Good afternoon,
I am looking for help with a problem as I am a total newbie to electrical projects.
Here is the scenario: I had an existing metal box which has metal conduit attached to it from bottom and top. Bottom contains the hot wire which then feeds recessed lights on top.
I replaced the existing wall switch with a new timer and found that it would not fit in the existing metal box. I bought a deeper plastic box and replaced the metal box with it. After this, the recessed lights did not turn on. I noticed that my "pen" voltage tester now detects current throughout the metal conduit leading up to the recessed lights. I am assuming that by removing the metal box, I have a ground problem?
Inside the metal conduit there is only the hot wire and the neutral but no ground.
By removing the metal box you have likely removed the ground path. That should not have any affect on the operation of the circuit. A ground is used as a separate path back to the panel should a fault occur within the circuit. Don't rely on those non contact voltage testers they sometimes pick up stray current and give false readings. I would start by removing the timer switch that you installed and test for voltage at the hot wires with either a neon type tester or a volt meter. also you should replace that plastic box with a metal one to restore the ground.
Thanks - Here's another observation. I tested with the volt meter and assured that voltage was present at the switch. Here's what I found:
1. I tested the hot wire going into the switch - Power is ok.
2. I connected the switch, turned it on, and test the "out" wire on the switch to assure power was flowing through - power is ok.
3. I connected the "out" wire I just test in step 2 above to the wire going to the recessed lights. As soon as connect it, there is no power. I remove the connection to the fixture and there is power again.
What would cause this?
Thanks - Here's another observation. I tested with the volt meter and assured that voltage was present at the switch. Here's what I found:
1. I tested the hot wire going into the switch - Power is ok.
2. I connected the switch, turned it on, and test the "out" wire on the switch to assure power was flowing through - power is ok.
3. I connected the "out" wire I just test in step 2 above to the wire going to the recessed lights. As soon as connect it, there is no power. I remove the connection to the fixture and there is power again.
What would cause this?
With the switch in the off position, electricity does not flow through it so there is no power on the "out" side. Power does flow through it just fine without the fixture connected to it. The problem arises as soon as I connect the wire of the fixture to it.
This is a dimming switch. I eliminated the switch as being the problem by swapping it out with a working one.
Rivers, I am quite aware as to how a switch works, I was asking to ensure you know how voltage works, voltage measure across an open switch will read line voltage, voltage read across a closed switch will read 0.
Voltage read between a closed switch the end of one wire disconnected from a load to the load wire will read line voltage.
My guess is you either have a bad timer switch, or you didn't wire it correctly.
There may be a long standing mistake in the original wiring whereby the conduit was used as a neutral.
So do not touch the conduit with bare hands for the time being. For kicks, measure voltage between the two conduit ends as they enter the plastic box. Or, if you are daring, turn the switch to on and, with one hand in a pocket, get a piece of wire to bridge the two conduits for a minute or two. If that lights the light then you definitely have a ground fault (hot or neutral touching the conduit.
After that, keep the breaker off until you find and eliminate the accidental or intentional connection between neutral and ground.
Thanks for all the input guys.
AllanJ - I tested voltage between conduit ends and found a reading of 70 to 75 with the switch off. When I turned on the switch, the reading went up to 110 to 115. I wasn't daring enough to bridge them by holding the wire. Can I get the same result by taping a wire to both ends with electrical tape?
Mark, I've tested that switch with another one in the house by swapping it. Wired it the same way and it worked in another location.
Thanks for all the input guys.
AllanJ - I tested voltage between conduit ends and found a reading of 70 to 75 with the switch off. When I turned on the switch, the reading went up to 110 to 115. I wasn't daring enough to bridge them by holding the wire. Can I get the same result by taping a wire to both ends with electrical tape?
Mark, I've tested that switch with another one in the house by swapping it. Wired it the same way and it worked in another location.
OK.
Then:
One wire from bottom conduit is connected to switch
One wire from top conduit is connected to switch
The other wire from the bottom conduit is connected to the other wire in the top conduit.
Is this correct?
Yes, it was working just fine for about a week. I got around to changing the box afterwards.
While I don't understand the logic, I'm going to replace the plastic with metal again. It is the only thing that changed.
In your one reply you said you measured voltage on the conduit. This is not good:no:
Do you have access to the wiring in the first light the switch feeds?
Wouldn't the metal conduit act as a ground? If he read from hot to the conduit, that does not mean the conduit is hot. Right? Just means it is grounded. Right?
Now if he got voltage from the conduit to the neutral, different story, right?
Normally conduit has to be used with metal boxes, but you can get away with this: bond around the plastic box. This is done by clamping a bare jumper wire of the same size to both of the conduit ends. (Breaker off when you do that)
But before you bond around the box or before you put back a metal box, you need to verify that there is no ground fault (or intentional use of the conduit as a neutral) on the far side (downstream; away from the panel). Use a 5000 ohm resistor instead of the bare jumper wire and after that turn the power back on and measure voltage across the conduit ends.
The purpose of the 5000 ohm resistor is to make sure you are not measuring phantom voltage. An experienced person can tell the difference between phantom voltage and an actual ground fault using other methods.
Imagine the following circuit. Panel breaker to hot wire. Hot wire to switch. Switch to light fixture. Light fixture to far conduit. Far conduit and near conduit not touching because the metal box was replaced by plastic. Near conduit back to panel. Panel and neutral bus bar (terminal strip) bonded. In this circuit, if the switch is flipped to "on" you will measure considerable voltage across the two conduit ends and the light will not go on. In this circuit, if you bonded the conduit ends, the light will go on. I am not saying that what you have is exactly this but you need to investigate. The only proper way to get the current back to the panel is via the neutral wire.
You do not have to investigate every time you replace a box or work with conduit but the fact you found a possible problem means you need to find and fix it before continuing work.
Definition: Bond -- An intentional essentially resistance free electrical connection not counting a switch or a heavy load itself (such as a heater). If A is bonded to B and B is bonded to C then A is bonded to C.
Definition: Fault -- An unwanted not necessarily resistance free electrical connection.
I went ahead and replaced the box with a metal one this afternoon and it resolved my issue. I'm thinking of consulting an electrician to see if there is indeed a ground fault that needs to be taken care of.
I appreciate you all taking the time to help me through this.
I went ahead and replaced the box with a metal one this afternoon and it resolved my issue. I'm thinking of consulting an electrician to see if there is indeed a ground fault that needs to be taken care of.
No, this did NOT solve your issue! It made your light work, but it did not fix the serious problem. You have already verified thoroughly that you have an open neutral and a ground fault on the neutral side of the light fixture. The conduit is acting as the return path for the circuit.
This is DANGEROUS and must be fixed.
It's probably a very simple fix, too. Check the connections to the fixture. I'd bet the white fixture wire came out of the wire nut and is touching the box. If all the connections at the fixture look good then we'll progress to more diagnostics, but check that first.
Do NOT leave this as-is, it is a serious safety hazard.
Your conduit is providing the neutral path for the lights. The lights should have worked with the separated conduit since you have a neutral wire. I think you will find a neutral that has separated from a wire nut and is touching the conduit somewhere between the switch and the first light. Let us know what the electrician finds. Getting an electrician is a good move as you don't want to leave it the way it is.
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Related Threads
?
?
?
?
?
DIY Home Improvement Forum
3.1M posts
319.5K members
Since 2003
A forum community dedicated to Do it yourself-ers and home improvement enthusiasts. Come join the discussion about tools, projects, builds, styles, scales, reviews, accessories, classifieds, and more! Helping You to Do It Yourself!