Dimming of LED lights

[greencircle]

I have a 150W outdoor LED light fixture that doesn't turn on with 100% brightness. Appears it's only about 25% bright. Cant understand how it's possible. There ain't any dimming controls in them. Then what else could be causing this?

 

[MacGyverS2000]

How do you know it's only 25% of its designed brightness? Have you plugged the bulb into another socket to verify the brightness changes? Have you tried a different type of bulb (incandescent, fluorescent, etc.) between the two sockets to verify it's the socket and not the bulb?

Dan - Owner

http://www.Hi-TecDesigns.com

 

[greencircle]

No I am not sure it is 25%. I should have simply said its dimmed down a lot. My bad. At this point, unfortunately I am not able to switch power to different input. This is LED street light. All other fixtures on the street except two of them are full bright. I am trying to come up with possible reasoning before bringing down the fixture. Want to keep minimal downtime.

 

[OperaHouse]

I imagine that the LED lamps will suffer some of the same issues as the low cost CFL lamps. Is this 150W equivalent or actual wattage? Sounds like a controller problem though when some LED fail they go into low output mode.

I've just been playing with some cheap E27 9W LED lamps. A Philips for $2.50 easily runs at low power with a .05uF capacitor in series or 70,000 ohms. That is like a neon lamp! Looking with a scope it is a simple full wave bridge with a resistor in series and a smoothing cap. I have one in the basement operating at low power so the dog can see the steps at night.

Tried the same thing with a $2 Zilotek, an identical looking package. That unit would flash at about a half second. Obviously an electronic ballast that waits for a capacitor to charge up before the inverter starts.

Both lamps have a 9-10 year life expectancy indicating they are over driving the LED. The base of the lamps get quite hot.

 

[mcgyvr]

multiple LEDs/strings inside the bulb are burnt out..

 

[greencircle]

I believe that the LEDs are all in series. So it there is even 1 bad LED the whole string would not light up. In my case the string does light up (dimmed down). So I believe the LEDs are OK. I suspect that the input current to the LEDs might some how be reduced. This also mean that the Driver output is reduced.

I guess I will have to bring down a fixture to verify this.

 

[IRstuff]

"I believe that the LEDs are all in series. So it there is even 1 bad LED the whole string would not light up. "

Not generally true. Almost all LED stop lights have either parallel strings or bypassing of bad strings. Hard to imagine that street lights would be designed differently.

TTFN
I can do absolutely anything. I'm an expert!
homework forum: //

http://www.engineering.com/AskForum/aff/32.aspx

faq731-376 forum1529

 

[itsmoked]

Possibly the LED driver is a dimming type and the daylight sensor is bad.

Are these stepped brightness lights? If so the motion sensor could be bad so it never steps-up.

One of the series LEDs can fail and that restricts the current by presenting too high a resistance to ALL the other LEDs in the string(s).

There can be a bad solder joint that does the same thing.

There are usually current limit adjustment screws. It may be set wrong or the pot is bad.

Infant mortality of the driver or any one of the LEDs will do the same.

Trouble shooting: Wearing dark glasses measure the voltage drop across each LED. Make sure you do it including the wires that feed the LED so you include the connections. You do that by measuring from the terminals on the adjacent LEDs.



Keith Cress
kcress -

http://www.flaminsystems.com

 

[greencircle]

Yes it is a Dimmabale driver. But there is a remote rotolock photocell that triggers power to all the poles on the street. I don;t suspect the rotolock photocell becuase there are few poles lighting up right.

So, at this point I believe I may have to measure the resistance across LEDs.

Current limiting screws? where?

 

[itsmoked]

If they exist they're usually small holes in case of the driver.

Keith Cress
kcress -

http://www.flaminsystems.com