AC voltage experts, check this out

[ChangLi]

I'm replacing an outdoor security light I hooked it up to a 120v circuit and it lasted about 40 seconds and quit. In trouble shooting the problem, I tested voltage and got 160volts....HTF can you get 160v ?
I know it's bleeding over from somewhere, but 160? Any ideas?

 

[thomsbrain]

better test it again with a fork, just to be sure.

 

[hamburglar]

160V continuous? Or just peaks?

 

[BooneRebel]

Originally posted by: thomsbrain
better test it again with a fork, just to be sure.

LOL . . . Really! That's one of the few times in recent memory I've laughed out loud at something here...

 

[blahblah99]

that sounds like peak to me.... I wonder if he set the meter to read correctly.

 

[Heisenberg]

It probably is peak. 120 * (2/2^.05) is about 170.

 

[ChangLi]

I actually checked it thrice and one of the three for almost a minute...it varied from 144.9 to 162.5 constantly. Please explain the "peak" formula I really would like to learn.

 

[Heisenberg]

Originally posted by: ChangLi
I actually checked it thrice and one of the three for almost a minute...it varied from 144.9 to 162.5 constantly. Please explain the "peak" formula I really would like to learn.

Huh. Assuming your meter is okay you might want to have an electrician look at it.

As for the peak formula, here goes. AC power is in the form of a sine wave. The voltage your meter reads when set to AC is called root-mean-square (RMS) voltage. This is not the same as the actual voltage at the peak of the sine wave. RMS is a special sort of average voltage (without going into too much detail, it's found by integrating the voltage over one complete cycle in a certain way). Anyway, for a sine wave the relationship is peak voltage = (2/2^0.5) * RMS voltage and RMS voltage = (1/2^0.5) * peak voltage. When you measure DC, you simply measure the voltage absolutely, because it is not changing with time. If you try to measure AC with the meter on DC, you'll get a value closer to the peak value, instead of the RMS one.

 

[etech]

The peak of the sinewave of AC voltage is around 170 volts. When you hear 120 volts that is the RMS(root mean squared) voltage.

Check your meter(manual)and see if it reads RMS. It should and every meter I have ever seen reads RMS AC voltage, but I've been surprised before.

Check another outlet(s) in your house and see what it reads.

Read betwen the ground and the neutral wire and see if it reads 0 volts. I have seen strange readings when neutrals weren't hooked up but that shouldn't blow out the light.

Be very careful, something is wrong and even more than normal caution is advised.

 

[ChangLi]

Heisenburg.

Thank you! I will spend some time with the formulas and becom their friends...I really do appreciate. The readings were from a new/almost digital multimeter. To fill the hole, I put the light back up and about an hour later, it was working..... How long can I expect a "commercial" 120watt sodium fixture to last at 160 +,- volts? Any ideas on how 160v? No probs with the rest of the building.

 

[ChangLi]

My bad...thank you etech for the formulas.

 

[Heisenberg]

Originally posted by: ChangLi
Heisenburg.

Thank you! I will spend some time with the formulas and becom their friends...I really do appreciate. The readings were from a new/almost digital multimeter. To fill the hole, I put the light back up and about an hour later, it was working..... How long can I expect a "commercial" 120watt sodium fixture to last at 160 +,- volts? Any ideas on how 160v? No probs with the rest of the building.

I really don't know how long it will last. I'm sure the lifetime of the bulb will be reduced quite a bit though. If that circuit really is at 160v then something is wrong somewhere, and I would have an electrician look at it before doing anything else.

 

[etech]

Sounds like you may have gotten a fingerprint on the first bulb. Any oil on the bulb can cause them to burn out rather quickly.

A sodium light uses a special ballast to give it a higher voltage. The 160 is probably normal though I don't know what voltages they use.

Next time, tell us it's not a normal incandescent.

 

[ChangLi]

Double my bad....I so do SUCK at this internet thing......mad props to H for the formulas.....AAAAAAAAAAARRRRRRRRGGGGGHHHHHHHHHHHHH it's been a long MF day. Thankx to all for your input . JM

 

[ChangLi]

sorry. I guess I knew it was Sodium, I kinda expected you guys to know too. I'm still trying to figger out out I'm gettin 160v to that light only. Every other circuit is within bounds. Thankx guys for your input.

 

[etech]

Where on that light circuit are you measuring the voltage?

 

[ChangLi]

Man I really do suck at this...the voltage was measured from the "incoming" hot wire to ground.

 

[Heisenberg]

Were you measuring the voltage to the light bulb or to the fixture itself? The voltage to the fixture should be 120, but etech was right about sodium lights using a different voltage for the bulb. In general, anything other than a normal incandescent light will have a voltage other than 120 to the bulb.

 

[ChangLi]

I measured the voltage from the "incoming" hot wire to ground. I again ask how can you get 160v. If it were a 240v line, .70% +,- would be 160v, but I am certain this is 120v....no doubts.

 

[etech]

Depending on the fixture and where the ballast is located you may still be measuring after the ballast. I'd have to see the fixture to know for sure.

I'll agree with Heisenberg this time, if the voltage to the fixture is really reading 160vAC I'd call an electrician. I can't come up with a reason it would be that high but it isn't normal and could be indicative of a serious problem. If you are measuring after the ballast caution is still advised. Some lights use very high starting voltages.

edit, I forgot to add this article, it's more on adding sodium lights to fish tanks but has a good diagram of the ballast.

diagram

 

[ChangLi]

Etech, Heisenberg,

You two are the best, thank you for your input and your assistance. This has been my very best ever anandtech experience.

I' guess I'll call an electrician.

One last stupid ?
The light was dead...I replaced the bulb, I replaced the photocell, the light was dead. I checked voltage with the power wires still conected, could the ballast effect my readings? Remember please that the light came on for about 30-45 secs and then went out?

 

[cressida]

Root mean square ... like everyone else says.
multiplay 120 by square root of 2, you should get the peak value of the sinusoidial wave.

 

[Heisenberg]

Normally, I would so that no, you should get 120v regardless of whether or not the wires were connected. However, it is possible something strange is happening inside the ballast and messing up the voltage you measure. It's not real likely, but possible. The only way to know for sure it to disconnect the light entirely, and then measure the voltage. If you are SURE you can do this SAFELY, it would really narrow down the problem. If you have any doubts, though, let an electrician take care of it.

 

[etech]

I can't think of how the ballast would affect the voltage on the input but it may be possible. If it is I would say that there is a wiring problem.

Disconnect the light from the circuit and measure just the input wires. Measure from hot to neutral(120vAC), Hot to ground(120vAC) and from ground to neutral( 0 vAC). If any of those don't match with the light disconnected, call an electrican.

With the light hooked up they should read the same.

Since I can't see the fixture I have to ask, are you sure you measured before the ballast and not after?