Problems with J type TC readings

[LV_Pro]

Hi all,
    Going back to a customer's this morning where I'm having some problems that the answer to isn't leaping out.

My customer, after 3 years of running ok, started having problems with the test system computer  locking up, etc., and decided to move everything over to an new computer running XPpro, etc. and to upgrade from LabVIEW 7 (apparently they had upgraded once as it was originally written in 6.1) to 8.0 .Well after days of fighting with it they called me in (I wasn't available last week, in Austin). Well I removed everything they had and started from scratch. Installed 8.0 and Max 4.0.3. First problem is that on the old machine the MAX traditional virtual channels are all broken (no DAQ card) but the channel configurations (chan #, etc.) seem ok. No way that I have found to export them to be read in on the other machine, but after about 1/2 hr of creating new one on the new machine all seemed ok. Couldn't use LV 8, breaks too much of my original program, they don't want a new program, just their old reliable one back. Uninstalled 8.0 and rolled back to 7.1 and it loads and runs the program except that my TC measurements are all "Whack".  Any thoughts to a tired (didn't plan to stay overnight, have yet another cheap shirt to add to my collection) engineer? Going to move card back to original this morning to confirm everything hardware wise works, then start from there.

Thanks,

[LV_Pro]

Put the DAQ back in the old machine, works ok. Figured how to export the "traditional" configuration, and was able to import on the new machine. But on the new machine my TC readings are 2 orders of magnitude higher than on the original. I have "fixed" this by intorducing a scaling factor of a 1000, but am uneasy not understanding what has changed between. I don't see a scaling factor on the original machine (MAX 3.03) and it seems to work on the new machine (MAX4.03). The DAQ is a 6052E with SCC-TC02 modules between it and the TC's. I wrote and integrated this program over 3 years ago and the customer's engineer that I worked with that has used it for most of that time is no longer here, so there very well could be a "gotcha" that I have forgotten. Just don't want to walk out of here not fully understanding what's changed.

Thanx,

[garretth]

Putnam,

This is strange indeed, especially if the values seem to be off by exactly 1000. With the SCC system there is no selectable gain settings and there is nothing about the virtual channels that I can see causing this. It almost seems that there must be some difference in the application (a scaling factor or custom scale?). Did the application change at all? Were both created from thermocouple channels or could one have been a scaled analog input channel?

[garretth]

Putnam,

The only other thing that I could think of is just to make sure that your SCC chassis and module were correctly configured in MAX. I know you've probably already done this. The only reason I mention it is because the TC02 module has a gain of 100. If the driver doesn't know this, then it might explain why you are seeing the large scaling offset.

[LV_Pro]

Good Morning,
   Well I ended up just configuring the effected channels to have a scale of -1000 to 1000 which "fixed" the problem. Didn't have a lot of time on the system at that point, they had been down for several days before calling me, so I didn't get to determine the ultimate cause of what we were seeing. The application hadn't changed, the hardware (other than the computer) hadn't changed, so it probably was something that I wasn't seeing in the MAX configuration. I did export the "traditional" settings on the original machine, after reinstalling the DAQ card, and was able to import them on the new one, but still, as mentioned earlier, had to scale the TC channels. If I ever get a chance to look at the system again I'll explore it more, but since it ran for approximately 3 years before they called me in it may be a while.