SRI 8610 vs HP 5890 GC - any recommendations?

[lespaul]

Does anyone have any experiences or pros/cons that they can share on using a SRI 8610 GC/FID. We will primarily be using the GC for aroma analysis on food, but just wondering if the SRI was a decent system or am I better off going with a good old (ancient) HP 5890 GC?

The only advantage that I see with using the HP is: (1) its reputation, the HP is a proven workhorse, and (2) if I ever want to add a MS unit I can (though I highly doubt I will ever need to, at least for OUR purposes).

[GOM]

Hi

OK i will jump in first

Most GCs will be similar in performance and we all have our own preferences- at the end of the day it comes down to ease of use, reliability, your local servicing and backup.

I have looked at the SRI 8610 GC/FID on Youtube. I can only say that, from my experience, the 5890 is phenomenally reliable 24/7 for 20 years without fail, easy to use and has plenty of backup in consumables and advice from forum members. I am not sure how much Agilent currently supports it

As for adding an MS - that would not be cheap or easy in either case, but you do say that that is a very low priority

Regards

Ralph

[lespaul]

Thanks Ralph for the input.

My only concern with the HP 5890 was its age and lack of replacements parts/support, but I see that its clearly still being used as a primary workhorse in many labs.

Should I be concerned w/software? That is will it only run on antiquated versions of Windows (eg. Win 3.1, Win 98, etc) or can I use any 3rd party software?

Also, just out of curiosity how much is a MS? Are we talking $10-30k or just a few grand? I know there tons of MS units (quad poles,etc), but just in general?

Again, I plan to run individual standards to determine retention times for select compounds of interest.

[rb6banjo]

I'm running up against this in my lab. I'm in the situation where if the pc that's controlling/collecting-data from my 5890A fails, I will have to antiquate the instrument - even though the GC is working well. It's HPIB communications from the pc to the instrument and the pc's you can get now-a-days can't accommodate the control board. A pc with an ISA bus is a tough thing to find, even on E-Bay or ubid.com. If my pc were to fail, I have a 3rd party data acquisition package that I think I can get to work but it won't communicate with the instrument. I'd have to punch the methods in by hand on the instrument keypad.

[lespaul]

Thanks rb6banjo for sharing this. I hear ya, I been years since I've seen a pc with a ISA board. I may have an old one lying around, but Im hoping I could use a 3rd party software and/or converter for this. Some old pc's can be so bulky.



So I guess having access to a old pc is one of the "hidden" important issues to consider when going with the HP over any other modern GC system.

[aidnai]

Just to chime in on this -- I am using a 5890 series II+ with Chemstation B.03.02 and windows XP. This is running in a virtual machine that is running inside windows 10 with very modern PC hardware. The trick is a $500 USB-GPIB converter from agilent. I bought one on ebay for half price -- it didn't work. Got the real thing from Newark and have been running this way ever since. I have successfully used virtualbox (from Oracle, free) as well as VMWare to provide the virtualization. VMware has given me no issues I can remember. Virtualbox occasionally disconnected the GC and had to be restarted.

I have also tried a demo version of chemstation C.01.05 running on windows 7 (virtualized again) -- also able to control the instrument.

Besides getting rid of the ISA/PCI GPIB issues, it's easy to manage your virtual PC in a way that is safer should you be running a deprecated OS (such as XP). You can also get away from other obsolete hardware (I had the RAM, hard drive, and video card all fail on my XP box over a period of several years, leading me to move to the virtualized solution).

There are a couple tricks to getting it set up, but search the forum and you should find it.

Regards,
Aidan

[KM-USA]

Yeah, the old 82335 HPIB boards will not fit in more-modern computers.

I have some used 82341 and 82350 GPIB boards available reasonably (if you're in US or Canada) which should fit in more-modern computers; post E-mail address and I'll contact you. We went LAN on our 6890 and 1100 systems so these were no longer needed.

[rb6banjo]

Thanks for the tip on the USB-GPIB converter idea. Lesson learned - if you need a gizmo to do something bizarre, look it up and see if someone's built one!

[HippyLabRat]

First off, what's your budget?
I'd say look to see if you can spring for an Agilent 6890 (NOT a Hewlett Packard! This is actually important for hooking it up to a computer). It'll work with the modern versions of Empower and Chemstation, and has that really nice digital control panel in the front that makes troubleshooting much easier.

http://www.labx.com/item/agilent-6890n- ... ew/1140104

I don't work for the company selling this one, but its worth looking into for $13,500 USD. It'll also play MUCH nicer with a Mass Spec upgrade later on, too.

[aidnai]

I bought used from American lab trading, had a decent experience. When the injector zone fan broke and blew up a chip on the mainboard, they were also able to rush a replacement (not warranty but it still kept us up and running).

KM-USA -- while you can certainly still find PCI slots on some modern computers, you can't attach PCI devices to a VM like you can USB devices, meaning you won't be able to run inside a virtual machine. So if you need an old OS (XP or older) now you're stuck again with hardware -- no EFI/UEFI, no windows update, no driver updates etc. The USB converter gives you some good options to get away from old hardware.

[KM-USA]

KM-USA -- while you can certainly still find PCI slots on some modern computers, you can't attach PCI devices to a VM like you can USB devices, meaning you won't be able to run inside a virtual machine. So if you need an old OS (XP or older) now you're stuck again with hardware -- no EFI/UEFI, no windows update, no driver updates etc. The USB converter gives you some good options to get away from old hardware.

Aidnai - you sound like you know what you're talking about. However I don't understand any of the above. But they can't take my Phi Beta Kappa and Magna cum Laude away !!!

[osp001]

I connect to an HP 5890 using a 19242-60030 (RS-232C) board, which feeds into a 19242-60500 cable (simple enough to make yourself, but the "J"-style connector isn't immediately available- I'd have to look up that part). That in turn terminates in a DB-25 connector, and then a crossover connector to a DB-9 female plug. The DB-9 plugs into a serial-to-USB adapter.

There's a crossover connection between the DB-25 and DB-9:

DB-25 ////////// DB-9
2......................2
3......................3
7......................5

Simple enough to breadboard, test it, and make permanent with crimps or a soldering iron.

Not sure if it works with the old HP software, but I use UniChrom and it does fine.

[aidnai]

Aidnai - you sound like you know what you're talking about. However I don't understand any of the above. But they can't take my Phi Beta Kappa and Magna cum Laude away !!!

Sorry! I like to think of myself as a GC chemist, but actually I spend more time on the IT end I think... don't worry about your magna cum laude either...

I guess all I am saying is I am using the USB-GPIB converter (this is the correct model http://www.amazon.com/Agilent-USB-GPIB- ... B00HXBC318) and a 5890 and a modern computer and I like this setup very much.

...
Not sure if it works with the old HP software, but I use UniChrom and it does fine.

HP Chemstation does not interface with GCs via RS232. Looked into this very carefully a few years ago and found it was a dead end. Built the cable and had it communicating with Unichrom as you have done, but Chemstation was a no-go. As far as I know, unichrom was only geared to MS detectors (I have TCD) and so was not suitable for my purposes.

edit: screwed up the quotes

[osp001]

As far as I know, unichrom was only geared to MS detectors (I have TCD) and so was not suitable for my purposes.

Other way 'round. UniChrom isn't for MS, just other detectors. I recommend giving Anton an email if you're interested; he's been very helpful.

[AICMM]

I'll skip the MS part, 'cause if you are getting to the point of needing an MS, you are playing a different game altogether. Having said that, there are TONS of parts out there for 5890's because there are TONS of 5890's out there. It is a true, flexible (thank God) work horse.

As far as data acquisition, if you don't need the full scale of the detector (all trace or all bulk) then you can use any number of third party data systems to collect the data from an HP without much difficulty. If you need the full scale of the FID, then you can look at DataApex or Chemstation (which can be done on XP's without too much work.)

Best regards,

AICMM