LED Emergency Light Fixtures Blowing

Status
Not open for further replies.

nathanslc

Member
Location
Utah
We are installing new 277V LED emergency light fixtures that have a battery backup. The light fixtures are controlled by a switch with the normal lighting circuit (also LED same brand) and an un-switched circuit for the battery. The emergency light fixtures are failing after approximately 2 months. They completely burn as if it were a phase to phase short in the light fixture when switched. The breaker(s) do(es) not trip. This is happening randomly across several circuits and different phases. We have tested using battery power on the same phase as the switched power but they still fail in the same manner. We have tested amperage as well as voltage as well as breakers and there are no issues. The manufacturer keeps sending replacements only to have them fail again after approx 2 months. The existing fluorescent emergency light fixtures have functioned fine for years according to the building owner. Any ideas?
 

greenspark1

Senior Member
Location
New England
Ground could be bad/floating?
If it's just the emergency fixtures it sounds like a manufacturing problem. Poor voltage regulation or something. Without attaching a voltage logger it's hard to say. Have you used these fixtures before?
 

nathanslc

Member
Location
Utah
Can you post a cut sheet or at least a link to the product you installed?

I have never used these fixtures before. Although there is a ground, the operation of the fixture itself is not effected by it. I have disconnected the ground in case there was a transient voltage and this also does not work. The neutrals have been checked and appear to be solid (no open neutrals).
 

mgookin

Senior Member
Location
Fort Myers, FL
How many emergency light fixtures did you install and how many of the emergency light fixtures have failed?

Have you contacted an engineer at Cree?
 

nathanslc

Member
Location
Utah
How many emergency light fixtures did you install and how many of the emergency light fixtures have failed?

Have you contacted an engineer at Cree?

We have installed approx 30 and a total of 13 have failed so far. When it first started happening we thought we had 1 or 2 bad fixtures but then it just kept happening. We have disconnected them at this point to prevent any further damage. We are currently waiting for a reply from the manufacturer but that may take time so I thought I would try this forum as a resource. I am hoping someone has had a similar situation and found a resolution.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
140221-2015 EST

Take a look at the maximum allowed ambient temperature. Totally unrealistic. Do you know how to convert C to F?

In what manner do these fail?

You need to do some bench testing where you over stress the fixtures in various ways, and simultaneously make measurements.

.
 

ELA

Senior Member
Occupation
Electrical Test Engineer
What is your input voltage measuring?

The specs say 120-277V. Therefore the maximum is 277V, in lieu of a specification indicating a +/- percentage.

Could be that they include an MOV that is marginally specified and your 277V supply, possibly being a little on the high side takes the MOVs out?
May not be intended for a 277V circuit + 10%.
 

nathanslc

Member
Location
Utah
What is your input voltage measuring?

The specs say 120-277V. Therefore the maximum is 277V, in lieu of a specification indicating a +/- percentage.

Could be that they include an MOV that is marginally specified and your 277V supply, possibly being a little on the high side takes the MOVs out?
May not be intended for a 277V circuit + 10%.

The input voltage measures a consistent 282 across several circuits. I cant imagine that 5 volts would cause such a problem. I have been told that this type of technology (emergency battery on LED 2X4) is less than a year old and that this is one of very few projects that actually have it. Has any one else installed emergency battery backed up 2X4 LED fixtures?
 

ELA

Senior Member
Occupation
Electrical Test Engineer
The input voltage measures a consistent 282 across several circuits. I cant imagine that 5 volts would cause such a problem.


They completely burn as if it were a phase to phase short in the light fixture when switched.
Was that imagined or real?

How did you determine that? Take one of the failed units apart and figure it out.
 

nathanslc

Member
Location
Utah
Was that imagined or real?

How did you determine that? Take one of the failed units apart and figure it out.

It is real. They definitely burn when switched. We initially thought it was a phase to phase short between the emergency and normal circuit however; some of the fixtures that burned were on the same phase (normal and emergency). We have taken the fixtures apart. We cant "figure it out" or I wouldn't be asking on this forum. The wires completely burn off of the board.
 

Attachments

  • 20140117_141247.jpg
    20140117_141247.jpg
    143.4 KB · Views: 0

ELA

Senior Member
Occupation
Electrical Test Engineer
nathanslc,
Thanks for the picture.
I was implying that the design might be marginal.
I understand that it may be difficult to figure out without additional input.

It is a bit difficult to see in the picture exactly what is burnt? What wires burned off the board ( a missing components wires)? I see two pads at the top. Is there normally a component located there?

Can you clean the soot from components that were not actually burnt and attempt to find the component/s that are burnt or missing? Is there a silk screen text marking on the board where a missing component was?
A typical failure could be a MOV that is designed to prevent transient events, however they can explode when exposed to a constant over-voltage.

If you have access to a working unit can you identify which component/s t have failed by comparison with the failed unit? If so can you read any markings on the component in question from the good unit?

I understand that this type of investigation should probably not be a part of your job, just trying to help since you did ask.


You are seeing 282V at one moment in time? Can you log the voltage over time?
 
Last edited:
Status
Not open for further replies.
Top