deuterium lamp intensities

[grumbledook]

I had to replace the deuterium lamp of my Dionex DAD-3000RS for the first time.
(It's a Hamamatsu L2D2.) The lamp was on for about 2200 hours. The original
intensity ~24 hours after first ignition was >8.5x10^6 cps. I decided to replace
the lamp after the intensity has dropped to ~2.3x10^6 cps, causing a significant baseline drift
already. I installed a new lamp. And after >24 hours its intensity is at ~3.7x10^6 cps
(and not increasing further). I was wondering if such differences in initial lamp
intensities are common or if the new one is some kind of "Friday afternoon bulb".

[DR]

Things to investigate:

alignment of the lamp (even though it may appear to be a "drop in" component, minor adjustments can make a difference as there is some lamp to lamp variation to overcome)

age of the new lamp (D2 does have a half life)

age of the mirror and other components comprising the optical bench (they get foggy over time, need replacing eventually)

Are the noise and drift reduced with the new lamp?

[HPLCCONSULT]

'DR' has made some excellent comments, but I would like to add one or two more points that I teach in chromatography classes.

(1) D2 lamps have a SHELF life. Sitting on the shelf unused for one year can cause a loss of intensity/signal. Some manufacturer's date code their bulbs to help you keep track of this. *This may or may not have nothing to do with the OP's comment, but I wanted to toss it in.

(2) Intensity ? Who cares. Better than nothing for sure, but not scientific if you are comparing actual signal outputs from lamp to lamp. Measure the actual baseline noise and signal to noise of a standard to compare lamp to lamp output. High intensity does not mean the lamp is stable or will provide a nice signal. Until you actually measure the bottom (noise level), you have no way of comparing outputs. *We use actual noise levels and actual signal to noise values to determine when a lamp is due for replacement, not intensity.

[grumbledook]

Thanks for your comments. I found the first remark in HPLCCONSULTs post interesting. The serial numbers of both lamps indicate that the "new" lamp is actually older than the "old" one. ... Of course, the noise level and baseline stability is better with the replacement bulb but the expected lifetime of it until noise and baseline stability become unacceptable again will be significantly shorter compared to the first lamp. I only want to know if I would have to accept such differences or if this could be a reason to complain. In the end it is also about buying one or two of the lamps p.a.
Regarding the optical bench: The instrument is now operating for one year. I had to replace the VIS lamp a couple of weeks ago. And for this one the initial intensity was as high as for the first one (< +/- 10 %). So I doubt that the mirrors or lenses got foggy. And I there is no play at all. So if there is need for adjustment/alignment, I would call for the Dionex service.

[DR]

As long as the thing is working well enough...

Shimadzu lamps also more or less drop in w/o room for adjustment, BUT you often have to adjust the mirror a little to get both sample and reference energies up to their max. values.

The take home lesson is to buy your D2 lamps shortly before you expect to need them. If you bought the extra lamp nearly a year ago, you probably have little or no reason to complain. You may not get 2000 hours out of it, but it should be fine at least for a while.

[HPLCCONSULT]

All sounds normal to me. Expecting two D2 lamps to be identical is unrealistic. That is why each detector should be re-calibrated and adjusted (in the software) when a new bulb(s) is installed, before use in analysis. *Make sure you test the bulbs under identical conditions so you have a real way to monitor and document changes (An SOP should be created for this).

Note that I said to check for a product DATE stamp on the lamp or lamp packaging. Not all manufacturer's place serial numbers in sequential order on their products (I know this sounds odd, but trust me on this as I know of a couple of companies whom manufacture the lamps first and place them all randomly in a container. Later on, the lamps are pulled out one-by-one and numbered). All lamps will change over time and that I why we suggest you not "stock" spare bulbs. If you do obtain a new lamp to put aside for future use we suggest that you write the date you received it on the packaging for future reference.

BTW: tungsten lamps ("visible lamp") are very hearty and not nearly as complex as the D2 lamps so I would not expect any issues with them.