T5 fluorescent tube issues

[067256]

Hi there,
have posted this before but have been called back again and now I am really stuck for answers. Over a period of 6/7 months this office area has been changing lots of t5 tubes in there modular fittings due to ends being black and not working. I have been to change tubes. I installed crompton types 3 months ago but got called back again this week and these same tubes are no good now. Black ends!!
the fittings have warm start electronic ballasts and the fittings are controlled via various ceiling pir presence and microwave detectors. Timers are set to max.Also including reactive dimmer switches aswell.
does anybody have any ideas on why the lifespan of the tunes is so short?
many thanks.

 

[Marvo]

the fittings have warm start electronic ballasts

What make and model number are the fittings?

and the fittings are controlled via various ceiling pir presence and microwave detectors.

How often are the fittings switched by the sensors? Also why a microwave detector?

Timers are set to max.Also including reactive dimmer switches aswell.
does anybody have any ideas on why the lifespan of the tunes is so short?
many thanks.

What is the timer duration (in secs or mins) when set to max?
Why are the fittings dimmed? Do the fittings have dimmable ballasts?
Make and model number of the reactive dimmers please.

 

[067256]

ELAN 600x600 modular fittings 4x14watt T5. Dimmable ballasts fitted. Main office area dont use the dimmer facility anyway. Timer max setting is 99 minutes.
MK K4900 retractive dimmer switch.
head stratch!

 

[Marvo]

I'm still itching to know about the microwave detectors....

Were the dimmable ballasts a retrofit or original spec with the fittings?

Have you run the usual tests on the supply circuit? What were the figures?

Are the frequent tube failures an even occurrence across all the fittings or are some worse than others?

 

[MarkieSparkie]

Your short life tube problem is almost certainly due to frequent short time period switching caused by the PIRs and other detector/controllers.
Short time period switching of discharge lighting is usually accepted as any ON time shorter than 30 minutes.
Discharge tubes and ballasts can be killed extremely quickly by frequent short period switching, the typical life span can be reduced a 1000 fold in extreme cases, meaning a typical 4000 hour tube is killed in just 4 hours of operation! Any discharge lighting installation where the switching is controlled by any type of active detector(s) requires special, and sometimes expensive, pre-heat control gear on all effected lamp fittings to minimise this.

 

[Farmelectrics]

markie were did you get the 30 min time from take on board what your saying and its true but on the buildings were looking after they have a delmatic lighting system controlled by sencors and a main pc the switching is 15 min from last detection and all tubes have a good life span having said that the lights very rarely switch off in office hours due to constant movement just interested to see were you got this 30 min switch period from thanks in advance

 

[MarkieSparkie]

markie were did you get the 30 min time from take on board what your saying and its true but on the buildings were looking after they have a delmatic lighting system controlled by sencors and a main pc the switching is 15 min from last detection and all tubes have a good life span having said that the lights very rarely switch off in office hours due to constant movement just interested to see were you got this 30 min switch period from thanks in advance

The information came from notes made on the Osram Lighting Installers Course 1996. The 30 minutes ON time was stated as the minimum practical to assure typical average service lamp life.
The fluorescent tube life is based of the IEC standard 3 hour switch cycle; 2 hours and 45 minutes ON 15 minutes OFF. Lamps are tested in random selected batches, the IEC switching cycle is repeated until the lamp fails, the average lamp life of the batch when 50% of the lamps have failed is then taken as the typical average service life.

 

[MDJ]

I'm still itching to know about the microwave detectors....

?

Probably specified by an engineer who designed the original scheme Marvo, I have done Jobs where they insist on them being used, No idea why but they do like them haha

 

[Marvo]

Hmmm, they're obviously a fashion that's not reached my neck of the woods yet.

It might be worth checking that none of the control devices have a solid state switch output instead of a relay. Triacs and diacs can cause high voltage spikes on the leading edge when switching. That said it would be more likely to cause premature ballast failure rather than tube failure so I'm probably clutching at straws. Possibly a poor relay contact or intermittently chattering relay might be worth a look.