Stay connected with us...

• entries
168
492
• views
28,035

# What is Ghost Voltage?

8,601 views

Ghost voltage is a term that you'll hear used in tech circles, and often incorrectly. Ghost voltage is the name of a very specific phenomenon, but I've seen it used variously to refer to failing under load, high resistance connections, and even simple open circuits. What does it really mean?

What we call ghost voltage is transient, seemingly sourceless voltage. It does, of course, have a source. You know how when current flows through a conductor, it produces a magnetic field? Well that magnetic field can, in turn produce voltage in another conductor. That produced voltage will then just hang out in that conductor, just like charge does in a battery or a capacitor, until it is provided a path to neutral.

Does this mean that ghost voltage can power a load? Not at all -- the moment there's a path to neutral, ghost voltage bleeds off instantly. So why is it that your voltmeter can detect it? That's because a standard voltmeter (or VAC function on any multimeter) is designed to provide as much impedance to current flow as possible. The intent is so that you can get a measurement while affecting the circuit as little as possible. This high impedance means that your meter won't bleed off the ghost voltage, instead reading it as some funny-looking number of volts.

The bottom line is that, when doing AC voltage measurements, you should always use a loading meter. There's simply no reason not to. It will give you more accurate readings, it will confirm whether a power supply is actually capable of passing current, and it will keep you from getting tricked by any spooky ghost voltage.

This short video will give you a show you a real-world ghost voltage situation:

• 5

Respectfully I disagree with this.

When you are measuring voltage you are measuring the voltage potential difference between your two probes. So if your reference is L1 (verified supply) which has 120v (reference to N/Ground), and your other probe is on the open side of the element relay, you DO have a voltage difference 120V. both sides of the element are open, not connected to anything so its voltage potential is ov, it can’t do any work. L1 does have the potential to do work in the amount of 120v so the difference between those two points is 120v.  But since both sides of the element are open there is no path to ground, so yes, when you try and measure voltage with a load it is going to drop out.

But to measure between those two points and come away with an answer that there is no voltage (potential difference between your probes) is even more confusing and tells you nothing constructive, and logically would most likely lead to one of two incorrect assumptions. First it could make you think there is NO voltage on L1 (but you would get shocked if you touched it) — if you are thinking in terms of L1 being same voltage level as open element, output of open relay); second, it could lead you to think there is 120V on the open element circuit, which there is not — if you are thinking the open element has the same voltage as L1.

The real ghost voltage is actually about 12v. Because you are measuring between L1 and a floating wire, you have one probe on L1 that is a true 120v (verified source) but your other probe is connected to a point that can float. In this instance it is floating about 12v above ground (120-108) so when you look at your potential difference between the open element (output of relay, when relay is open) you are measuring from that floating level of the open element to L1 at 120v. Now if you took your meter and measured voltage from the open element to ground/neutral (high impedance) you might get that 12v and then if you switched over and measured it loaded (loZ) it would go away. Because there is no real potential difference between that open element and ground.

When testing for voltage you have to have a known reference point in which to test from, reference or you don’t know what your results are telling you. Like on an oven that only switches L1 and not L2 if you have the bake circuit on (bake not working) and you are getting 120v (reference to ground) at a given point, you don’t know if that voltage is from L1 or L2.  But if you reference L1 (meaning one probe on L1, and you have verified your supply) and you get 0v you know that point has the same potential difference as L1 — 120v,  and not L2 (that would give you 240v).

Stronghammer, there should be no potential voltage difference because P1 is floating when the bake element is off; there is no path for electricity to flow. The voltage measured was stray inductance coming from nearby wires and that is what ghost voltage is as I understand it.  What I’ve taken away from this post is the LoZ function helps us as techs by verifying the voltage we are measuring is real.  Also, measuring LoZ voltage where there shouldn’t be any can alert us to the possibility of a dangerous faulty wiring situation.

48 minutes ago, 10slaughter said:

Stronghammer, there should be no potential voltage difference because P1 is floating when the bake element is off; there is no path for electricity to flow. The voltage measured was stray inductance coming from nearby wires and that is what ghost voltage is as I understand it.  What I’ve taken away from this post is the LoZ function helps us as techs by verifying the voltage we are measuring is real.  Also, measuring LoZ voltage where there shouldn’t be any can alert us to the possibility of a dangerous faulty wiring situation.

Yes, pin 1 of P1 is floating when bake is off since both sides of the element are open. But L1 is at 120v. I guarantee you that you would get a shock if you touched L1. More than that, if you took a light and connected one wire to pin (4 of P1) L1 and the other to neutral it would light up and stay light up as long as it was connected. When you have one of your probes on L1 (and L1 has 120v) unless you have your other probe on another point where L1 is also present you should have 120v.

If you did this tested they did in the video and instead of going between pin 4 and pin 1 of P1 you went from pin 1 of P1 to ground or neutral then any voltage you saw would be a ghost voltage.

Posted (edited)

15 minutes ago, Stronghammer said:

If you did this tested they did in the video and instead of going between pin 4 and pin 1 of P1 you went from pin 1 of P1 to ground or neutral then any voltage you saw would be a ghost voltage.

I don’t think you understand what is meant by ghost voltage. What you’re talking about here is actual real voltage

Edited by EthanRanft

You are correct that there is no path for electrical flow through the open bake circuit. But testing voltage isn’t testing current flow, to test that you check amps. Voltage tests are to determine a potential difference. I.e. if the path is completed current will flow. But the way you would test for ghost voltage on the open bake element is to ground, not to L1. When you test high impedance from L1 (pin 4 of P1) to the output of the bake relay (pin 1 of P1) the impedance is preventing them from being connected electrical through the meter. hence the reason you can have one probe of your meter on a hot line and you can touch the other without getting shocked. The reason the voltage goes to 0 when the test is switched to low impedance is then the bake element becomes electrically the same as L1. While doing this LoZ test, If you took another meter and checked from the bake element (pin 1 of P1) to ground or neutral you would see 120v because L1 is being allowed through the meter to the element.

5 minutes ago, EthanRanft said:

I don’t think you understand what is meant by ghost voltage. What you’re talking about here is actual real voltage

That’s my point, L1 has real line voltage on it. If you want to check for ghost voltage on the bake element you need to go from the element to ground, or take the difference between L1 and what you read on the meter (120-108=12) 12 is your actual ghost voltage. If checked high impedance from the open bake element to ground you would see that 12v and then switching to LoZ it would go away.

4 minutes ago, Stronghammer said:

L1 has real line voltage on it.

Yes, but with respect to ground or a good neutral. But not with respect to P1-1. With the L2 relay open, P1-1 isn’t connected to anything, so there’s no voltage potential across L1 and P1-1

Posted (edited)

18 minutes ago, Stronghammer said:

If you want to check for ghost voltage on the bake element you need to go from the element to ground

Why would “checking for ghost voltage” be a thing or give you any troubleshooting information? Again, I don’t think you understand what ghost voltage is

Edited by EthanRanft
• Team Samurai

@Stronghammer, it sounds like you're not talking about the same concept we are. Before this discussion goes any further, you should review these materials from Fluke on ghost voltage:

And check out this webinar we have on the subject as well. It goes into the topic in great detail:

• Team Samurai
58 minutes ago, EthanRanft said:

P1-1 isn’t connected to anything, so there’s no voltage potential across L1 and P1-1

Just one small correction to this, @EthanRanft -- there actually is a potential difference. But this is capacitively coupled voltage created by the changing polarity in Line voltage. It's not connected to a power supply, so it can't sustain electron flow when a load is placed across it. No electrons = no power = no work  And power (watts) is the product of voltage and current:

P = I x E

if I is 0, P is 0

In other words, loads are all about watts. Watts are what make a load do its work. And ghost voltage is incapable of delivering watts.

Think of it like static electricity. It's a real difference in potential, but as soon as a path to neutral/ground is provided, the ghost voltage instantly discharges, just like static electricity does when you touch a doorknob.

2 hours ago, Son of Samurai said:

Just one small correction to this, @EthanRanft -- there actually is a potential difference. But this is capacitively coupled voltage created by the changing polarity in Line voltage. It's not connected to a power supply, so it can't sustain electron flow when a load is placed across it. No electrons = no power = no work  And power (watts) is the product of voltage and current:

P = I x E

if I is 0, P is 0

In other words, loads are all about watts. Watts are what make a load do its work. And ghost voltage is incapable of delivering watts.

Think of it like static electricity. It's a real difference in potential, but as soon as a path to neutral/ground is provided, the ghost voltage instantly discharges, just like static electricity does when you touch a doorknob.

Ah, so the more accurate statement would just be “no voltage” as opposed to “no voltage potential”?

5 hours ago, EthanRanft said:

Why would “checking for ghost voltage” be a thing or give you any troubleshooting information? Again, I don’t think you understand what ghost voltage is

5 hours ago, Stronghammer said:

Yes, pin 1 of P1 is floating when bake is off since both sides of the element are open. But L1 is at 120v. I guarantee you that you would get a shock if you touched L1. More than that, if you took a light and connected one wire to pin (4 of P1) L1 and the other to neutral it would light up and stay light up as long as it was connected. When you have one of your probes on L1 (and L1 has 120v) unless you have your other probe on another point where L1 is also present you should have 120v.

If you did this tested they did in the video and instead of going between pin 4 and pin 1 of P1 you went from pin 1 of P1 to ground or neutral then any voltage you saw would be a ghost voltage.

Based on your comment that when you see a voltage high impedance that then goes away when switch to LoZ means it is not supposed to be there and there is a problem, is exactly why I don't agree with this video. Because pin 4 of P1 is supposed to have 120v on it. The difference between the measurement of 108v and the 120v it is supposed to be, is the ghost voltage (12v) and that doesn't tell you there is a problem, it really just says you are not using a good reference to determine if you have proper voltage input to your L1 bake relay.

Posted (edited)

47 minutes ago, Stronghammer said:

and there is a problem

The video doesn’t indicate there’s anything wrong with this oven.

47 minutes ago, Stronghammer said:

and that doesn't tell you there is a problem, it really just says you are not using a good reference to determine if you have proper voltage input to your L1 bake relay.

The method shown in the video is not intended to show the proper way to verify L1 to the bake relay.

In it’s current condition, off, there should be no voltage across P1-1 and P1-4. Not 120V, and not 108V as is read with a high impedance meter

Edited by EthanRanft
• Team Samurai
16 hours ago, Stronghammer said:

Both of your measurements were taken with a probe on P1 pins 1 and 4. Pin 4 of P1 being tied to L1 and such has real 120v present, such that hooking a load from pin 4 of P1 to neutral would accomplish real work. Like turning on a light bulb, operating a motor, etc.

I think I see what's tripping you up about the example shown in the video. This is a 240 VAC circuit, and so a reading of 108 VAC was identified as ghost voltage in this particular circuit configuration, since we're looking for a good 240. Doesn't matter if you have a good L1 at P1-4 and you can light a light bulb by connecting P1-4 to either chassis or Neutral— in this circuit, there is no Neutral and you need a good L2 and L1 in order for the load (the bake element) to function. L1 by itself is incapable of delivering electrons to the intended load without a return path to the power supply. That's why the voltage difference in that circuit is identified as ghost voltage for the intended load

But the L2 conductor is floating, and as you can see from the measurement of 108 VAC, it has significant ghost voltage capacitively coupled between L1 and L2. With this 240 VAC circuit as it is physically configured, this is definitely ghost voltage for the intended load (the bake element).

Important point: ghost voltage can occur in ANY open conductor in close proximity to a live conductor. It does not need to be Neutral.  When we talk about ghost voltage, it is always in the context of between two specific points in a specific circuit and a specific intended load. If a circuit uses L1 and L2 (no Neutral) and L2 goes open, then you can read ghost voltage across L1 and L2 in that circuit

My LoZ reading told me that one of those Lines was missing. You critiqued not testing from P1-4 to Neutral for voltage. That may have been my next step: physically half split the circuit to determine which Line was missing. But the specific scope of this video was to demonstrate ghost voltage in a live circuit (as stated in the title of the video). This was not a troubleshooting video per se, but a demonstration video showing a phenomenon that techs should be aware of when working with AC circuits.

One other concept to keep in mind: we never talk about voltage at just one point. Voltage is always the difference between two points. There is no 120 VAC “out there” somewhere; voltage is always relative to some other point. That’s what we are actually measuring with the two probes of our voltmeter: the voltage difference or the difference in electrical potential between those two points. That’s why technically you can’t simply say you have 120 VAC between P1-4 (L1) unless you specify that that value is with reference to Neutral. With reference to P1-1, it’s just ghost voltage that bleeds off on LoZ.

So we don't say that the voltage at a particular point is ghost voltage, we say that the voltage between these TWO particular points is ghost voltage. In this case, the voltage between P1-4 and P1-1 (the circuit for the load of interest) is ghost voltage but the voltage between P1-4 and Neutral is not.

• 2