Waters 486 UV detector error 402 code fix.

[Shralping]

Hi all.
I thought I would advise that I had a error code come up on an older waters detector. The code 402, going by the paperwork tells either the lamp is failing (not enough UV light) or the cell is dry. I was sold this unit and told the lamp would need replacing someday soon as it had seen a lot of hours.
I couldn't wait for delivery of a new one so I removed the top off chassis, removed the cover box screws and lifted it out to check the glass lens.
The lens had a solar burn in it. This was clouding the lens to the point that I figured it would not let light through well.
To fix this I removed the lens, and since the burn was offset, not in centre of lens, I spun the lens around so that the burn was opposite its old position.
Then the UV light would pass through the metal slot and enter clear glass.
Upon start up and calibration. It began to work. No error message, and I could see the wavelength read out.

I have noticed however, that the absorbance value ( above wavelength on screen) drops further and further into the negatives over time.
If I hit the Auto zero button it resets back to zero, but over time, it will drop again.
I don't expect this is because the UV light strength is wearing away that fast but it may be.
Is there any other reason for the absorbance to fall into negative numbers over time without running any solvent through the cell? I can check this when the new lamp arrives to see if this is usual.

[javascript]

Was this absorbance drift happening before you starting the error?

I am not totally sure there is a problem.

As the instrument heats up, the readings from the A/D converters could increase, and to compensate for this increase the auto zero factor is also increase.
I could go how this may happen, but just know that there are (probably?) two 'light' detectors, one before the flow cell and one after.

If you were running with fluid you would just want to make sure to do and autozero before starting your injection.

Just make sure that the environment that the instrument is in is at a proper temperature (see the manual, it is probably around room temperature).

I would recommend calling/emailing some sort of customer service just to get a confirmation

[Shralping]

Hi JavaScript,

I could not see if the absorbance drift was there prior to seeing the error code simply because the unit ( and HPLC) is new to me and the error was the first thing the techs saw when they installed computer software and set her up for me.

I guess you have answered my question if it is true that the absorbance drifts at all like this.
I did not see two light detectors, but I didn't look. I did notice with the older shimadzu LC6 series detectors, some have a green led light to indicate the UV bulb is on, while other units of the same vintage and stamping seem to have a second green led light to indicate if the second (non UV light I think) is on. I noticed this by looking at various ebay ads. That did make me wonder if they ALL have two globes or not.

I thought that maybe the absorbance read out changes with the generation of ozone build up, but temp might be it too as globe warms.
The room is ok for stable temps.
I was watching this absorbance value change but now that my controller screen is not longer showing functions etc I can't run a solvent through
her to see what range the drift is over say a 8 minute run. When I do a run with solvent, then I can make note of this and how that might intefere with the represenation of peaks or not. I also have a new UV globe on the way, so I will replace the older, and see if there is less or more drift.

If its not severe or is somewhat expected, hence the Auto zero button, then all is well. Thanks for your comment!

[ScottHorn]

I'd like to offer another suggestion regarding this detector that may or may not be related to the drift you're seeing. Either way, it won't hurt to try this. I don't have the manual handy, but there is a diagnostic function you can access that will show you the reference energy of the lamp. You press the "diag" key, then a 2 number code listed in the manual. The display will then show the reference energy of the lamp. I believe something around 100 is what you should expect from a new lamp and good optics. Once you see this display, open both ports on the reference cell (the one to the left of the actual flow cell). Now take a syringe and gently push a quantity of methanol through the cell from the outlet to the inlet (reverse of normal). As you do this monitor the reference energy. In my experience it will climb for a while then stabilize. Once it has done so, while maintaining pressure on the syringe, close the inlet, remove the syringe from the outlet, and close it as well. I find that I have to repeat this procedure once a year. I can tell when it is time to do this, because I will see a drift in my baseline that corresponds EXACTLY to the cycling of our air conditioning system in the lab.

Scott Horn

[Shralping]

Excellent information Scott. I never had much of a manual but the small amount of paperwork I do have speaks of what you mention.

I found that after setting the wavelength to 230nm ( it was set at 330) you press the diagnostic key followed by 19 then enter key. This gives a reference value at the top of display. This was 23, then fell to 22 then to 21. I assume that your desirable value of 100 is under a situation of reading 230nm?

This is without pumping anything through though. I shall do the pump solvent test shortly. Thanks. It seems that my number is way lower than 100 so it is probably a tired old lamp. One is coming in the mail soon. This will be a good test for how old this one is. However there is a system in the manual to set up the lamp height and degree position to possibly maximise the lamp read or get it within a desired range but hopefully I will sort this as the lamp arrives.

Thanks heaps for chiming in. This helps me a lot.

[ScottHorn]

The lamp alignment procedure you have to do when installing a new lamp can be a bit tricky. I've put 2 lamps in mine so far, and each time, the optimal position (rotation-wise, not height) was at one extreme of the range over which you can adjust. It took an extra pair of hands to "firmly" hold the lamp in place while tightening the screws to lock the lamp in position.

I don't quite remember where I heard that "100" value quoted as being good for a new lamp, it may have been on this forum. I remember the first lamp I changed out had dropped to something like 13 before I took it out, and the last one I installed and aligned yielded a value of 87 (maybe?) after all was said and done. Before you put in your new lamp, try the flow cell flush and check your lamp alignment. You may be able to squeeze a few hundred more hours out of your current lamp before switching to a new one.

edit: Forgot to add, yes, 230 nm was the wavelength I used. You can monitor the reference energy value while you do the reference cell flush to see when you've flushed enough. Same thing for the lamp alignment, you just watch the reference energy value as your raise/lower/rotate the lamp.

[Shralping]

Thanks again Scott, again, great info.
I received the new lamp tonight.
After replacing it in the more or less same positions as the last lamp, the reference read at 23 again. Same as old lamp.
Maybe the old lamp is still ok for a bit yet. So then remembering your 100 value, I started adjusting and the most I could get out of it was
60. This was after adjusting height, then rotation then height again. I am happy with that for now.
Ill do a flush soon, however my LC 10 ATVP pump was coming up with an error code "not protected" and I pulled her down, she was filthy inside with dust. Changed the memory battery but the error persisted. So she is in the shop for repair for now. As such cant do a good cell flush.

The other thing it could be reading a little low is because of the solar burn in the eye lense maybe. Even so, am I wrong to assume that
as long as this 60 value stands for a while and it lasts strong for the stated 1500 hours, wont the reference standards I will use see the thing is calibrated to this amount of UV? So will be just fine? and yes the manual page does say 230nm for the diagnostic.

One other thing I noticed was the old german tube had some kind of plastic piece inline of the red power cable that had a digital display screen.
Crystal diode I think it is called. Like a calculator screen in a plastic rod. I didn't notice if it was working to show a display or not. Though the new lamp had no such arrangement attached. I can only think it might read amps or volts in and was someones idea as a check.
Any ideas what that device is? By the way mine sites 2mm anti clockwise from its farthest rotation. Which is 2mm back from where it used to be locked up. They had put texta in to show position. So their position was like yours but potentially that may have been to get around this units solar burn in the eye piece which is now out of the way. Might be worth checking your lens for solar burn next change.

[ScottHorn]

60 Should be just fine.
I have no doubt my optics are solarized. I bought this instrument used and it has seen heavy service in the time I've had it. I actually think I have a replacement window for the lamp compartment (the part you describe as being solarized). The problem is last time I was inside the unit doing a lamp replacement, I couldn't figure out how to get that little window out. I was hesitant to use too much force for fear of misaligning some part of the optics.
In the mean time, while your pump is out of commission, you could always flow your mobile phase through the reference and sample cells using a syringe. The label on the cells say they can handle 1k PSI, so you should be safe with a syringe.

[Shralping]

Ok Thanks again, I'll give the syringe a go.

I have a feeling I might have been experiencing power surges since a lap top repairer said thats what crashed a lap top. For the lense, I had to unscrew the main "black out" box and lift her out vertically. That's the cardboard looking thing that has a bunch of screws all around and isn't a square shape.

This then allows you to access the lens window real easily. If you tried to pull it out while sitting next to the lamp it would be harder.

My lens was held in place with some kind of black silicone or sikaflex. Im not sure if this is factory standard though. I was able to peel the lense out without losing the black silicones shape.
Attempting to clean the lense of the burn in failed. So I spun it around and placed it back in, then got some roof silicon and went around it to make sure it wouldn't fall out. So far so good I think. It might give me another 15 years. Might not. If you saw a 100 value you are looking good.