I just bought a USB datalogger from our friends over at omega... =20 designed directly for a single thermocouple into a usb port. I was very=20 impressed with it for the cost ($69). Using it for process profiling=20 (reflow and wave solder operations). They have some really remarkable 'higher end' products for not a lot of=20 money (and of course some for a lot of money) See=20 http://www.omega.com/toc_asp/subsectionSC.asp?subsection=3DC&book=3Ddas&all= =3D1 for=20 the overall category... Personally I'm intrigued by some of the=20 omegabus products such as=20 http://www.omega.com/ppt/pptsc.asp?ref=3DDIN-100&Nav=3Ddasc01 .. I should also mention that I noticed that a few of the thermocouple=20 input devices also will read other sensors in a limited voltage range...=20 may be able to do simple signal conditioning on the front end to bring=20 them into range. I can't address the software functionality and options and whether they=20 match up with your needs or not. I just cared about simple=20 logging/graphing. -forrest On 12/16/2011 10:24 AM, Dwayne Reid wrote: > Good day to all. > > I'm on the hunt for a low-cost data acquisition system / data > logger. Although a stand-alone unit would be nice, I'd be perfectly > happy with something that is tethered to a PC or laptop running > Windows XP or Win 7. > > This is to replace a couple of older systems that run on Windows 98 > and earlier only. Its just getting to be too much hassle keeping > those old Win98 boxes running. > > Ideally, the unit I get would be fairly close to the nicest old unit > that I have - its something called an IDVM (Intelligent Digital Volt > Meter). This is a parallel-port connected system that is > opto-isolated from the PC and has an 18 or 20-bit a/d > convertor. Four of its 8 input channels are selected by latching > relays, allowing those 4 channels to be separated from each other by > several hundred volts. The other 4 channels use an analog mux and > those channels must sit within the common-mode range of the mux. > > What I liked so much about this unit was the software running on the > PC - it allowed direct input of the polynomial coefficients needed > for accurate thermocouple measurement. It also has a cold-junction > temperature sensor right near the input connector. Finally, the unit > had a lab-grade internal voltage reference and extremely accurate > input attenuator resistors. If I recall correctly, the overall error > budget was something on the order of 0.02% or so. This was > astonishing accuracy when I purchased it back in the mid '90s. > > The problem is that this unit came out during the DOS / early Windows > era. Although the software ran just fine under DOS, and then Windows > 98, there is absolutely NO way that it will run on a modern > OS. That's too bad - I really, *REALLY* like this unit. > > > My other unit is a Quatech PCM-12H pcmcia card: 12-bit a/d, 16 > single-ended or 8 differential inputs, with a x1 / x10 / x100 / x1000 > Programmable Gain Amplifier in front of the 12-bit a/d > convertor. Its problem (aside from the incredibly lousy software > that it came with) is that the PGA is just too noisy. It was > designed to allow direct measurement of thermocouples on the x100 or > x1000 gain ranges but the readings are completely unusable at those > gain amounts - there is dozens to hundreds of LSBs worth of random > noise. I was unhappy when I purchased it and after coming back to it > several years later, I'm still unhappy with it. > > The final nail in the pcmcia card's coffin is that, although it was > sold by Superlogics as being supported by Windows XP, the card's > manufacturer (Quatech) has never written XP drivers for it. I just > now finished spending several days fighting with trying to get the > card recognized by XP before tracking down the new owners of Quatech > and getting confirmation that XP was never supported. I didn't find > this out when I first purchased the card because the laptop it was > being used in was running Windows 98se. Did I mention that I think > that the software that came with it is possibly the WORST ever PC > application that I've EVER seen? > > > So: now I'm looking for an inexpensive but reasonable quality data > logging system. My great preference would be to find something akin > to the old IDVM system that I have: > > 1) isolation barrier between the input signal(s) and the computer > that the unit is tethered to. > > 2) slow a/d convertor with lots of bits that can directly read > thermocouple inputs *OR* a 12-to-16 bit convertor with a > decent-quality PGA in front. > > 3) 4 or more inputs. > > 4) programmable attenuator that can handle inputs up to 200Vdc (or more) > > > Nice to have: > > 5) isolation relays on several inputs that allow multi-hundred volt > separation of low-level signals. > > > The sad thing is that I could design and build the hardware myself, > given enough free time. What I can't do, though, is write the > PC-side software. My talents just don't go in that direction. > > > For what its worth, I'm waiting for Dataq's DI-155 > to become > available - I have my name on a waiting list to be notified when it > becomes available sometime this month. > > The DI-155 sells for US $149 and has 8 differential inputs feeding a > 12-bit a/d. Its FS input ranges from plus-minus 2.5V through 50V > (attenuation only). I'd need to add appropriate preamp channels for > reading thermocouples. > > It also lacks isolation from the PC - I'd have to purchase a USB > isolator for when I need the isolation. > > The DI-155 mostly suits my needs but its not immediately > available. Thus my hunt for something that perhaps suits my needs > more closely. > > Finally, I'm not adverse to spending a little extra to purchase > something that can run stand-alone (not tethered to a PC). The main > requirement is for multiple channels, preferably with enough internal > gain to directly read thermocouples. However, it must also be able > to read standard 0-5Vdc inputs as well. Again - programmable > attenuation allowing inputs up to 200Vdc or greater is also appreciated. > > Any suggestions gratefully accepted. > > Many thanks! > > dwayne > --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .