Large peaks from Agilent 1200 DAD

[Inst Tech]

Hi,

I have 2 Agilent 1200 DAD systems sat next to each other, bought at the same time, exactly the same 'spec' (exactly the same everything).

One of our analysts came to me and said one of them has 4 times larger peaks for the same runs...

I have run an injection on each system of my own PQ sample using the same column, same phases even the same vial of sample and one system has peaks that are ~ 1/3 bigger than the other!!

My 1st thought was the injector was picking up too much sample but I performed a 'lift' test (weighed a vial of water, carried out 10 x 10ul inj then weighed the vial again) and found that it is picking up the correct amount (10ul).

I spoke to Agilent and they do not know what the cause is...

Any one else had a similar problem?

Thanks

[bhuvfe]

Maybe you can try to switch the flow cells between the two DADs and see if you still have the same factor 3.

Do you run the same kind of samples on these two 1200 systems?

[HW Mueller]

We are not being told whether

"bigger" refers to peak hight or area,

the settings are the same on the detectors, etc.,

the flow rate is the same

both detectors operate in their linear range

selfchecks of the detectors have been performed and what these results are.

Also, is the difference 33% or 300%?

[Inst Tech]

The 'bigger' refers to peak height. The analysts are getting 4 times but I get 300,000mAU when I should be getting 200,000mAU.

The settings are the same on the detectors and rest of the system, the flow rate is the same - basically these are set up with hardware, software settings, samples, phase and column (for the samples, phase and column I moved them from one system to the other) are all exactly the same..

"both detectors operate in their linear range" sorry I am not sure what you mean by that...

"selfchecks of the detectors have been performed and what these results are. " I will run these again and make a note of the results..

[Peter Apps]

Leaving aside the rather wide range of different differences between the two systems, do you know which of the systems is "correct", either one is producing peaks that are bigger than expected, or the other is producing peaks that are smaller than expected. Maybe the autosampler on the system that gives smaller peaks is picking up less than it should ?

Peter

[HW Mueller]

What I was asking is whether you might have a blackout, or nearly so, in one of the detectors.
Are your units rather µAU?

Did you try bhuvfe´s suggestion?

[Hollow]

I don't have exactly a clue, just writting what came to mind:

- what is your software?

- are both systems connected to the same PC or has every system its own?

- if two PCs, have they the same software version? (as far as I remember, we observed differnces of factor two between two similar systems (1200RRLC) on different versions of EZ Chrom...)
-> maybe you can interchange the systems/detectors and PCs to see if it's a hardware or software issue

- does this difference cut it out during quantitation (means standard and samples are different by the same factor?)

- are there any hardware switches different (maybe on the back of the systems or even on the electronic boards inside?)

- maybe you can export the raw data to an ASCII file and "evaluate" it in Excel (with EZ Chrom there are some additional information in the file header of the exported file such as "scaling factors" for the ASCII data)

good luck

[Consumer Products Guy]

Check that DAD wavelength bandwidth and any reference wavelength is set the same.

[Inst Tech]

What I was asking is whether you might have a blackout, or nearly so, in one of the detectors.
Are your units rather µAU?

Sorry the units are µV..

Did you try bhuvfe´s suggestion?

Am waiting for the other system to become free so that I can 'borrow' the flow cell..

Check that DAD wavelength bandwidth and any reference wavelength is set the same.

They are

what is your software?

Empower 2

are both systems connected to the same PC or has every system its own?

Both are connected to the same 'lace box'

are there any hardware switches different (maybe on the back of the systems or even on the electronic boards inside?)

Not that I am aware of - there shouldn't be

maybe you can export the raw data to an ASCII file and "evaluate" it in Excel (with EZ Chrom there are some additional information in the file header of the exported file such as "scaling factors" for the ASCII data)

Will see if I can do this from Empower..

Leaving aside the rather wide range of different differences between the two systems, do you know which of the systems is "correct", either one is producing peaks that are bigger than expected, or the other is producing peaks that are smaller than expected. Maybe the autosampler on the system that gives smaller peaks is picking up less than it should ?

The one with the smaller peaks is definately correct as I have tried the 'sample' on other systems and it is the same..

[ym3142]

are the two UV lamp energy level the same?

is it possible one of the flow cell has more contamination than the other?

[thohry]

Have you checked the method carefully?, especially the detector's parameters. How about the retention times and peak shape? are they nearly identical? then If I were you, I would swap the flowcell and the lamp one another.
Hopefully, you will find the culprit.

[Inst Tech]

are the two UV lamp energy level the same? is it possible one of the flow cell has more contamination than the other?

I have checked the lamp intensity of both detectors without the flow cell in and get the following:-
System----------------------------Bad------Good
190-220nm (>2,000)-----------42,096----12,671
221-350nm (>5,000)-----------69,408----25,172
351-500nm(>2,000)------------51,067----17,525
501-950nm (>4,000)-----------52,853----16,516
190-350nm (<450,000)-------222,618----78,045
700-950nm (<300,000)-------210,744----76,350
D2 Alpha line (<1,200,000)---331,253---170,467
(Sorry for the dashes it's the only way I could get a table)

As you can see the 'bad' system has a much higher intensity than the other...

Have you checked the method carefully?, especially the detector's parameters. How about the retention times and peak shape? are they nearly identical? then If I were you, I would swap the flowcell and the lamp one another.

I have checked the methods (and got someone else to check them) so they are the same. The retention times are spot on and the peak shapes are similar (except size) both are giving good chromatography other than that..
I am unable to swap the lamps due to the amount of 'change control' (qualification tests) that would be required! Although soon I might have to resort to taking the good system out of use and just sit playing with them both..

[ym3142]

I appreciate your constantly coming back with your data.

I unfortunately do not know the relationship between the UV absorption (out of DAD) and the lamp energy level. We need someone who are really good at physics.

Luckily, you use the data of the reference standard and samples from the same instrument. The difference we are discussing should not effect the testing results.

[Hollow]

as you've said your detectors are on a LAC/E Box, maybe you can interchange just the ports and see if the differences belongs to the detectors or are "generated" in the box.

As you're using Empower, compare the methods by looking at the print-out of the "instrument method report" for your instrument methods.

We're aware of an issue, where sometimes our pump module (600E) just have a different "initial" parameter set than we can see in the gradient table on the computer screen.
(this happens sometimes when we change from "isocratic" to "gradient" modus in the instr.method. On the screen everything looks fine but it sometimes hold the old initial settings from the prevoius isocratic method. Can be seen on the instr.method report.
After reentering the value for the inital step and save it, the problem then is solved)

[bisnettrj2]

You said an analyst came to you and said "one of them has 4 times larger peaks for the same runs..." Was this an 'all of a sudden' problem? Did any maintenance precede this change?

One quick check idea - in the DAD Control screen, are your Analog Output voltages the same? I've no idea if that makes a difference, but maybe with the LAC/E box, it does? Probably not.

Also - is the Vis lamp off on both instruments?

Did you check the DAD control window to make sure the "bad" detector isn't set up to switch to a different wavelength at the start of the run?

You can also run the DAD Self Test on both just to confirm that both detectors are functioning within spec. If they both pass, I'd run the Cell Test to check the cleanliness of the flow cell.

Also - what do the Intensity Tests look like on the other instruments? Are they all closer to the good instrument, or the bad instrument, or are they all over the place?

Finally, what does the noise look like on the "bad" system? Is it also three times greater than the good system?