18W 12V lamp issue

Status
Not open for further replies.

1793

Senior Member
Location
Louisville, Kentucky
Occupation
Inspector
I could use a little help. I went to a customer?s today to look into his LV Landscape Lighting. These lights were installed mid February of this year. There are 9 18W, 12V Malibu Tulip style walkway lights, 6 35W Malibu Well lights. There is a 600W transformer, 2 300Ws.
There are 3 of the 18W and 3 of the 35W on one half of the transformer, the rest are on the other half.

The issue:
For the last few months there have been no problems, about a week or so ago 8 of the 9 Tulip light bulbs burned out over a two day period. 3 of the 8 new bulbs went out in about three days.

The lights are on at night for about 9 hrs. The bulb package states that the bulbs should last ? year based on 3 hrs of burn each day. By my calculations, with the amount of time these are on each night they should last about 2 months, this sounds right for the original bulbs, what I can?t see is why the replacements are burning out so quickly.

Any ideas why this is starting to happen?

Here is a photo of a few of the old bulbs.
IMG_3993.jpg
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
I could use a little help. I went to a customer?s today to look into his LV Landscape Lighting. By my calculations, with the amount of time these are on each night they should last about 2 months, this sounds right for the original bulbs, what I can?t see is why the replacements are burning out so quickly.

Any ideas why this is starting to happen?

One quick thing to look out for is over voltage. (Possibly from L1-L2 voltage imbalance at the source?) The applied voltage is usually set up so that after subtracting the resistance losses in the transformer and all of the wiring, the bulbs are getting a voltage that runs them at an efficient temperature without over driving them.

But once a few bulbs have burned out, the current draw will be lower and the voltage at all of the other bulbs (but particularly the ones downstream of the failed bulbs) will see a higher than normal voltage and will burn out quicker themselves.
Did you replace them all or just the ones that were burned out? If you start out with half the bulbs being old, then the failure rate will definitely jump.

Are the new bulbs the exact same current and voltage rating as the old ones? If they look whiter (higher in color temperature) than the old bulbs, they may be running hotter for some specification-related reason.

The blackened inside is definitely a sign that the filament is evaporating and coating the inside of the bulbs. An air leak burning out the filament will typically leave a gray, ashen color inside the bulb instead.

Look into retrofit LED replacements while you are on the subject.
 

1793

Senior Member
Location
Louisville, Kentucky
Occupation
Inspector
The voltage at the fixtures is anywhere from 10V to 11V. Heat is my thought.

Our temperatures are stating to rise as Summer approaches. I had all new bulb, lamps, in today, this morning. The air temp. was about 70 deg., I had the lights on for about 20min. with direct sun beating down on the fixture. I use a FLUKE mini IR Thermometer and got a reading of ~ 275 deg. up inside the shade. If I took a reading on the bulb it was 365 deg. This seams very hot.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
The voltage at the fixtures is anywhere from 10V to 11V. Heat is my thought.

Our temperatures are stating to rise as Summer approaches. I had all new bulb, lamps, in today, this morning. The air temp. was about 70 deg., I had the lights on for about 20min. with direct sun beating down on the fixture. I use a FLUKE mini IR Thermometer and got a reading of ~ 275 deg. up inside the shade. If I took a reading on the bulb it was 365 deg. This seams very hot.

Bulbs too high powered for the fixture? Poor ventilation for some reason, not just high ambient temperature?

18W and 35W are enormous for a 12 volt system bulb. Are those the sizes that actually came with the fixtures or did somebody "upgrade" them?
Are the new bulbs halogen or straight inert gas? Halogen bulbs have to get hot to work properly. Not sure that means 365F on the outside though.
 

1793

Senior Member
Location
Louisville, Kentucky
Occupation
Inspector
Bulbs too high powered for the fixture? Poor ventilation for some reason, not just high ambient temperature?

18W and 35W are enormous for a 12 volt system bulb. Are those the sizes that actually came with the fixtures or did somebody "upgrade" them?
Are the new bulbs halogen or straight inert gas? Halogen bulbs have to get hot to work properly. Not sure that means 365F on the outside though.

Bulbs are incandescent

The 18W is what comes with the fixture

The 35w is what comes with the Well light

No problem with the Well lights.
 

iwire

Moderator
Staff member
Location
Massachusetts
The voltage at the fixtures is anywhere from 10V to 11V. Heat is my thought.

Our temperatures are stating to rise as Summer approaches. I had all new bulb, lamps, in today, this morning. The air temp. was about 70 deg., I had the lights on for about 20min. with direct sun beating down on the fixture. I use a FLUKE mini IR Thermometer and got a reading of ~ 275 deg. up inside the shade. If I took a reading on the bulb it was 365 deg. This seams very hot.

I really do not think that is hot.

Sure, hot to you and me but not hot for glass, metal and filaments.

The filament of an incandescent lamp approaches 3000 Celsius, it will not matter if the outside temp is 10 F or 100 F.

http://hypertextbook.com/facts/1999/AlexanderEng.shtml
 

iwire

Moderator
Staff member
Location
Massachusetts
Consider that standard incandescent lamps are used inside home ovens without an issue with oven temps during self cleaning mode well over 800 F.

I agree that over voltage would be the culprit or a batch of bad lamps. Maybe try getting the lamps from another source.
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
PE (Retired) - Power Systems
....or a batch of bad lamps. Maybe try getting the lamps from another source.

This happens a lot with lamps. Average life means that 50% of lamps burnout earlier and 50% later, so if 10% last twice as long then 10% can burn out immediately.

Definitely find a second source, not just more lamps from the same store.
 

1793

Senior Member
Location
Louisville, Kentucky
Occupation
Inspector
This happens a lot with lamps. ... can burn out immediately.

Definitely find a second source, not just more lamps from the same store.

I have purchased some new lamps from a local lighting wholesaler. I have my customer watching for the life of the replacements I put in the other day and as they go out I'll make sure he uses these new lamps.

Thanks to all who have shared.
 

1793

Senior Member
Location
Louisville, Kentucky
Occupation
Inspector
Update...

Update...

My customer just sent me a text to let me know one of the new lamps burned out last night. This was day 3. It had rained most all of the day and had quit by the time the lights came on. I have not had a chance to get back out there with the different lamps.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Excessive heat being introduced by a poor connection in the base/socket?

The total power dissipation (bulb plus connection) in that case should still be lower than the power dissipation in the bulb itself without a bad connection. The possibility of damaging the socket is real, but IMHO there is no way that the bulb filament could end up hotter. And the deposited filament metal on the inside of the bulb sure looks like a sign of overheating.
Take a look at some of the bulbs that have not actually failed yet and see if they are starting to blacken on the inside, or if it just happens when the filament breaks.
 

iwire

Moderator
Staff member
Location
Massachusetts
The possibility of damaging the socket is real, but IMHO there is no way that the bulb filament could end up hotter. And the deposited filament metal on the inside of the bulb sure looks like a sign of overheating.

I agree, the filament is hot hot hot hot, a little more heat is not going to change things. :cool:
 

K8MHZ

Senior Member
Location
Michigan. It's a beautiful peninsula, I've looked
Occupation
Electrician
I have worked on 12 volt vehicle systems for years.

For some reason, the 'wedge' style bulbs seem to always fail with the inside of the bulb blackened. A typical example is license plate bulbs, like this 194.

ge194_1.jpg


In fact, I don't think I have ever seen a 194 fail where I could still see the filament supports. I can almost always see them and the broken filament in a bayonet bulb like an 1157 or and 1156.
 

K8MHZ

Senior Member
Location
Michigan. It's a beautiful peninsula, I've looked
Occupation
Electrician
The total power dissipation (bulb plus connection) in that case should still be lower than the power dissipation in the bulb itself without a bad connection. The possibility of damaging the socket is real, but IMHO there is no way that the bulb filament could end up hotter. And the deposited filament metal on the inside of the bulb sure looks like a sign of overheating.
Take a look at some of the bulbs that have not actually failed yet and see if they are starting to blacken on the inside, or if it just happens when the filament breaks.

If the connection is poor, thermal cycling may introduce small cracks in the base of the bulb allowing oxygen in, causing the filament to oxidize which then coats the inside of the bulb with oxidized tungsten.

I think the absolute worst connection design is the wedge style using small wires as contacts. Contacts with very little surface area.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
If the connection is poor, thermal cycling may introduce small cracks in the base of the bulb allowing oxygen in, causing the filament to oxidize which then coats the inside of the bulb with oxidized tungsten.
Which is typically white. A large air inrush will typically leave a white coating on the inside of the bulb. A slow leak may cause tungsten oxide deposition which is then covered up with metallic tungsten when the filament finally fails. Not sure whether the white would be noticeable then, but I suspect it still would be from the outside.
 

kwired

Electron manager
Location
NE Nebraska
The total power dissipation (bulb plus connection) in that case should still be lower than the power dissipation in the bulb itself without a bad connection. The possibility of damaging the socket is real, but IMHO there is no way that the bulb filament could end up hotter. And the deposited filament metal on the inside of the bulb sure looks like a sign of overheating.
Take a look at some of the bulbs that have not actually failed yet and see if they are starting to blacken on the inside, or if it just happens when the filament breaks.

Can you tell me why standard 120 volt medium base lamps have shortened life when the lampholder has loose rivets connecting either the center pin or the shell to the supply conductors?

Over many years I have had people complain about a fixture that doesn't seem to have lamps last very long in it, then find socket has bad contact in some part like that, replace the socket and they then last longer.

I realize these lamps have the filament operate at a pretty high temp, but that additional heat from something like that must be creating a problem with something.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Can you tell me why standard 120 volt medium base lamps have shortened life when the lampholder has loose rivets connecting either the center pin or the shell to the supply conductors?

Over many years I have had people complain about a fixture that doesn't seem to have lamps last very long in it, then find socket has bad contact in some part like that, replace the socket and they then last longer.

I realize these lamps have the filament operate at a pretty high temp, but that additional heat from something like that must be creating a problem with something.

I suspect that the problem is more likely the intermittent nature of the bad connection. That can cause the same effect as rapidly switching the bulb on and off.
A high voltage bulb has a relatively thin filament compared to an LV bulb. That means that the filament cools rapidly as well as heating up rapidly, and so gets a repetitive surge from any intermittent connection.
Any bulb will have a longer life in hours when constantly on than when cycled many times.

I just can't see the additional heat at the base causing any problems by itself except melting the solder or otherwise causing an air leak.
 

1793

Senior Member
Location
Louisville, Kentucky
Occupation
Inspector
Can you over ride the sensor/timer that turns on the lights and test voltage at "start-up". Maybe the transformer is surging when it first comes on?

Yes I can do that, good idea. I spoke with my customer today and I asked him if the other lamps were turning black and he said there is a slight "blacking" near the base of the other lamps.

These lamps are of course base up.

Again, there are some 35W Well lights on the same transformer(s) and they are just fine, no problems.
 
Last edited:
Status
Not open for further replies.
Top