Lab Equipment Headaches

[Err]

I don't work in our IT department at work but when a piece of my lab's equipment needs the computer worked on or upgraded, I make sure I'm there because I'm about the only one that's been there long enough to know the ins and outs of the software that makes them function. I'd requested a new computer for our AKTA FPLC (http://www.gelifesciences.com/webapp/wc ... 2/18190026). The one we have is about 12 years old and uses a proprietary software suite called Unicorn. An FPLC is basically an elaborate $60K pump with a UV monitor and a Fraction Collector. We use ours to purify antibodies. This model is very reliable and only needs routine seal replacements and the occasional UV bulb replacement. The computer that was running it was an old XP-32 bit machine. It' had become unreliable and would cause the USB connection between the FPLC and the computer to drop or the random crash. That sucks when you need something to run for 16 to 20 hours at a time. I'd requested a 32 bit machine and got a Lenovo ThinkCentre M series Tiny Desktop (http://shop.lenovo.com/us/en/desktops/t ... ries-tiny/). I can't remember the exact model off the top of my head but as long as it's 32 bit Windows 7, it should work. I've acquired multiple iterations of the Unicorn 5.xx software over the years ranging from 5.01 up to 5.32. I tried installing version 5.22 because it was the version we were running previously with this instrument. After installing the software and rebooting the computer, I turned on the FPLC so it could connect and it flat out refused to connect. I tried multiple times and even tried the 5.32 version that's supposed to support Windows 7. Nothing. There are two services that are supposed to run and they were. The device manager also showed that the FPLC was present. Next we tried installing the software in XP mode but that was also a bust. I then noticed that the Lenovo had no USB 2.0 ports on it. Every port was a 3.0. I went into the Bios and found that we can enable debugging on port 1 to make it a USB 2.0. I tried that and still couldn't get it connect in Windows 7. The IT guy then went and grabbed an older HP Windows 7 -32 bit machine and it worked perfectly. I have never seen software/drivers that are as picky as the stuff GE writes. We spent 4 hours fussing with the Lenovo.

[Executioner]

That brings back memories for me also. We have in the main calibration lab a supermicrometer that can measure down to 0.000030 inches. I upgraded the supermicromter to use a laser instead of relying on the mechanical lead screw. The accuracy improved to 0.000010 inches. The upgrade came with a PC running Windows 98 (with DOS underneath). After 12 years, the pc crapped out. It had an ancient SCSI hard drive. IT was no help, because it was not on the network. So I took it home and was able to copy the files, and rebuild another pc running Windows 98. I tried to run it under 2000, but it would not work. To upgrade, the cost is almost $20k.

[Pugsley]

Welcome to the world of industrial PCs! Our machines are the same way except they need very special motherboards that can mount on top of our real time IO boards and have all the interconnects in the right places. Good old Biscuits PCMs.

At least our new touchscreen has separate IO boards that communicate via canbus and serial so the PC can be a bit more flexible.

[Losbot]

I do have to say though, we have a bunch of those Lenovo tiny ass PCs and they're small powerhouses if you spec them right. Sucks it couldn't resolve your problem but they're really nice and take up almost no space. You can even get the VESA bracket to mount it to the back of your monitor.