5890 G1701BA can't get split vent open in method

[MSCHemist]

Hi all I've started working with my 5890. It is a Series IIa with FID and 7673 hooked with HPIB and remote cable. Anyways i am using Windows NT 4.0 and G1701BA Chemstation MSD to control it. I find the instrument parameters redundant and confusing like where you have to enter initial oven temp and init pressure even though you also enter it into the constant flow calc right below it. Anyway I have it to where it controls the detector, gas flow, oven temp, and ALS fine but I can't get it off splitless. I entered in the split flow (65ml/min), split ratio(50:1), and column flow (1.3 ml/min) in the injector pressure program section. I increased the total flow using the knob.

I looked at the purge valve section
There is an A and B (I don't have a rear injector) with -init value on/off -time on -time off and a check mark for splitless. I have splitless unchecked I thought the on and off times are only for splitless injections. What should the values be?

Then there is the valves/relay section. There are no relays but 4 valves listed with #3 enabled. I am unsure what they are.

I entered the above and every time it sends information to the GC it sets the split to 0.0:1. I watch it on the front key pad by hitting flow until it shows the split value. I can manually set the split flow to 65 ml/min on the GC keypad by hitting gold and flow however chemstation msd sets it right back to 0.0:1 the next time it sends information to the GC. What do I need to do?

On a related note I am so frustrated I started looking at third party software. I've heard Unichrom and GCwerks mentioned in old posts. Unichrom I'd need to hook of the 7673 g1512a box using the inet cables according to the instructions there. GCwerks I am uncertain if I'd need to hook it up using RS-232 which I don't know much about and haven't been able to find much information other than I know my GC has an RS-232 port and I may have some of the cables lying arround.

[cleh]

I know this can be frustrating and hope I can help. The constant flow calculator just helps to calculate flow and split rates. Once you have the initial pressure and temp, enter them above. The method only controls the carrier flow. The split flow is set manually with a bubble or electronic flow meter connected to the split outlet on the front of the instrument. Use the knob above it to adjust the flow. The purge valve section controls the valve state. Purge A is usually the front injector and B the rear. For split injections, initial time should be 0 and both values should be on. For splitless injections, initial time should be .5 or .75 min (or whatever you need for the column you're using) and valve off and then on. The only thing the splitless check box does is turn the valve off each time the method loads so you don't want that checked for split injections. This is a back pressure injector so the split flow will be whatever you set it to in the on and off positions. The split flow is redirected in the injection port and comes out of the split outlet on the front of the instrument. I know this is confusing, but you can call me if you want me to walk you through this (857) 277-9858. This software isn't bad, but you may want to upgrade because it's hard, but not impossible, to find parts for the PC. I'm not familiar with Unichrom or GCwerks, but I'm sure someone on the forum will have info. Good luck.

[MSCHemist]

So the times shouldn't matter in a split injection.

I use an electronic flow meter on the split vent with the control knob to set the flow. When I am running a splitless method I set it to 15 ml/min. For a 50:1 split I adjusted it to 65 ml/min. However on the GC display it keeps going to 0:1 split whenever chemstation sends method data to it and I can't figure out why. Is the valve is portioning so if I set it to 100 ml/min it will maintain the proper split and column flows and the rest will just bypass the injection port?

So the flow out the split vent should always be 65 ml/min if that is what the knob is set to. During a splitless injection the EPC lets 1.3 ml/min into the injector and of course the septum purge flow and the rest bypasses the injection port and just goes right out the vent.

I found some instructions for Thermoquest Chromquest which seems to have identical parameter interfaces to Chemstation. I found this notation

To enable split injection mode, check the initial state box for the desired inlet. It is not
necessary to have an On or Off time.

[MSCHemist]

I did some more reading. So the only thing the EPC is capable of doing is reducing the inlet pressure to below the supplied pressure to maintain the proper column flow. The split flow is controlled by the black total flow knob and the GC has no way to regulate the split ratio other than on or off for splitless injection.

I still don't get why when I toggle throught the flow displays on the keypad I see 0:1 split when I set the split flow with the flow meter to 65 ml/min nor why I am able to enter anything into the split flow by hitting gold flow [flow paramters].
http://www.chem.agilent.com/Library/Sup ... a16027.pdf

[MSCHemist]

Is the stuff under flow parameters on the gc just there as a built in calculator and doesn't affect any instrument settings?

Seems like having to set the split flow/ratio manually makes repeatability an issue in setting the split ratio rather than have it calibrated and controlled electronically on new instruments. Plus the lack of gas saver sucks. Fortunately I will be doing predominantly splitless work on it.

[MSCHemist]

Still a bit confused. I think injection port A EPC (front) is valve 3 and Injection port B (back) EPC is valve 4 because theb 5890 method file says set valve 3, off
set valve 3, ON, 0.50
set valve 3, Off, 5.00
and the same for 4 which would be the back injection port B.

I was setting it up for splitles so I select init value off, On time 0.5, and off time 5.0 with splitless checked. However, I kept getting erros and I'd set purge valve init value off and splitless checked and then I'd go look at the GC and it would be switched on and so forth.

Do I need to set stuff up under the valves/relays for the EPC configuration?

[KM-USA]

The valves/relays are not for the EPC.

We had this version on a GCMS until a few months ago, now decomissioned, so I can only go by memory for the other stuff. We used split 99% of the time.

[MSCHemist]

Yea I can't understand how the purge valve kept ending up open when I was making a method and saving it with splitless enabled init value off and on time .5min and off time 5min for a splitless method. The softare seems kind of buggy with data saving and purge valve control but I'm not sure if Unichrom or any other third party software can handle the instument as configured (2 HPIB cables and a remote cable hooking together the GC, ALS, and computer). Unichrom seems cheap but according to the stuff on the site the 7673 needs to be hooked up via INET cables.

[cleh]

Still a bit confused. I think injection port A EPC (front) is valve 3 and Injection port B (back) EPC is valve 4 because theb 5890 method file says set valve 3, off
set valve 3, ON, 0.50
set valve 3, Off, 5.00
and the same for 4 which would be the back injection port B.

I was setting it up for splitles so I select init value off, On time 0.5, and off time 5.0 with splitless checked. However, I kept getting erros and I'd set purge valve init value off and splitless checked and then I'd go look at the GC and it would be switched on and so forth.

Do I need to set stuff up under the valves/relays for the EPC configuration?

It is possible to use the valves to control split/splitless injectors, but I've only seen that a couple times. If I were you, I'd set initial valve 3 on and initial valve 4 on with no off time for split injections. I'd also set purge A and B initial time to 0 and both values on without splitless being checked. No matter what software you have, you will still need to figure out the hardware configuration. Open the small door on the left side of the GC. Sometimes there's a label on the inner side or inside saying where the valves are assigned.

[MSCHemist]

I know the 5890 has a 6 port EPC kit with A controlling the only inlet. The rest of the cords just hang out the back of the instrument unattached. I think at one time it was hooked up to a Tekmar LSC 2000 purge and trap. It was also hooked up to a 5972 MSd prior to 2006.

[Consumer Products Guy]

Idea: couldn't you just pull the little connector off the solenoid (take off left panel) that operates the splitless mode? Then it would always be "split" mode. We have one 5890 that has defective solenoid for that, and we've never bothered to have it fixed, so we use that one a "split only".

[MSCHemist]

I don't think there is anything wrong with the valve itself. The instrument controls it fine it is just the software seems to be having trouble communicating to the 5890 when it should be opened and when it should be closed.

I think if the software gives me any more issues I'll switch to a third party software though we don't want to put any big money into the instrument. Unichrom looked good though I am concerned about having to hook the 7673 up via inet/HPIL. Any other recommendations?

[cleh]

The purge and trap and MS would not have used the EPC connections so that wouldn't be an issue. Try using the GC panel to turn purge A on and off and put your finger on the black solenoid on the injector pneumatics inside the left side of the GC. You should hear a loud click when you turn it on and a softer click when you turn it off and you should feel it move. After that, do the same with valve 3 If you hear something, is it the same as when you turned the purge on and off and do you feel the solenoid move? Let me know what happens and we can then deal with the software.

[MSCHemist]

I'll give that a try when I get back to work but I know for a fact the valve works. I did both splitless and 50:1 split injections last week and it worked fine. The software is just giving me issues. First it was quitting after the first line of each sequence and not saving data so I ended up deleting everything in HPChem and replacing it with archived files (I don't have the disk anymore). That fixed that problem. Then like I said I had trouble getting it off splitless. I'd check the appriate boxes and all of the sudden when I start the method it would do the wrong thing (purge valve open when supposed to be closed and vice versa and I'd check the method and it wouldn't be how I just saved it sometimes).

I'll play with it some more later this week. If it is still giving me issues I'll replace the software (and computer). It'd be nice to get off of that old HP Kyak XA with Penium II and Windows NT anyways. I need to check whether I have the ISA or PCI GPIB card. Or i could but the wires and hook both the 7673 and 5890 up to dual serial ports (RS-232C). my 5890 has the dual option card.

[MSCHemist]

It seems to be working well today. The method is operating the purge valve like it is suposed to and the valve is definately opening and closing like it should. Not sure what changed but sometimes old software (or even new) acts unpredictably.