Page 1 of 1
Peark area is 90% less when switiching from 5890 to 6890.
Posted: Wed Dec 02, 2009 8:00 pm
by Nikko
When injecting a sample on a both a 5890 and 6890 we are seeing only 10% of the sample on the 6890 when compared to the 5890. The column and liner are the same on both instruments. The split and column flow are the same. Even the syringe used to inject the sample as well as the operater are the same. We have done this test on two different 5890 GCs and got similar results. Why would the 6890 show such a low conc? We have done troubleshooting for leaks, changed to a new column, and nothing changes the results. The signal value is lower on the 6890 when compared to the 5890. Would this have an effect? Any suggestions or comments would be appreciated.
Posted: Wed Dec 02, 2009 8:34 pm
by Bricevan
What kind of detector is on each of the instruments?
Posted: Wed Dec 02, 2009 9:04 pm
by Nikko
They are both FID.
Posted: Thu Dec 03, 2009 1:58 am
by tom jupille
seeing only 10% of the sample
Could you elaborate a bit on the problem? Do you mean that the calibration plots look OK on both instruments and calculated sample concentrations are down 90% on one? Or do you mean that *all* responses (calibrators and samples) are lower?
If it's the latter, check down around the bottom of you calibration curve and look at the signal/noise ratio. It's possible that you may just be seeing a voltage scale difference. If your signal and noise are both 10%, your sensitivity is actually the same.
Re: Peark area is 90% less when switiching from 5890 to 6890
Posted: Thu Dec 03, 2009 3:30 am
by thohry
When injecting a sample on a both a 5890 and 6890 we are seeing only 10% of the sample on the 6890 when compared to the 5890. The column and liner are the same on both instruments. The split and column flow are the same. Even the syringe used to inject the sample as well as the operater are the same. We have done this test on two different 5890 GCs and got similar results. Why would the 6890 show such a low conc? We have done troubleshooting for leaks, changed to a new column, and nothing changes the results. The signal value is lower on the 6890 when compared to the 5890. Would this have an effect? Any suggestions or comments would be appreciated.
How about peakshapes and retention times?
Have you measured the split ratio manually?
Also please check the column installation at both ends.
Posted: Thu Dec 03, 2009 2:43 pm
by GasMan
How are you measuring the output of each GC. Are you using the analog output to an integrator or A/D device. If yes, the fact that you get a signal 10% of the 5890 suggests to me that you could have a difference in the Range settings. This is the only control on the 5890/6890 that changes things in steps of 10.
Gasman