Why does my LED lamp have AC voltage from case to ground and is it dangerous?

Here's the deal:  3-section under-counter LED light strip in an aluminum housing, with a 2 wire AC cable and plug for 220v.
I found however, that when I set a voltmeter to AC and measure between the case and a ground (water faucet, natural gas line or the ground from an electrical socket) I get 74 volts with one unit and 89 with another.

The seller first pointed out that a grounded cable should be used. I replied with pictures showing the unit is built to accept only two wire and comes with a two wire cord.

Then they said using a sink as ground didn't truly represent ground and that the measurement was faulty and that the current is not dangerous regardless.  I took additional measurements using an electrical ground and the metal pipe from the natural gas line. Same result.

However, I'm having trouble measuring current. I put the probe into the ten amp side, set the meter to the mA/A position and got nothing, although I did hear a buzzing sound emanating from the lamp housing--that can't be good.

I then put the probe into the mA socket and tried again.  Same buzzing, no reading. 

My questions: 1. What's wrong with my setup that I see no current measurement.  2. Is this a dangerous lamp? I took the one near the sink down. The other one is not within reach of an accidental grounding, so it is up until I'm ready to return both units.


sort by: active | newest | oldest

In general a metal housing requires a ground connection or some protection against contact with any internals.

But I know from old hifi equippment that they often had a "buzz" on the case - you tough it gently and it feels like it is vibrating.

They showed similar voltages when tested against ground but basically no current no measure.

In most cases all of this stoppen when reversing the input power cables - give it a try.

Most cheap LED lamps only use a resistor and a capacitor to lower the voltage and of coure the rectifier.

The better ones use a constant voltage driver - they can produce an audible sound when working, usually a very high pitched sound almost like a mozzie.

bobcov (author)  Downunder35m2 years ago
Hi, I felt the buzz, and that's why I got out the voltmeter. Tried reversing the plug in the outlet, no change.

I would assume they use an inverter to drive the LED's and that this inverter causes HF interference which is caught by the alu housing.

Nothing serious as you already checked the voltage.

If you check the frequency you might see something in the 20, 40 or 60kHz range - the normal frequencies of standard inverters.

In case the buzz when touching is of any concern you add an earth connection to the housing, either with a seperate wire or by replacing the power cord with something that has the added wire.

Only an option if you can open the lamps and see if they provide and usable connection as otherwise you have to make a small screw hole.

Toga_Dan2 years ago

Hey, Bob,

Are your outlets not grounded?

Where are you located?

Can you send a pic, or make + model #? a link to this unit online? Is it made for 220?

It does sound scary. But it sounds like you have no Amp potential. I guess that you are using a digital meter, yes? An analog meter might require enough current to damp down the volts reading. I wonder if the buzzing would also happen if you did ground the case solidly.

I have seen a ground wire register voltage relative to a fixture, and perhaps even relative to pipes. I don't recall how that was resolved.

bobcov (author)  Toga_Dan2 years ago
I can only assume the outlets are grounded. I get the same reading whether I use a water pipe, the ground contact in the outlet or the piping for the natural gas. Since the reading is the same, I'm guessing they're going to a ground source, and not just tied to each other.

The unit is made for 220 v. Using a digital meter, yes. When I grounded solidly, the buzzing stopped, but there was more of a crackle whenever I moved the contact on the casing. I really think if I left it grounded, some component of some sort would fail and then I'd get some real current, but that's just speculation on my part.

Between electrical ground and natural gas pipe ground, there's 162 mv, or basically nothing.

Here's a picture of it.

I just wonder why it is designed so that some part of the circuit is attached to the case. I'm thinking it has something to do with using the case as a heat sink for the LEDs. That's probably what they're doing.