Chemstation calibration table issue

[Michal.Galkowski]

I'm running Agilent Chemstation with GC6890N for the atmosperic composition measurements.
My system is configured so that there is exactly one injection per method and in each method
i have between 5 or 6 compounds. What I would like to achieve is to have a txt report that
lists the compound names (e.g. CO2 or CH4) followed by the peak detailed data (RTs, heights,
areas and so on).

I've successfully managed to do that using the calibration table of the Chemstation, but after
some 30 analyses my Chemstation is displaying an error "Undefined symbol N", and you need
to accept the message for the sequence to go on. This is unacceptable, since my measurements
run continuosly and I cannot monitor them every 10 minutes.

Has anyone encountered this problem before, or perhaps know a better solution to my problem
than using the calibrations table?

[dblux_]

...
Has anyone encountered this problem before, or perhaps know a better solution to my problem
than using the calibrations table?

You absolutely need calibration table for every method, Otherwise your analytes won't be identified nor quantified.

Maybe you have a problem with datafile names accepted by ChemStation ? (What version of ChemStation controls your GC ?)

[Michal.Galkowski]

I'm using the Rev. B 04.03. [16] version.

I have calibrations table for every method that I use. Also, datafile
names should not a problem, since the error occurs randomly (it seems). The same method can be run for a number of times without any troubles and still crash at some point.

I did a little research and I've found an interesting clue in the sequence log file:

Method Method completed 18:05:09 08/20/13
7da 41e2 5213afd7 0
Method Loading Method 02_03.M 18:05:11 08/20/13
443 41e2 5213afde 0
Method Method started: (F) line# 32 blank run inj# 18:05:18 08/20/13
67 41e2 5213afe0 0
Method Instrument run started (front) 18:05:20 08/20/13
0 1da5 5213b27f 0
6890 1 Not ready:Multiposition valve at runtime 0.0 18:16:31 08/20/13
0 1da5 5213b27f 0
6890 1 Not ready:Aux 3 pres 399 kPa at runtime 0.03 18:16:31 08/20/13
0 1da5 5213b27f 0
6890 1 Not ready:Aux 5 pres 474 kPa at runtime 0.03 18:16:31 08/20/13
0 1da5 5213b27f 0
6890 1 Not ready:Aux 4 pres 194 kPa at runtime 0.18 18:16:31 08/20/13
875 41e2 5213b27f 0
Method Saving Method 02_03.M 18:16:31 08/20/13
67 41e2 5213b280 0
Method Instrument run completed 18:16:32 08/20/13
70 41e0 5213b280 0
CP Macro Analyzing rawdata FID000032.D 18:16:32 08/20/13
70 41e0 52145133 0
CP Macro ======================================== 05:33:39 08/21/13
70 41e0 52145133 0
CP Macro In Command ObjHdrText$ in macro CopySampleVa> 05:33:39 08/21/13
70 41e0 52145133 0
CP Macro rs 05:33:39 08/21/13
70 41e0 52145133 0
CP Macro at line 1176 in file C:\CHEM32\CORE\SIGNAL.M> 05:33:39 08/21/13
70 41e0 52145133 0
CP Macro CX 05:33:39 08/21/13
70 41e0 52145133 0
CP Macro Error 16420 occurred: 05:33:39 08/21/13
70 41e0 52145133 0
CP Macro Item doesn't exist 05:33:39 08/21/13
70 41e0 52145133 0
CP Macro ======================================== 05:33:39 08/21/13

The fragment in red is what triggers the error, it seems. It produces an error window with "OK/Accept" button. When pressed, it continues the sequence as if nothing happened. Except that the measurements are on hold until you do that.

[dblux_]

Whatever this CP macro is - seems that it stops GC run due to errors.

Can't help you with macros - sorry.

[bisnettrj2]

I can't help you either, but Googling macro error 16420 gave this page from Agilent:

http://www.chem.agilent.com/Library/Sup ... iance.html

Search for 16420 on the page to find the associated line. Doesn't look like there's a fix, at least yet, but maybe you can call Agilent and see if there is something they can do to help you develop a work-around. The problem described is not really like your problem, except that it seems to be related to automatic generation of reports, and it's the only reference to 16420 on the Agilent site.

[Wissenschaft]

Have you tried re-creating the method from scratch under a new name? I've noticed that every so often methods become corrupted on Chemistation and throw up funny (and annoying!) problems/messages.

[Michal.Galkowski]

Have you tried re-creating the method from scratch under a new name? I've noticed that every so often methods become corrupted on Chemistation and throw up funny (and annoying!) problems/messages.

I did save the method as a new file and then tried to go, but it didn't work.

I've noticed that after every properly executed method there are two files in the data directory: sample.mac and sample.mac.bak. However, when the "N problem" occurs, there are also 2 different that are created, but this time sample.mac and sampleu.mac (no .bak versions). The second one has only N letter inside, so I guess that is is the macro that is executed and produces the error.

When I run the offline version and try to load the signal, I get an error message "Item doesn't exist", and cannot produce a report file, even manually. But if I delete sample.mac and sampleu.mac, I can analyse the data without any problem.

Here are both macros:

SAMPLE.MAC:

Name SAMPLEINFO
ON ERROR Print ""
SAMPLETARGETMASS ""
ON ERROR
CLRRPTSMPLINFO
SAMPLECALISTD 1,0
SAMPLECALAMT 0,1,1
REMOVE SAMPLEINFO
RETURN

SAMPLEU.MAC:

N

[Wissenschaft]

Have you tried re-creating the method from scratch under a new name? I've noticed that every so often methods become corrupted on Chemistation and throw up funny (and annoying!) problems/messages.

I did save the method as a new file and then tried to go, but it didn't work.

I don't know a lot about macros, so I can't really help you there. However what I meant was that you make the method fresh not just resave it. Using the same method many times, even saved under a new filename, can get corrupted. I would open a new blank method template and enter all the parameters like I was creating a brand new method.

I'd love to know the solution if you figure out the macro issue though!

[Michal.Galkowski]

We've tried fixing the software using the Chemstation install disks.
"Repair" option did not work, so we've uninstalled it and reinstalled the whole thing yesterday.
So far no errors, but it's still too early to tell if the problem has been solved.

[Michal.Galkowski]

Just a quick follow-up: problem was not solved. We've reversed to Chemstation B. 03.01 using the WinXP virtual machine. Working for weeks on end now.

[James_Ball]

Just a quick follow-up: problem was not solved. We've reversed to Chemstation B. 03.01 using the WinXP virtual machine. Working for weeks on end now.

We had a problem once when trying to use the MSChemstation on a new PC. It had the correct XP SP2 installed and it would install and run the program for a few days then it would not even allow Chemstation to start. This was even a PC from HP that was a recommended model by Agilent. We finally just got a different PC and it works perfectly after that, we took the PC that didn't work and installed the LC Chemstation A.09 on it and it works flawlessly. Sometimes there is just a single file on the system that messes things up completely, either in the Chemstation install or even the operating system.

[GasMan]

It looks to me as though somebody has tried to delete the contents of the Sample.mac file, but did not remove the last letter ( the N in RETURN ). Remove this letter and the macro should run.

Gasman