SHort life of G10 and Mains Downlights

[Hysteresis]

When you replace a lamp do you use a tissue or clean cloth to hold the lamp ?

If you touch the glass with your fingers then there can be a deposit left on the glass. When the glass is hot this deposit then acts as a seed to start a change in the glass leading to failure of the lamp. Even clean and dry fingers can leave enough residue on the glass to start the process.

The GU10 unit is a design which encloses the actual Hologen 'bulb' and as such the requirement to not touch the bulb is illiminated. The exposed halogen bulbs as in the G4 unit does require a tissue when handling but as far as I am aware this requirement is not relavant in the case of GU10's. I hope I have that right, it is certainly my understnding.

 

[Hysteresis]

I have 20 fittings around the house, and oer the last year I have changed 22 lamps.

No real problem then.
On average, each one is lasting about a year. Although some have lasted longer than this and some have been replaced twice, that is just how averages work.

I see where you are coming from flameport, but the published life expectancy of a GU10 is 8000 hours. I probably have mine switched on for an average of 10 hours a week, as in my living areas I use incandescent table lamps, and in the areas I have the GU10's they are usually off and only switched on when those areas are occupied. Now 10 hours per week averages out to about 500 hours on time per year, so even if I reduce the operating life to 10% of the published life expectancy, I should not experience a failure for Two years. That is even reducing the life expectancy by 90%. So satistically my failure rate is 40 times greater than expected even with the reduced life expectancy.
Averages taken over a small sample are notoriously inaccurate.
Taken another way one could calculate that 20 bulbs with 8000 hours life expectancy should give me 160,000 bulb hours, My average failure rate is 250 hours for the 20 bulbs. That is a total of only 5000 bulb hours. This calculation still gives me a failure rate of 30 times the expected published figure.
So Yes there is a problem.
If you want to take your averages example, then you must look at the figures. I am failing 20 bulbs per year out of a total of say 20 bulbs installed. That is I am getting a maximum on average of one year per bulb. Each bulb is on for approximatly 500 hours per year. So on average I am getting 500 hours burn time per bulb. This is still 16 times less than the published figure - on average.
Do you still think I have no problem?
Incandecent bulbs have a better life cycle than this and GU10's are marketed as having a greater life expectancy that incandescents.

 

[riveralt]

I see where you are coming from flameport, but the published life expectancy of a GU10 is 8000 hours. I probably have mine switched on for an average of 10 hours a week, as in my living areas I use incandescent table lamps, and in the areas I have the GU10's they are usually off and only switched on when those areas are occupied. Now 10 hours per week averages out to about 500 hours on time per year, so even if I reduce the operating life to 10% of the published life expectancy, I should not experience a failure for Two years. That is even reducing the life expectancy by 90%. So satistically my failure rate is 40 times greater than expected even with the reduced life expectancy.

I would write to the manufacturers of the bulbs to complain.

Averages taken over a small sample are notoriously inaccurate.
Taken another way one could calculate that 20 bulbs with 8000 hours life expectancy should give me 160,000 bulb hours, My average failure rate is 250 hours for the 20 bulbs. That is a total of only 5000 bulb hours. This calculation still gives me a failure rate of 30 times the expected published figure.
So Yes there is a problem.

If you want to take your averages example, then you must look at the figures. I am failing 20 bulbs per year out of a total of say 20 bulbs installed. That is I am getting a maximum on average of one year per bulb. Each bulb is on for approximatly 500 hours per year. So on average I am getting 500 hours burn time per bulb. This is still 16 times less than the published figure - on average.
Do you still think I have no problem?

Do you ever think that you might be the problem.
You have been given advice on this subject for 13 months now and have done absolutely nothing about it. As your previous, long and monotonous posts, have shown you have a digital multimeter - you have stated you know how to test - so stop your britching about the amount of bulbs you go through and get on with it.

 

[Hysteresis]

Oh Riveralt -
You miss the point. I am looking for advice and help not abuse.
Like so many posters on this web site, you seem to want to be abusive and not helpful. It is why so many good Contributers have just abandoned the site.
If you can't be helpful then just don't post.

 

[stillp]

I'm surprised nobody's asked you if you're dimming these lights, which will stop them reaching full temperature and prevent the halogen cycle taking place. This will drastically shorten the bulbs' life.
If you have dimmers, replace them with switches for a while, and if the lights are too bright fit lower wattage bulbs.

 

[ban-all-sheds]

the published life expectancy of a GU10 is 8000 hours.

This still doesn't explain your failure rate, but does make it less extreme:

When a manufacturer says that his lamps have a life expectancy of X hours he means that by that time 50% of them will have failed.

 

[riveralt]

Oh Riveralt -
You miss the point. I am looking for advice and help not abuse.
Like so many posters on this web site, you seem to want to be abusive and not helpful. It is why so many good Contributers have just abandoned the site.
If you can't be helpful then just don't post.

But that is the point you have been given advice and yet to continue to ponder on this possibility and that possibility.
You have a simple choice either get an electrician to check your lighting circuit or check it yourself. You have a multimeter don't you?
No one here is going to be able to identify the specific fault you have (if you have one) without getting a screwdriver to look at the terminal connections and/or a meter to check the cables.
This post has been going on for 14 months now, it clearly doesn't worry you that you are regularly replacing bulbs - otherwise you would have got it fixed by now.
I can only conclude that you are a sad and lonely Troll for which you have my sympathy.