Experiences with HP5890

[freaklabs]

Hi. I'm considering picking up an HP5890 to use on the farm I'm on to teach the local community about GC and food analysis. I don't have much of a budget but am willing to fix things which is why I'm going with the 5890.

Does anyone have experiences bringing up a used 5890 they can share? I'm interested in things to watch out for. I'm also worried about the software used to drive it. Many of the systems I see on eBay do not have the chemstation software. How does Unichrom or Openchrom compare?

Thanks.

[Consumer Products Guy]

The HP 5890 is a great unit. I've been using them since the 5890A which used the old 7671 autosampler which was on tracks and tilted, and an integrator. Later we got 5890 Series II and Windows 3 and ChemStation. Earlier this week I used a 5890 without EPC to assay a product for glycerin and propylene glycol, that computer was using NT 4.0 and ChemStation A.06. Two weeks ago I replaced a heater/sensor in one of the inlet blocks, the first repair on that I had done in over a decade (took less than 90 minutes).

We've retired some 5890 systems in the last few years that were still operationally fine, just had space issues because of arrival of a brand-new system (capital budget replacement, ongoing plan).

So older systems will not have EPC control, not a big deal. These can run with hydrogen carrier too.

[MSCHemist]

Keep in mind the 5890 can be a series I or II. The series I I believe is trickier to hook up to an autosampler. You also will need a computer with a GBPIB or old RS-232 port or there are GPIB-USB adpaters. You may also want an EPC module to have constant gas flow during a nonisothermal GC run. Also keep in mind the 5890 doesn't have a gas saver built in that reduces inlet flow to 15ml/min while maintaining column flow when not actually performing an injection. To achieve a 50:1 split on a typical .25 column you will set the total flow to 50ml/min and it will stay there which will get expensive with helium. I just always operate mine with a 10:1 split so that keeps the total inlet flow at about 10 ml/min

Overall it is a very robust machine. My configuration is 5890 series IIa with EPC and FID. I have the 7673 autosampler with g1512a controller box. It is hooked up to an old HP Kayak computer with an ISA GPIB interface card running Windows 2000 SP4. I Use Chemstation MSD B.02 version. It has an FID detector. I haven't been using it frequently lately. I do most stuff on the GC/MS and have the 5890 with a db-5 and retention gap set for Scoville Heat Units.

I had thought of using it for FAMES but my company doesn't really have much interest in fatty acid profiling. I also considered setting it up with a thick wax column for free fatty acids.

[James_Ball]

For your application I would imagine the non-EPC unit would be great. You seem to be looking more for occasional demonstration type work than constant or routine analysis needing accurate quantitation. The good thing about the non-EPC unit is that you don't have to worry about the EPC solenoid going out, which some of the early EPC units had problems with. We had one of the first EPC units and had to replace that solenoid a few times before they got all the bugs worked out, after that they were really robust units. Purchasing a used one you could not be sure just which one you were getting.

If you will be doing hand injections then I believe UniChrom or OpenChrom would work fine. I imagine they work well with autosamplers also but that just makes configuring them more work. Simple software would work well for your application as long as the GC has the RS232 port installed. Overall the 5890 is one of the best and easiest to use units you will get your hands on, and there are so many used ones you should still be able to find parts for a while even though Agilent doesn't officially support them any longer.

[Yama001]

If Agilent still made 5890s I would buy one in a heartbeat. Getting a used GC to work with a data system can be a real headache, though. Gentech can set you up there, but it will likely be a few grand.

[MSCHemist]

I recall hearing they made an accessory for the 5890 before gas saver was standard on the 6890 called the gas miser. Does anyone know anything about that?

Never mind found it. It just creates two manual setpoints for the total flow controller. You can achieve the same thing tuning down the flow controler or leaving it at 10:1 split like I do.

[segaloco]

They've behaved pretty well for me. The old computer interface is a bit of a pain, because I've grown very attached to ethernet/NIC connections, but otherwise it's a tried and true workhorse of HP/Agilent machines. Definitely if you're getting one off of someone check everything, there are a number of good videos on YouTube on maintenace (albeit on 6890s and 7890s for the ones I've seen) as well as a wealth of info here. I'm sure it'd be a great investment, and, with the right care, a longlasting one as well. Our lab still uses a couple of them alongside 6890s and a 7890, so I can tell you they definitely do have quite a bit of longevity.

[freaklabs]

Hi everyone.
Thanks for all the help. Based on the feedback, I'm considering getting an HP5890 Series II and will probably not need the autosampler initially. A lot of it will be preparing samples by hand and manually injecting them in. The main idea is to teach farmers "advanced" methods of chemical analysis so they can analyze the chemicals in their produce. Its kind of an experiment to make organic food claims more quantitative but I can also see it helping in things like soil, water, or air analysis out here.

I'm a bit concerned about the software though. Is OpenChrom compatible with the HP5890 Series II? Also, the UniChrom demo has a 3-minute retention limit (I think). Will this be enough as a learning system and for chemical analyses of things like pesticides?

The initial system I get will most likely be a learning system and I'm thinking to use OpenChrom or UniChrom on a Windows XP or 7 machine if thats possible. It sounds like the connection would be via GPIB or RS-232 which is fine for me. I guess I'll need to get gas supplies too. Would it be okay to start off with nitrogen and helium tanks? Are any other gases needed?

[segaloco]

Well, the gas all really depends on what you're testing for. So what you're looking for in this analysis is to promote your food as organic quality? In that case I'd assume you'd run quality tests for certain pesticides and herbicides and such. From my reading Helium and Nitrogen will be right up your alley. Another important factor is your detector. Going off of EPA methods 8081, 8141, and 8151, an ECD would be a good place to start for some of the more simplistic stuff. If you need organophosphate pesticides of course you'll need a different configuration as stated in 8141. EPA methods aside, there may be other detector/gas combinations that will better suit your needs, but that was what I gathered on first inspection.

[Steve Reimer]

An ECD is required for low level testing of chlorinated and brominated pesticides but the detectors are radioactive sources so they come with extra paperwork. Most of the commonly used pesticides now are organophosphates, carbamates or others that would need an NP detector.

[James_Ball]

For gasses, I would go with Hydrogen carrier and Nitrogen makeup gas. Helium is quickly becoming very expensive, and if you see you use quite a bit of gas, you can buy generators to make the Hydrogen and Nitrogen you need, along with the Zero Air if you go the Nitrogen Phosphorous Detector route. This would be the best way long term for both safety and cost.

http://www.restek.com/chromatogram/sear ... pesticides

Look down this list for example chromatograms along with the columns and temperature profiles you would be using for pesticides. Restek is a good place to start for information on setting up what you are wanting to do. They also sell parts and supplies for the 5890 if you need them.