Light Bulbs Dimming

Page 1 of 2  
This just started happening. In one room I have a ceiling light/fan which has been there for 15 years and working fine. Fan works fine. Just recently, when I turn on the light, it starts out bright, then dims after several minutes. I have only CF's in the house, but have switched the bulbs and all act the same. Will try to find an incand type today and check it out.
If I turn the power off for 5 seconds and back on, it's back to full brightness again. Have tried this from both 3-way switches.
Most of the ceiling lights are on the same circuit and I only see this on the one fixture.
Any ideas how to start troubleshooting this?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 11/17/2015 08:18 AM, Snuffy "Hub Cap" McKinney wrote:

If you stand next to the light, do you hear a crackling sound? If so the socket is bad.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 11/17/2015 8:26 AM, philo wrote: ...

Good chance't with/without any sound effects; the tab may be weak and when gets warm loses even more contact and has a film built up.
Try a pencil eraser to "bright it up" a bit and give it a little pull back into position and the symptoms should disappear. If so, still probably not a bad idea to replace the socket on general principles.
Or, could be a poor connection of the hot/neutral to the socket and the socket itself is fine...
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 11/17/2015 9:02 AM, dpb wrote:

Or, it could be a faulty/failing switch...
--



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 11/17/2015 8:02 AM, dpb wrote:

Oooo... I so dislike suggesting things like that! Always afraid the person will forget to turn off the "switch"; the switch will be in the neutral conductor instead of hot; etc.
So, I always suggest: - turn light "on" (wall switch, switch on fixture, etc.) - verify light is illuminated! - turn OFF breaker for that branch circuit - verify light is extinguished!! (hope you didn't flip the WRONG breaker and the bulb burned out "by coincidence" at the same time :> ) - tag circuit with "man on line" (if there is any possibility that someone will come along unbeknownst to you and flip the breaker back ON while you're working on it)
A lot of times, these tabs are a springy material that just fatigues over time. So, even lifting it back up isn't a guaranteed (long term) solution.
Or, the contact has a portion that "turns downward" and acts as a spring, of sorts, to buoy the contact point upwards. Once this gets deformed (cuz someone screwed in a bulb over-zealously), it is almost impossible to restore its utility.

Look closely at the base of the lamp. The center contact should show no signs of scortching. It's usually a blob of solder so it is fairly soft. Make sure it has a uniform "shape" -- no flat spots that might be giving poor contact.
[I had a batch of bulbs that *all* had the same "defect" -- if you want to call it that -- in their manufacture that made them flakey in a particular light socket (garage door opener)]
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Don Y wrote:

If you've got a soldering iron or gun you can try "extending" the center contact on the bulb by adding some solder, turning the bulb base down and remelting the solder with the iron so it "hangs down" a little before it cools.
Jeff
--
Jeffry Wisnia
(W1BSV + Brass Rat '57 EE)
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 11/18/2015 1:12 PM, Jeff Wisnia wrote:

In my case, a *wood* file (it's just lead so very soft) made short order of taking off the high spot (so the "low spot" was more nominal).
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 11/17/2015 9:18 AM, Snuffy "Hub Cap" McKinney wrote:

I recently replaced all the ceramic sockets in our kitchen ceiling fan for flickering lights.
I installed the fan in 1988
Over time the rivets that hold the wire leads in place in the lamp sockets loosened up, causing overheating and damaging the wire lead insulation.
The insulation was brown, dried out, a cracked and broke away from the wire when bent.
Could have been a real fire hazard.
Steve
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 11/17/2015 9:18 AM, Snuffy "Hub Cap" McKinney wrote:

Let me pose a totally different problem/solution, which I personally had. I put CFLs in my 2 ceiling fans and they seemed to work ok. But, after a time an after being on for some about 1/2 hour, I started noticing some blinking/dimming/etc. I traced it to a little module, in the fan, that dims or shuts off the lamps due to a too high wattage situation. This is to protect the fixture and sockets from someone putting in too high wattage (incandescent) bulbs. I went back to regular incandescents and there was no problem. Since one of the fans is really high up, it would be nice to have more light, so I put in CFLs and removed the high wattage detector. All is fine except for 2 problems. Sometimes, when you turn on the fan or change speeds, it causes a slight vibration and the CLF bulbs make a tick-tick noise when they bump the glass shade. The 2nd is that these CFLs take, seeming forever, to get to full brightness. Now that you can get mini candelabra base LED bulbs in 60 watt equiv. fairly cheap, that might solve all the problems ... except it takes a big extension ladder to get up there and we really don't use them that much. Someday.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

which has been there for 15 years and working fine. Fan works fine. Just recently, when I turn on the light, it starts out bright, then dims after several minutes. I have only CF's in the house, but have switched the bulbs and all act the same. Will try to find an incand type today and check it out.

brightness again. Have tried this from both 3-way switches.

this on the one fixture.

regular

really

slight

to

in

This brings up a question I have wondered for some time. Has anyone ever measured the actual current flowing to a CF to see if it really does save power/money?
Not related to this, but I notice when changing a CF after a few months or years, that the base is blackened. Also notice the base gets really hot. The glass is hot too, but not as hot as incand bulbs. But having the base getting so hot and being close to the ceiling seems like it could be a hazard in some places where there is little clearance from the ceiling, for example.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 11/18/2015 10:03 AM, Snuffy "Hub Cap" McKinney wrote:

If "you" try it, chances are you won't have the proper tools to do so. (Hint: a DVM is NOT "the right tool" -- even with an "amps" scale)
The stated wattage on a package of CFL's is the "actual" power that the bulb(s) consume. E.g., "13W for 60 equivalent watt bulb". This is the "real" power CONSUMED by the device.
[Power represents "work done" -- for a loose definition of "work" :> ]
The reason that "you" won't be able to measure this with your DVM is that CFL's are highly "reactive" loads; said another way, their "power factor" is not "1" as it would be for a non-reactive (i.e., resistive) load.
Incandescent lamps are practically perfect resistive loads. Their power factor is "damn near 1.0" (power factor can vary from -1.0 to +1.0).
What this means is that the current *through* the device is exactly in phase with the voltage *across* the device. Current is defined as the voltage across a device divided by the resistance of the device. If the resistance is constant (not varying), then, in theory, current is always directly proportional to voltage: double the voltage, double the current; halve the voltage, halve the current.
Power is voltage times current. So, if you are drawing 10 amps at a potential of 120 volts, you are consuming 1200 watts (10 * 120).
If the "resistance" of your load isn't constant, it is considered "reactive" (it changes!).
For a CFL, there is an electronic "ballast" in the base of the lamp/bulb. This is a little power supply that converts the 120VAC to the much higher voltage used to excite the fluorescent bulb. It does so in much the same way that the power supply in your PC, TV, etc. operates.
Mains power is AC (alternating current). It's not a "constant" 120 volts but, rather, a "wave" that varies from (roughly) -170V to +170V and back again -- 60 times each second. So, an ideal *resistive* load would see the current flowing through the wires go from some negative value (that is proportional to the resistance presented by the device) to some *identical* positive value and then back again -- always in lock-step with the variations in the mains voltage.
This is what happens in an incandescent lamp.
In a CFL ballast ("power supply"), the mains voltage behaves exactly the same way (of course). But, the ballast essentially draws *no* current from the mains UNTIL the voltage is pretty close to the peak. THEN, it draws a LOT of current and continues to do so until the mains voltage drops below that "high" level. Thereafter, it draws *no* power until the mains voltage approaches the same "high" *negative* voltage -- at which time, it draws another big slug of current.
So, while the mains voltage is below that "high" voltage, the bulb looks like it's not even connected to the circuit! It's drawing *no* power! But, for that small period of time near the (positive and negative) voltage peaks, it draws a *lot* of power -- to make up for all the time that it wasn't drawing ANY power!
Between these peaks, the ballast operates on the "extra" power that it has stored in itself (capacitors). So, the power that is actually used (consumed) in the ballast looks like 13W. But, the *apparent* power seems much higher -- when you look at this big "slug" that it is delivered as (icky sentence structure).
For a CFL, the "power factor" -- the ratio of "real" power to "apparent" power -- is about 0.6 or 0.7. So, a 13W CFL *appears* to consume 20W. You only pay for the 13W that are actually consumed. But, the wires in your house experience "surges" comparable to what a 20W load would present.
[And, the utility company likewise experiences those surges! If your entire house was powered like this sort of power supply, the utility would be annoyed because they would have to size their system to deliver huge "slugs" of power to you -- instead of a smooth and continuous flow!]
For Art's "wattage protector" problem, the design of it probably assumes a power factor of 1.0 -- for a normal, incandescent lamp. Instead, it is seeing these big surges and thinking the lamp is actually bigger than it really is (because it isn't noticing all of the time that the lamp isn't drawing ANY power between surges).
[Sorry, this is not the simple explanation I had hoped but the best I could do in the morning and without pictures! :> Google "power factor" for more information ]

Some CFL's don't like to be mounted "base up". This traps all of the heat in the ballast (power supply) and can lead to more rapid failure (electronics don't like heat).

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

I'm sure glad that LED bulbs are available and affordable now. CFL bulbs have always been a pain in the ass (in many ways)......
Ten years from now, they probably wont even sell CFL bulbs anymore. They had a very short life span !
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload


light/fan

fine.

then dims

switched

today

full

this

personally had.

after a

some

that

This is

and

would be

wattage

turn on

bulbs

that

that you

cheap,

extension

Someday.

ever

save

]

phase

voltage

resistance

halve

lamp/bulb.

higher

same

volts

back

see

(that

*identical*

variations

the

it

"apparent"

months or

hot.

the base

a

ceiling, for

I don't have one now, but the best way to measure actual AC current in a circuit is with a clamp on power meter. I don't know if this particular one's any good, but the basic description is good for anyone not familiar with this type of measurement. http://www.harborfreight.com/digital-clamp-on-multimeter-95652.html
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 11/18/2015 9:46 PM, Snuffy "Hub Cap" McKinney wrote:

It won't give you an accurate reading -- that was the reason behind my lengthy power-factor description. The clamp on ammeter doesn't expect the load to be reactive so doesn't know how to get to "real power" (real current).
You need an "RMS reading" meter (of amperage, voltage, etc.) not a "typical DVM/ammeter/clamp-on ammeter.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thursday, November 19, 2015 at 12:27:33 AM UTC-5, Don Y wrote:

Get a Kill-a-Watt meter. It will give you the watts and the power factor. My CFLs have a .6 power factor which surprised me.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

is

(i.e.,

"apparent"

20W.

wires

would

current in a circuit is with a clamp on power meter. I don't know if this particular one's any good, but the basic description is good for anyone not familiar with this type of measurement.

http://www.harborfreight.com/digital-clamp-on-multimeter-95652.html

my

expect the

factor. My CFLs have a .6 power factor which surprised me.
How much power does one of your CFLs draw compared with the number on the box?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Wed, 18 Nov 2015 22:28:01 -0700, Don Y

Even an RMS reading meter will not give you a power factor corrected "real power" reading. You need to use either a power factor meter or the "three voltmeter" test as shown in http://www.giangrandi.ch/electronics/cosphi/cosphi.shtml
A wattmeter will give you the actual wattage as well - unlike using a voltmeter and ammeter which only gives you volt-amps.
A simple Kill-A-Watt or Watts Up AC powermeter is very accurate.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Wed, 18 Nov 2015 09:03:44 -0800, "Snuffy \"Hub Cap\" McKinney"

After a few CFLs let out the magic smoke I was even less entralled with them than when I first started using them and had them fail in under 6 months.
High quality LEDs are standing up better. No-Name Chinese mini-spots -MR16 and the like - are proving to be a royal pain in the (three feet below the) neck.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

light/fan which has been there for 15 years and working fine. Fan works fine. Just recently, when I turn on the light, it starts out bright, then dims after several minutes. I have only CF's in the house, but have switched the bulbs and all act the same. Will try to find an incand type today and check it out.

full brightness again. Have tried this from both 3-way switches.

this on the one fixture.

personally

But,

in

regular

really

problems.

slight

the

get to

bulbs in

really

ever measured the actual current flowing to a CF to see if it really does save power/money?

months or years, that the base is blackened. Also notice the base gets really hot. The glass is hot too, but not as hot as incand bulbs. But having the base getting so hot and being close to the ceiling seems like it could be a hazard in some places where there is little clearance from the ceiling, for example.

What is that smoke, anyway? I noticed that with a few of them when they are first installed. And when I remove one, it still smells only not so strong. If it's got mercury in it, I'm going be pissed at Gov Yoga Pants and his go-green political machine.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Wed, 18 Nov 2015 20:49:33 -0800, "Snuffy \"Hub Cap\" McKinney"

If the bulbs still worked, it wasn't the "magic" smoke that came out. Magic smoke is what makes electronics work. When the smoke comes out, nothing works any more.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Related Threads

    HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.