Film-Tech Forum ARCHIVE: Xenon Bulb Life Span

This is topic Xenon Bulb Life Span in forum Digital Cinema Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=16;t=002895

Posted by Justin Hamaker (Member # 2165) on 12-16-2016, 12:48 AM:

Tonight I was changing out a 4K bulb on one of my 3D houses, and I noticed the anode and cathode are in almost new condition and the glass showed now sign of darkening. Because warranties we change our bulbs religiously at the rated life (1500h in most cases).

The question I have is what actually causes bulbs to fail? If a bulb shows little signs of wear once it reaches it's rated life and still produces a bright and stable light, what is the reason the bulbs should be changed at the rated life?

When we were running film, we had a couple Christie lamp houses which would regularly give us 5000-6000+ hours with a 2K bulb and still put up a good picture. But with the 3K bulbs on the Xetron lamp houses we usually couldn't go much past 3000 hours before the bulb would start to dim and/or flicker. On those houses with the 2K bulbs, I wonder if it was because we were usually running the power at the lower end of the bulb's rating and not putting as much strain?

Posted by Steve Guttag (Member # 268) on 12-16-2016, 07:49 AM:

First off...there are MANY factors.

Cooling...lets start there. Air flow over the lamp and through the electrodes go a long way to allowing a lamp to reach its best potential.

Lamp construction. Ushio/Christie have, historically, done the best job at keeping the electrode material on the electrodes and not depositing on the envelope. Once a lamp darkens, it starts this downward spiral. It increases heat while reducing light, which begets one to increase current, which increases heat, which increases decay...etc.

A Xenon lamp will have a safe operating range that it can run in. Too low and the arc is unstable, too high and you exceed the current capabilities of the construction. Within that range, generally, the lower the better. Higher current is higher heat and likely faster decay of the electrodes.

Igniter and power factor in. Each lamp strike is a potential for decaying the electrodes (in-rush current). Noisy power (ripple) will decay the electrodes.

So you have electrodes that can be damaged from the power/ignition that can lead to an unstable arc.

Normal wear and tear. As a lamp burns, the electrodes will experience a "burn-back" which opens the arc some (making it less efficient and harder to strike). Digital lamps typically have smaller arc gaps to begin with. This lets them be more efficient with light. Ideally, you want a point-source of light (lasers are good with this). As such, once the electrodes decay enough from consumption as well as sagging (affects larger, heavier hotter electrodes more than smaller ones) you will get to a point where a lamp that was steady before becomes radically unstable the next instant. It really is like going over a cliff in a short arc lamp. In the traditional film lamps they would decay more gradually.

On the upside of the digital lamps, their decay is much more shallow and the newer longer life lamps from Ushio/Christie are VERY shallow...UNLESS you run them at 100% (or above in a Christie "overdrive" situation). Heat is the enemy and see above with the death spiral of running a lamp at the top of range.

Catastrophic failure (explosion) is either a defect in manufacturing (may not even survive transport), bad packaging that allowed the lamp to stress in shipping or deal with the pressures of operation or, most commonly, the breakdown of the seals used to keep the xenon in and under pressure. One cause of seal failure is...you guess it heat. If air flow isn't adequate, the seals will heat up.

There are other factors in there too that are projection equipment related that can affect lamp life. Christie, on their 20 and 30 series projectors have the lamp tilt with the projector. This is hard on a lamp since the arc now is try to deal with those forces too. Christie's smaller projectors, Barco and NEC mount the lamp so the projector tilt as zero effect on it.

Another thing a film system had going for it was the shutter. That 48Hz flicker hid some evils of the xenon (and carbon) system. You are much less apt to notice an arc flicker superimposed on a flickering image unless it is really wandering. For digital, there is no such mask. Any arc movement is there for all to see on screen. Furthermore, people are much more emboldened to run solid white test patterns (or other solid colors) which will allow one to see the problems more.

I can tell you, I have one customer that went some 8000 hours on their xenon lamp in digital so it does happen but if a lamp like that lets lose, think of the damage it could cause. These lamps set right behind a VERY expensive light engine and glass reflectors. Most have taken the position that they rather not have to buy the damage it could cost (plus the lost performances). Heck, if you have Dolby 3D, you could have another $10,000 device damaged on top of it all.

Posted by Dave Macaulay (Member # 813) on 12-16-2016, 10:05 AM:

All true.
The warranty hours are a technical and business decision. Technically a lamp will eventually fail catastrophically or start extreme flickering, the hours to that depend on all the stuff Steve said.
The business decision - they want a warranty life low enough that failures in warranty are rare, they don't like paying for mirrors either. They need to compete as well, warranty hours and upfront price are the things interest the accountants.
The really big chains may have negotiated lamp deals for quite low prices but with no warranty - or a very very limited (like 5 hours) warranty - and may even run the lamps well past the normal warranty hours until they flicker or blacken. Replacing mirror sets now and then is cheaper for them in the long run than paying for the warranty "insurance".

Posted by Jim Cassedy (Member # 4115) on 12-16-2016, 11:34 AM:

Some years back, Christie/Strong put out an excellent 'white paper'
titled "Xenon Bulb Failure Analysis". I've kept it as a reference document
on my computer for years. But, I just discovered there's also a copy
RIGHT HERE in the FT Warehouse.

Happy Reading!

Posted by Mark Gulbrandsen (Member # 72) on 12-16-2016, 02:29 PM:

Another reason everyone here over looked is that short gap lamps operate at much higher internal pressures than did cinema lamps. 3 to 4 times higher internal pressure. That alone is enough reason for me to switch them out when the warranty goes ding.

Mark

Posted by Leo Enticknap (Member # 534) on 12-16-2016, 04:06 PM:

Quite apart from the risk/reward/Russian roulette equation of trying to squeeze a few more hours out of it vs. the chance of a kaboom, their light output declines as their design life progresses.

The choice of bulb model you make in the first place will depend on how much light you need to achieve 14Ft-L on a peak white, given your screen size, throw and the other variables that are specific to your theater. As Steve notes, you ideally want a bulb that will give you that operating at the lower end of its designed current range (but not off the bottom of it). Part of the reason for this is that as the bulb ages and the arc gap increases through cathode and anode wear, you have "wriggle room" to increase the current, such that you will be able to maintain that 14Ft-L throughout the bulb's design life.

Most d-cinema projectors will take care of the adjustment automatically for you (Barco calls this feature Constant Light Output, or CLO: it's based on a sensor in the light path). You install the bulb, adjust the current such that it's giving you 14Ft-L, calibrate the sensor (tell the projector: "this is the light level that I want you to maintain, so please keep adjusting the amps as necessary to maintain it"), repeat for the LSC files of all the other aspect ratios and 2-D/3-D combinations you have presets for, then forget about it (apart from taking occasional light readings - I try to do that every two weeks or so in our screens) until the next bulb change.

However, the time will come when, even with the amps maxed out, the bulb can no longer give you enough light for your projected image to be within the DCI spec. This time should not come until at or near the end of the bulb's warranty hours, but when it does, you need a new bulb.

Posted by Carsten Kurz (Member # 5396) on 12-16-2016, 04:27 PM:

There may be issues building up within the glass or glass/metal joints that may not be visible to the naked eye, even when the electrodes still look good.
Neither deteriorated electrodes nor a blackened envelope will make your bulb explode, but a weak glass will.

- Carsten

Posted by Monte L Fullmer (Member # 2797) on 12-16-2016, 05:03 PM:

I remember xenon bulbs in the 1980's were prone to explosions within the first few hours of life where some of it was construction of the bulb itself, lamp construction - to allow bulb expansion when they reached temperature, and the similar.

I remember vertical bulbs can really last past their warranty-like 8k hours, esp in Cine lamphouses, and even CFS consoles if the air flow and cooling were very strong.

Then, we had the cheaply constructed bulbs of: ORC, Perkin Elmer, LTi, and the similar that were hard to trust to keep any form of duration. Even Osrams, where they were paramount in quality, but since they got absorbed with Sylvania, the quality seems to have really plundered.

-Monte

Posted by Justin Hamaker (Member # 2165) on 12-16-2016, 05:17 PM:

One point I can add is that on our Barco 32B we had the issue with the magnet being installed incorrectly. After flipping this around we started getting the full 500 hours on a 6.5K bulb. Prior to that we started to have striking and dimming issues at around 350 hours.

Posted by Steve Guttag (Member # 268) on 12-16-2016, 06:00 PM:

Another big clue that you have an inverted magnet is the darkening on the envelope. If the magnet is backwards, that envelope will darken QUICKLY.