unit is doing wrong, what is the cause, how they determined the cause of the problem, and what their proposed solution is. Truth be told, I believe this approach could work in a lot of fields besides HVAC/R. The leaders in any field are always people who have recognized and defined problems and then devised solutions.