Outdoor A/C compressor cut-off safety

mike31

After a temporary problem with an a/c compressor, flipping the circuit breaker a few times before it came back on, I noticed (via cheapo AC detector wand) AC being detected on the cover plate of the (outdoor) cut-off switch. It's one of those typical 6"x10" or so metal boxes where the circuit enters, there's a switch on the outside that's metal with a plastic handle, and the box feeds power to the a/c compressor unit.

Details: The circuit takes up three breaker spaces (rated 50A) in the main circuit breaker. (edit: removed possibly incorrect assumptions about the A/C condenser)

Anyway, breaking out a multimeter, if I touch the screws on the cut-off switch box to a surface screw on the a/c compressor unit, there's 60v between them, about half of a single 120v circuit, so there's got to be a connection to one of the 3 circuits, not a good connection, but enough to be dangerous? Testing the cutoff switch box cover against the a/c compressor screws shows anywhere from <1v to 8v depending on where on the cut-off box I'm testing. Testing the metal conduit that leads into the cut-off box from the main circuit breaker shows similar low voltages... but it should be zero I assume.

There obviously isn't enough current flowing to trip the breaker, but I have no idea how much there is and I can't shut it off just now to diagnose it because I need the A/C. (opening the box with it energized is not my idea of fun even with thick gloves... I'm not an electrician, but I stayed at a holiday inn last night...). Also, the screws on the cut-off box can't be very well grounded or there wouldn't be such high voltage between them and the screws on the a/c compressor... I think if they were well-grounded it would completely short and the circuit breaker would trip?

Is this a problem worth looking into, or can there be 60v vs ground on these exterior cutoff boxes without serious consequences? AFAIK it's been like this for a while. I only noticed it because of a power outage today that temporarily screwed up the a/c compressor.

edit... the other a/c compressor, on a different circuit, has what I would expect... wand detects no a/c anywhere around the box or conduit, and voltmeter shows no differential between the cutoff box screws or enclosure and ground.

McSteve

I'd start by turning off the power and opening up that disconnect box to look at the connections. Post up a picture here if you want.

stickboy1375

You should NEVER read a voltage on a metal surface, something is amiss....

__________________

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.

gregzoll

No contact testers will always beep when ever voltage is flowing on a circuit, nearby circuit, or around a metal encased appliance.

Does not mean anything. Unless you have felt a electric shock when touching the chassis, disregard about the beeps that no contact told you.

As for the voltage you read, suggest reading up on Phantom voltage.

__________________

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: Now listen, Cadet. I've got a job for you. See this button?
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: Don't touch it! It's the History Eraser button, you fool!
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: So what'll happen?
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: That's just it. We don't know. Maybe something bad, maybe something good. I guess we'll never know, 'cause you're going to guard it. You won't touch it, will you?

stickboy1375

Gregg, just for your info...

Reading voltage on a de-energized conductor to neutral or ground that is part of a multiwire cable or feeder in conduit where another conductor is energized is referred to by several terms in electrical literature Fluke refers to this as “Ghost” voltages.

basically, Im saying, I think you need to read up on 'Phantom Voltage" And you are nowhere near qualified to dismiss this problem.

__________________

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.

mpoulton

You're rarely wrong, but I think you missed this one. A non-contact tester will rarely ever give a positive test on a properly grounded metal surface, and you definitely will never read a "phantom voltage" (or any significant voltage, for that matter) between two properly bonded metal surfaces. There is clearly a real bonding problem here. To the OP: post pictures of the connections inside the AC disconnect, please, and also of the connections in the main panel if you can.

Auger01

First thing I would do is open up the AC unit and take a good look at the capacitor and then check for continuity between the ground lug and the disconnect ground. It should be almost zero but I dont think yours is.

gregzoll

mpoulton, if they are touching around the disco with the no-contact, depending on how sensitive it is, it will pick up the live voltage. Mine will do it if I get within a foot or so of a light the disco for the a/c, an appliance, etc., without even touching it.

The thing about non-contacts, they are only as good as not being able to find a pencil when you need it. In other words, do not trust them, and leave them where you last placed them.

__________________

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: Now listen, Cadet. I've got a job for you. See this button?
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: Don't touch it! It's the History Eraser button, you fool!
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: So what'll happen?
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: That's just it. We don't know. Maybe something bad, maybe something good. I guess we'll never know, 'cause you're going to guard it. You won't touch it, will you?

mike31

http://i.imgur.com/IWB4H.jpg

With main breaker on and cut-off lever in the "off" position (lever is to the side out of the picture frame), There's ~55v AC on the top three blocks, ~120v on the bottom three blocks (measured against the ground wire visible at the bottom).

I guess that's due to whatever's going on inside the compressor? If I left it off long enough that would go to zero? Measuring against the ground wire, there's no significant voltage on the cover or the metal conduit leading back to the main breaker panel (all <3v or so). The conduit from the cut-off to the compressor is not metal so there's nothing for me to check there.

So, I'm thinking there's no really anything obvious to be concerned about? Except... the multimeter was very clearly in AC voltage mode... should the 55v on the load side have been AC? I have no idea what the compressor circuitry is like.

gregzoll

Unless you discharge the capacitor, yes it can still give off residual stored voltage. Along with also most likely reading phantom voltage, even if power is off, the capacitor can still cause it with certain meters.

__________________

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: Now listen, Cadet. I've got a job for you. See this button?
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: Don't touch it! It's the History Eraser button, you fool!
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: So what'll happen?
To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
: That's just it. We don't know. Maybe something bad, maybe something good. I guess we'll never know, 'cause you're going to guard it. You won't touch it, will you?

mpoulton

Wait a minute... You're reading 120V between ground and the right terminal block with the white wire in it? This is totally screwed up. You mentioned that this circuit has a 3-pole breaker in the panel. We need to see that.

This is definitely not wired properly, but exactly what all is wrong is still hard to say. The green wire seems to be improperly used as a current carrying conductor, but I can't tell what for unless this is a 3-phase circuit. The white wire should not pass through the disconnect, it should remain connected at all times. The condensing unit appears to be completely ungrounded since the green wire has been misappropriated for some other purpose.

mike31

If the condenser is ungrounded, that would explain how the surface of the condenser (or a screw on the surface) is showing a voltage vs the screws on the cut-off switch box, right?

breaker for the circuit:
http://i.imgur.com/Jk3cK.jpg

On the previous image, the top three blocks were in the 45-55v range the first time I measured, and they were all approx. 55v the second. The bottom three were all 120v. The first two (left) on the bottom were 118-120 and the third (white), the right-most, was like 124v.

dmxtothemax

Clearly it is NOT grounded properly !
As for ghost voltages or phantom voltages,
Its not unheard of for some motors to inductively couple small currents into the frame.
But they are usually small, and will go away if grounded properly.
Get your grounds checked !

jlmran

After seeing the pics, I'm not surprised there are strange problems.

stickboy1375

No kidding, no grounding at all, on the load side... what a hack job.

__________________

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.