Tag: electrical diagnosis

There are many great diagnostic tools available to the service technician today, but I shaven’t found a tool as versatile than the simple isolation diagnosis.

There are many ways this concept can be applied but let’s start with some examples so you get what I mean.

Low Voltage Short Circuit Isolation Diagnosis

You arrive on a no cooling service call at a home and you find the system is off and doesn’t respond to any thermostat settings. You check the breaker and the condensate switch at the furnace by force of habit and look at the door switch, everything looks in order so far.

You take a look at the 3A low voltage fuse and SURE ENOUGH, visually blown.

Now there are many schools of thought of how to proceed from here but I prefer to use logical possibilities and process of elimination before starting to tear everything apart.

First,  look at any possible rub out points on low voltage wires in the furnace, visually inspect the safeties, go outside and check the control wire both inside and outside the condensing unit and ESPECIALLY wherever wires run across copper tubing. If you find nothing at these common failure points I will pull open the thermostat there and check that it’s wired correctly with no exposed copper.

You will notice that I didn’t just “replace the fuse”, because science and experience has shown me that fuses don’t just fail open on their own.

If I still find nothing then the “logical possibilities” part of the diagnosis has ended and we move into the isolation diagnosis part of our testing.

Some techs prefer disconnecting all the low voltage wires at the furnace and ohming them one at a time to ground and common at this point. Sometimes this works but I prefer to let the system work for me rather than pulling apart all the wires at this stage.

I replace the blown fuse with a 3a re-settable fuse / breaker like the one shown above with the thermostat off in all modes (to keep from wasting fuses). I then close the door switch and see if the fuse blows, if it does not then we can determine that there isn’t a continuous short circuit in R.

I then use a jumper wire and connect each switch leg (G,Y,W,O,W2,Y2 etc…) to the R terminal quickly and see if any of them trip the fuse or throw a spark (remember this is 24v it’s not going to hurt you). Often you will find the circuit with the short just that simply and then you can further isolate until to find the exact part of the circuit causing the issue.

Let’s say the Y circuit is the one blowing the fuse, you can then separate the two wires going to the condenser and the thermostat and tap one side at a time to R, if the conductor going outside is the one throwing the big spark / blowing the fuse then that is the direction to focus. You then go outside and disconnect the Y wire at the condenser wire nut or terminal block. If the short continues you know it is in the wire between the condenser and the furnace and not in the condenser. If it continues then you know it is in the condenser or in the contactor itself.

Shorted Compressor Isolation Diagnosis

Compressors shorted to ground or “grounded” is a pretty common diagnosis in the field and leads to an expensive compressor or system replacement. It’s important that we get that diagnosis correct and take away all the guess work so I use a combination of diagnostic tools and isolation diagnosis to ensure I never get it wrong.

When I walk up to a condensing unit and it has a tripped breaker, the first thing I do is reset it and see what happens…


The breaker tripped for a REASON and every time I reset that breaker I run the risk of creating a major arc. If that short is inside the compressor I add carbon and acid to the refrigerant every time I reset that breaker.

So once again we FIRST perform a full visual inspection of all the high voltage wires, terminals, contactor, capacitor, crankcase heater and the breaker itself. If that all looks OK then we pull the top and inspect the compressor leads and terminals themselves. Before you go pulling on the terminals or compressor plug make sure you are wearing gloves and safety glasses, it is possible for one of those terminals to blow out of the fusite at that moment and freeze your face or hands.

Another quick tip is take a photo of all wires and/or tag them before pulling them off. We all have cameras in our pockets nowadays which makes mis-wiring upon reassembly a thing of the past (I wish).

Now measure resistance to ground (In this case the copper stubs on the compressor) from each terminal. It is OK to use a megohmmeter if you want, just keep in mind that some compressors are still considered “good” by the manufacturer all the way down to 0.5 megaohms and some meters say “bad” at 20 megaohms.

Once you are certain as you can be that the compressor short to ground is the culprit or if it is reading between 0.5 and 20 megaohms to GROUND making you unsure I have one more task for you.

Now ISOLATE the compressor by taping and strapping up the plug or terminals so they aren’t touching anything. Now reassemble the unit and reset the breaker.

If it doesn’t trip and everything else (Condenser fan motor etc..)  runs properly then you can feel good about your diagnosis, if it trips again then it’s back to the drawing board.

Other Uses 

Isolation diagnosis can be used in things like finding system noises and even in finding open circuits by using jumper wires. Isolation diagnosis is taking a hypothesis and testing it with one component or conductor at a time allowing you to find the culprit through process of elimination.

On communicating systems where I wanted to be SURE it was the controller and not the wire I removed the controller from the wall and wired it directly to the fan coil board to make sure it still didn’t work even with no wire between. Sometimes my hypothesis was right and sometimes it was wrong but either way isolation diagnosis has saved me from looking dumb on many occasions.

— Bryan


P.S. – Bert Made this video a while back using isolation diagnosis to find a LV short

Disclaimer: Ghost voltage is a term used by techs to explain a phenomenon where they measure voltage they don’t expect or when the voltage they see doesn’t do the work they expect. More advanced techs know how to use the Lo-Z (Low impedance) mode on their volt meter if it has one to help eliminate this. The vast majority of what techs call “ghost voltage” is just a circuit with high voltage drop under load and not stray inductance from other conductors. I write this first so that more experienced techs understand the context of this article. 

This article serves two purposes. First, it is an article for technicians who have heard of the dreaded “ghost” voltage but never understood why it happens. Second, for my own apprentices and techs who I stumped this morning with a diagnosis problem that involved “ghost” voltage that they were unable to diagnose.

If they read my tech tips they will get the answer… sneaky right?

 So what is meant by ghost voltage?

In some cases, you will be diagnosing an electrical issue, usually controls / low voltage issue. You will be measuring potential on a circuit and then when the circuit is connected to the load the voltage will disappear … like a “ghost”.

For example, you make be measuring 24v at a condensing unit on the “Y” contactor circuit when the conductor (wire) is disconnected, but as soon as you connect it to the contactor/control board the voltage “disappears” when measured across the load (across the contactor coil) or more simply from Y to C.

In other cases the voltage may not disappear completely, it may just drop way down, or in other cases the contactor may chatter, circuit board lights dim etc…

I have heard all of these situations called “ghost” Voltage, but they are actually just voltage drop and these symptoms are caused by additional resistance in the circuit OTHER than the designed load.

Quick Note: there are also “induced” voltages that can appear as ghost voltage due to conductors running in parallel with other current carrying conductors. This is more common in Commercial and industrial applications where many wires are bundled or in close proximity over long distances. These charges are usually small and often “disappear” under load.

Rarely do we want more than one electrical load (resistance point) in a single circuit. When this does occur it is usually undesigned and caused by of long wire lengths, improperly sized wire and poor connections.

Now to CLARIFY, when referring to a circuit we mean one complete path between electrically different points (say L1 and L2 in single phase 240 or 24v hot to 24v common on a control transformer). Some think of parallel circuits as a single circuit, but while they may share conductors they have an individual load path.

To cut to the chase, whenever wire is undersized, runs of wire are too long or the circuit contains poor connections there will be additional resistance introduced to the circuit. When there is more resistance added in places other than the load (in this case a contactor coil) there will be a voltage drop and therefore the voltage applied to the load will be decreased. When a wire isn’t connected to the load this drop will be invisible because the load isn’t in the circuit and therefore you are simply reading across the OTHER, unintended load (resistance) which will often be the full voltage depending upon the exact issue and when you are making the measurement.

In every complete and independent circuit, including a series circuit, the amperage is the same no matter where in the circuit you measure it. Before the load, between loads, after the loads… it doesn’t matter. The amperage is dictated by the total applied voltage and the resistance (or more accurately the impedance) of the entire circuit.

The voltage applied to each load is dependent on the resistance of the load in comparison to the total resistance of the circuit. In the example below, you can see that the amperage is the same on each load and is dictated to be 500 micro-amps because the total circuit ohms is 18,000.

The voltage drop of each load in series is equal to it’s percentage of the total circuit resistance. Since  load R1 is 16.5% of the total resistance in the circuit, the voltage drop across R1 is 1.5V because 1.5 is 16.5% (0.165) of 9V.

There are a few other factors that make the trouble with voltage drop worse. Let’s say you use an undersized wire to feed a lightbulb, an undersized wire means that the conductor has a lower ampacity (amp capacity) than it should have. Once the circuit is energized the wire will begin to heat up, as it heats up the molecules in the wire begin moving faster which increases the resistance of the wire. The greater the resistance of the wire the greater the voltage drop across the wire resulting in a hot, dangerous wire, increased voltage drop at the bulb, less light from the bulb and decreased circuit amperage (less total work being accomplished).

In the case of many loads including inductive (magnetic) loads like a compressor contactor, the resistance in the coil isn’t just resistance you can measure with the contactor de-energized. This resistance that is created within an electromagnet once it is energized is called “inductive reactance” and it is measured in ohms of impedance. In order for the contactor coil to properly engage it requires the correct applied voltage and without the properly applied voltage, the resistance of the coil remains low. The crudely drawn diagram below (I’m no artist) shows a contactor coil circuit with no issues and a 0.5 amp  current at 48 ohms

When you add in a 200 ohm “bad connection” or any other type of resistance, not only does it create huge voltage drop, it also drops the impedance of the contactor coil itself with the result being a very low applied voltage (3.13V) on the contactor coil with it connected and under load. Under these conditions, the contactor won’t try to pull in at all. Under less extreme conditions it may chatter or become noisy.

Now, this is a hypothetical situation, but you will notice that the poor connection is AFTER the contactor coil in what we call the common circuit in 24v controls. It doesn’t matter WHERE in the circuit resistance is added, whether before the switch (in this case a thermostat) in the line side or after the switch on the load side. It could even be in common or in the switch itself.

Anytime additional resistance is added to a circuit it results in voltage drop when the circuit is intact. When we disconnect wires to test voltage or test voltage with a circuit that has an open switch we can create confusion and observe “ghost” voltage. In reality it is simply extreme voltage drop caused by additional resistance in series with the load.

— Bryan

Scroll to top
Translate »

Daily Tech Tip

Get the (near) daily Tech Tip email right in your inbox!
Email address