BajaNomad

ghost voltage?

Santiago - 8-15-2012 at 06:10 AM

Last week I finally got around to running a bunch of lights/plugs from my power center; I have a solar system providing 1500 watts of 110v. After completing the first twist up of everything, except the grounds, I noticed that one of the LED bulbs that was controlled by a switch had a faint glow with the switch off. When I threw the breaker, this glow went off. I moved the bulb to other locations with the same result.
I used my tester and found that with current flowing through the system I would get about 112 volts but when there was no load, i.e. nothing on or plugged in, I would get about .8 volts (less than one volt). None of my other LED bulbs would glow, just this one which is like a spot light and you can see the individual glow points in the bulbs.
I have not grounded the system yet and the ground wires are not even twisted together. Of course, I will do this next trip.
And before you go all techie on me, I have yet to find a 110v system in BOLA that IS grounded - I'm not saying there isn't one, I just have not found anyone who has bothered. Maybe in town with CFE, but off the grid homes, nada.
We pulled the cover of the load center (breaker box) and tested the voltage drop from the ground buss to the neutral buss and got zero; but the ground buss to the hot buss got 62 volts??? While both of us together don't add up to an electrician, it seemed to us this was wrong - why is there current flowing between the ground buss and the hot buss?
The ground buss in the load center is tied to an external ground buss that the inverter and charge controller are also tied to and that eventually this external ground buss will go to ground.
So two questions:
1. what is causing the .8 voltage in the circuit?
2. What is causing the 62 volt reading between the load center ground and hot buss?
Fire away......

woody with a view - 8-15-2012 at 08:11 AM

remember that electricity is invisible, and it hurts!

that's all i know about it. is there a crossed wire somewhere? good luck!

durrelllrobert - 8-15-2012 at 08:50 AM

Quote:
Originally posted by woody with a view
remember that electricity is invisible, and it hurts!

Never trust something you can't see :lol:

bkbend - 8-15-2012 at 11:28 AM

I moved into a BOLA casa with really scary wiring so it is now all redone and is a grounded 110V solar system. So there is at least one. In running all new wire it wasn't all that much more work. As for your issues, I'm not an electrician and am limited in knowledge but you might look at your inverter literature. Mine goes into an idle mode when there is no load but puts out a small amount of voltage so it can tell when a load comes on. It drove me nuts with one application that wasn't drawing enough load to trigger the inverter and I need to keep a parasitic load on that is not enough by itself to trigger the inverter but combined will do the trick.

Bob and Susan - 8-15-2012 at 11:47 AM

i have the same thing with my leds with some of my fixtures...i tried it ALL
ground...switch...wires...

i figure its the fixture and the led bulbs

remember the white wire does have current running thru it all the time
and it doesnt take much to light a led bulb

the draw...almost zero
i just let um glow
no power loss to speak of

and the differance between 4 watts and 13 watts is alot
the compact flouresant dont do that but take way more energy to run

[Edited on 8-15-2012 by Bob and Susan]

Cypress - 8-15-2012 at 11:49 AM

It's magic!

Bob and Susan - 8-15-2012 at 12:01 PM

i turned off the "no-load idle mode" when the solar people told me the on and off stuff may fry the boards inside...

i just leave it "on" all the time
the batteries charge everyday anyway

bkbend - 8-15-2012 at 06:04 PM

Thanks, I'll look and see if I can turn that 'feature' off. I don't think there are any obvious switches but the manual may show a dip switch internally.

Bob and Susan - 8-16-2012 at 04:22 AM

i still think the led "ghost voltage" is the common wire and the FIXTURE

i would try using another fixture to see if you still get "the glow"

some of my fixtures don't "glow" but wont fit in the location i have...
thats why i just live with the glow

RnR - 8-17-2012 at 10:01 PM

Quote:
Originally posted by Santiago
Last week I finally got around to running a bunch of lights/plugs from my power center; I have a solar system providing 1500 watts of 110v. After completing the first twist up of everything, except the grounds, I noticed that one of the LED bulbs that was controlled by a switch had a faint glow with the switch off. When I threw the breaker, this glow went off. I moved the bulb to other locations with the same result.
I used my tester and found that with current flowing through the system I would get about 112 volts but when there was no load, i.e. nothing on or plugged in, I would get about .8 volts (less than one volt). None of my other LED bulbs would glow, just this one which is like a spot light and you can see the individual glow points in the bulbs.
I have not grounded the system yet and the ground wires are not even twisted together. Of course, I will do this next trip.
And before you go all techie on me, I have yet to find a 110v system in BOLA that IS grounded - I'm not saying there isn't one, I just have not found anyone who has bothered. Maybe in town with CFE, but off the grid homes, nada.
We pulled the cover of the load center (breaker box) and tested the voltage drop from the ground buss to the neutral buss and got zero; but the ground buss to the hot buss got 62 volts??? While both of us together don't add up to an electrician, it seemed to us this was wrong - why is there current flowing between the ground buss and the hot buss?
The ground buss in the load center is tied to an external ground buss that the inverter and charge controller are also tied to and that eventually this external ground buss will go to ground.
So two questions:
1. what is causing the .8 voltage in the circuit?
2. What is causing the 62 volt reading between the load center ground and hot buss?
Fire away......


Firing.......

First off , a little background. A volt meter only measures the DIFFERENCE in voltage between two points, not the absolute voltage of a given point. The absolute voltage can only be measured against a point of zero voltage, most commonly a true "earth ground".

For example: If you had two wires, one at a potential of 1,000 volts measured to ground and the other at 980 volts measured to ground and you placed your meter between the two wires, you would only measure 20 volts. (The DIFFERENCE in voltage between the two wires) Ahh, a nice safe low voltage circuit ...

However, if your bare feet were standing on damp sand (true "earth" ground) and you touched one of those wires with your finger you'd be quite shocked at the result - 1,000 volts (or 980V) coursing thru your body, not 20!

Now, a little bit of electrical safety: You ALWAYS have to make two measurements before working on a circuit. Wire to wire (hot to neutral) and each wire to "earth" ground. Let the meter simulate exactly where you are going to place your body. You are placing your fingers wire-to-wire AND you are placing your fingers/feet wire-to-ground!

Put in that "earth" ground!!!!

Now for your questions:

Question #2 - I agree with you. There shouldn't be 62 volts between the ground buss and the hot buss. There should be the full 112 volts that you measured!

However, this is easy to explain. You said that your ground buss is not "grounded" yet and therefore is not guaranteed to be at zero volts. I'll bet that if you measured between the ground buss and a truly "grounded" point, that your ground buss would be at a potential of 50 volts. (112v - 62v = 50 volts)

Where did the 50 volts come from?? It's probably just an induced voltage or "ghost" voltage. AC circuits are notorious for inducing voltages in other nearby conductors, especially in conductors that are not grounded to bleed off the induced voltages.

Your "earth" ground at the inverter will also provide a "reference" point for the neutral side of the inverter circuits. You will always have 112V between your hot and neutral at the inverter but without the "earth" ground reference point your circuit voltages are just floating in reference to zero and can give some interesting and unexpected values.

This is the reason that CFE connects their transformer neutrals to "earth" ground at every transformer. Look carefully and you will see a ground wire coming down the pole from the neutral terminal or a jumper wire connected to the overhead ground wire on the poles.

Question #1 - Most likely an induced voltage. And, that particular type of LED bulb may have a VERY low "turn on" voltage.

Also, what type of switch is controlling the bulb? A make/break mechanical blade type of switch or a solid state device such as a dimmer switch or motion detector switch. Solid state devices ALWAYS leak a small voltage across the switch. It's inherent in their design. Haven't had much experience with LED's but the leakage is more than enough to flicker florescent bulbs. I've got two in my garage controlled by a motion/occupancy sensor flickering away right now!

Okay, my firing is running out of fuel .....

Buena suerte.

Bob and Susan - 8-18-2012 at 05:18 AM

the question is then

why do some of my led light fixtures have a "glow" and others dont


my switches are not dimmers...but regular toggles
all are the same

all wires run thru the same hot wire to the main panel and then a separate box in the casitas

all power comes form the same inverter

stray current

captkw - 8-18-2012 at 06:42 AM

a light/led that is glowing is like a mex test light !! you have stray current in your system and I would start at the BOX and work from there down ......K&T:cool:

Bob and Susan - 8-18-2012 at 06:46 AM

did that...every casita...right to the inverter

its only on one type fixture

ncampion - 8-18-2012 at 07:47 AM

RnR is spot on, he actually knows what he's talking about. Induced voltage can cause the LED to fire. LEDs are a problem for this reason as are CFL to some degree. Many timers, switches and dawn-to-dusk devices specificall say that they cannot be used with LED or CFL. You should probably have a profesional evaluate your entire system, it's not a good idea to have stray current running around, not to mention the fact that it may be wasting KW's that you are producing with a solar system.

Bob and Susan - 8-18-2012 at 07:58 AM

the hot wire goes to the box

thru breakers
to the switch
to the fixture

the led glows only on one type fixture not the others at all

isnt there always a small current on the common side anyway

i think the led lights are reading that small common voltage
its so small a reg bulb wont "fire"

RnR - 8-18-2012 at 10:48 AM

Quote:
Originally posted by Bob and Susan
the hot wire goes to the box

thru breakers
to the switch
to the fixture

the led glows only on one type fixture not the others at all

isnt there always a small current on the common side anyway

i think the led lights are reading that small common voltage
its so small a reg bulb wont "fire"



B&S - You're on the right track....

I did a little research on this LED issue and then tried to put 2 and 2 and 2 together and almost came up with 6...

So, here's my take on it:

The LED's that flicker are the MORE efficient ones. The one's that do not flicker are actually less efficient. Huh????

LED's have a "turn on" threshold that is measured in Micro-amperes not milliamperes of current. (One milliamp is a thousandth of an amp, one microamp is a millionth of an amp. That's why they are so incredibly efficient)

As I stated earlier, AC power circuits are notorious for inducing voltage/currents in other nearby conductors, hence, the many references to "ghost" voltages, "backfeeds", "hot neutrals", circulating currents, etc, etc, etc.

So, combine a super efficient LED with a circuit with induced currents and the LED is going to do it's job and light up. That's what it's supposed to do, isn't it?

Turns out that people don't like the flickering so the manufacturers have come up with a solution.

They have installed a high resistance (100,000+ ohm) shunt resistor internally across the terminals of the LED. The shunt resistor allows any small induced currents to "leak" through the resistor and bypass the LED proper without lighting the LED. When higher currents are allowed to the LED, (the switch is closed and the circuit is properly energized), a very small portion of the current continues to pass through the shunt resistor but the vast majority passes through the LED and it "turns on". So, depending on what value of shunt resistor the manufacturer has chosen to install, different LED's will "turn on" at lower or higher values of applied currents.

Obviously, the most efficient LED will pass all of the applied current through the LED and none through the shunt resistor. A less efficient LED/resistor combo will let more current pass through the resistor than through the LED, (and create heat rather than light), but it won't flicker!

So, the choice is yours. In actuality, ALL Led's are incredibly efficient compared to CFL's or incandescent's such that the flickering is only an aesthetic issue and not a practical issue.

How to stop induced currents?

1. Separate your hot and neutral conductors by at least a foot of space and don't let them run parallel to each other for very far. (That's why "twisted pair" communication wiring is effective at reducing noise, the conductors are not parallel to each other)

2. Use a double pole switch to break both the hot and the neutral to effectively isolate the device from any other conductors.

3. Place the switch very close to the device to minimize the lengths of the non-isolated conductors.

Buena suerte! :biggrin:

Bob and Susan - 8-18-2012 at 11:48 AM

santiago...a double pole switch is the correct answer

that would work if you really need to get rid of the glow

Santiago - 8-18-2012 at 03:36 PM

Good stuff; I actually like the 'glow' as it keeps the dorado slightly lit at night.

OOOH MYYYY!!! lol..GREMLEN'S

captkw - 8-18-2012 at 06:06 PM