I have a method set up on two GC's, a PE-Autosystem and a HP5890. Conditions are as follows:
Column: RTX 1701 0.25mmID x 1u df
split flow: 100ml/min
make up gas (argon/methane): 60ml/ min
carrier (He): 16.5psi
signal: 1mV = 80hz
My problem is when I analyze my sample (mostly halogenated hydrocarbons) I get 10x lower response on the PE instrument than I do on the HP, but when I analyze room air to determine dead time, the response is the same for both instruments. Can anyone help?
By Russ on Thursday, June 29, 2000 - 12:42 pm:
I am not familiar with the PE system. I am assuming from the information you provided that you have eliminated any difference in units from the detector response. HP and PE used to use different units for detector response. I know manufacturers use increased sensitivity as a major selling point. The difference you are seeing could be the difference in the two systems. However, we found the response you get from the HP ECD will vary depending on the level of the background signal at the start of the run. Using 3396 integrators with INET full ranging, this variation is somewhat linear for peak areas. That is an increase in the background signal level by a factor of 2 increased the peak areas by roughly a factor of two. This was confirmed by HP tech service. I don't know what type of variation you would see with the ChemStation software. What is your starting background signal level? It should be less than 100 and we got less than 30 for a new detector using a DB-1 column.
By Ron on Thursday, June 29, 2000 - 01:05 pm:
You should look at the signal to noise ratio, and not just at signal level. If you have equal signal to noise, then your sensitivity is equal, and the detectors should have the same detection limit. Obviously two detectors from different manufacturers (or the same manufacturer) will not be absolutely identical, but I would be surprised to see a 10X difference.
By Anonymous on Monday, July 31, 2000 - 09:43 am:
I haven't run a PE for awhile. As I remember, changes in make-up gas flow through the detectors REALLY affects the response. I believe the manual recommends 30 mL/min flow. If that's correct, try it, that should significantly increase the response. Also, if you have new ECD's, or ones that have just been cleaned the response will be less than a "dirty" ECD.
Also, I believe switching to N2 as makeup gas will also increase your response. Also, the ECD's on some of the earlier PE's seem to give less of a response than the newer ECD's but run a lot cleaner.
By PGPG on Saturday, October 28, 2000 - 08:09 pm:
CHECK THE PERFORMANCE OF YOUR PE
WE MAKE THE SAME COMPARISSON THE LAST WEEK AND
AUTOSYSTEM HAVE 2X SENSITIVITE THAN 5890
By Ndimele Louis on Tuesday, January 30, 2001 - 03:02 am:
I am operating an SRI 8610 GC fitted with an ECD. I had some analysis that were good 3 days ago, I left my machine on with N gas in line running overnight. However, I noticed that I no longer get responses on my GC. no response i.e. no pulse at all! I have tried heating the system The oven temp at 200oC for 24 hours and the ECD at 300oC, yet, I do not have any pulse on the system. What could be the cause of this problem. I am analysing Halogenated pesticides and the carrier and make up gas is nitrogen and the detector temp is 300 and the programmed temp moves from 150 to 250oC for the column.
By Anonymous on Tuesday, January 30, 2001 - 12:41 pm:
I haven't used that instrument. Here are some general suggestions: Is there any way the detector board might have been turned off? On the HP/Agilent GC's, the electrometer can be turned off. If this has happened, you will see no signal. Are you sure you are getting column flow to the detector? Try cooling the oven and detector down, turn off any detector make-up gas, and check the detector outlet for flow. If you have no column flow to the detector, check the usual suspects (septum leak, column installation, column fittings, etc.). I am assuming you are making injections from a syringe. Are you sure it is working properly? If the syringe is plugged, it may not be able to draw up the sample. You have probably already done so but, also make sure your inlet parameters are set properly (split vs. splitless, splitless purge time,etc.). If you are using an "external" sample introduction method (eg purge and trap), make sure it is operating properly. If possible, disonnect it and inject a sample through an instrument inlet.
By Michael Brown on Thursday, June 10, 2004 - 11:31 am:
I am also operating an ECD on an earlier model SRI 8610 GC, but unfortunately I had to shut down the GC to replace the oven heater. My research involves analyzing for halogenated disnfection by-products in water. I set the detector temperature at 300 C and usually start my analysis at 50 C and end at 200 C. I read from SRI that water may have formed inside the cell which could affect performance. So I heated up the ECD while having a N2 flow rate of ~40mL/min. The problem I am having is that the millivolt reading is correct, yet when I inject a 100ppb sample of carbon tet I get a low response compared to before.
Posting is currently disabled in this topic. Contact your discussion moderator for more information.