scxi 1001/1102/1303

[Fox farmer]

I am having trouble getting accurate thermocouple readings from my 1001/1102/1303 components.  I have 10 channels hooked up to the device.  All values are not accurate and are negative (in the -300 to -400 range.)  Further, when I unplug one thermocouple from its panel jack, all 10 channels respond to the input as if they've all been affected.  I installed the hardware in the from-the-factory configuration in regard to the bias/pull-up resistor networks.  Can someone help me figure out what I'm doing wrong?

[Kyle_A.]

Hello,

When you say the values are negative, do you mean they are in the -300 to -400 degrees C range?  What units are we talking here?  What type of thermocouple are you using?  It sounds like you may be seeing some saturation on your 1102 device, or the DAQ that you are controlling with.  Are you using a SCXI-1600 to control your SCXI chassis, or another PCI DAQ card? 

[Fox farmer]

Hi Kyle,

Yes, the values are all negative.  When outputing in deg F the values are around -450 or so, when in deg C they are around -250 deg C.  It is a J type thermocouple.  I'm using a PCI-6254 card to control the chassis.  When I switch from using the built-in thermistor to a constant cjc value there is improvement but I do not want to have to rely on changing the cjc value manually.  Any help would be much appreciated.

Thanks very much,

Jim

[Kyle_A.]

Hi Jim,

The first thing I would check is to make sure your 6254 card is correctly reading voltages and is not railing on the negative side.  Do you happen to have a CB-68 connector block that you can use?  If not, try disconnecting the SCXI Chassis from the card and then use Test Panels in Measurement and Automation Explorer (MAX) to read from your input channels and see if you can read 0V. 

Once you confirm that your PCI DAQ is in working order, then we can move to your SCXI Chassis.  Make sure your signal connections are secure and your wiring is secure.  Try several different channels on your 1102 and see if you can reproduce the measurement.  Also make sure that you are configuring your task correctly either in MAX or LabVIEW, which ever one you are using. 

Another good test is to read the voltage from a battery (1.5V) or a power supply outputting a steady low voltage and see if you can measure that with your DAQ card.  If you don't have a connector block then try putting this through the 1102 and see if you can read a positive temperature.  Might also be helpful to have a DMM handy to check the voltage coming off your thermocouple leads and verify that the voltage is correct.

[Fox farmer]

Kyle,

I do not have the cb-68 cable.  I did disconnect the cable from the Daq card and opened MAX.  Using pulled up the test panels and was able to look at input on the device.  Most every channel reads to the positive side.  Some ramp up to amplitude (units not shown) and fall off steadily.  Some hold steady at about 3.06 or so.  Should they be reading 0 if no cable/input is connected?

Earlier while I had everying hooked up, I opened MAX and slowed the sampling rate down significantly.  This did allow me to unplug/"open" one thermocouple from its panel jack and only see a response on that channel (the other channels with thermocouples did not "jump").  However, all the values were still in the -250 range using the built-in cjc sensor. 

While using test-panels, I also noticed the device had been factory calibrated at 48 C.  It was also displaying 47 C current temp, which it is not.  Could that be part of the issue?

Thanks,

Jim

[Kyle_A.]

Hi Jim,

When the terminals are open, there is no reference to the voltage you are measuring, and since the terminals are floating it could be anywhere.  To answer your first question, you will not always see 0 when you measure the voltage of open unconnected channels on your DAQ card.

Can you try differentially measuring a test voltage, such as a battery or a power supply?  This will gauge the accuracy of the channels in respect to a known stable voltage.  You can measure this with the 1102 by using an analog input voltage task and setting the measurement to differential.  Also, try several channels so we can confirm that we are reading correctly across the module.  Let me know how this testing goes.

In regards to the temperature calibrated, this is not the ambient temperature but the board temperature, which is why the temperature is so high.  After letting the module warm up, that is where you would see 47 C, which is 1 C off from its temperature that it was calibrated at.  There is no problems there.

[Fox farmer]

Kyle,

I tested the 1102 using a 9V battery.  I hooked up 5 channels to check (0, 10, 19, 25, 31).  I set up a MAX task to look at voltage data on these channels. All 5 channels showed values of 5.xxx even though I didn't have the battery connected yet.  Moving the battery from ch to ch did not seem to show much of a change. 

However, when I pulled up a test panel for each channel and connected the battery I was able to get the following values to show up:

ch 0: 1.05

ch 10: 1.01

ch 19: 1.01

ch 25: .998-1.0

ch 31: 1.0

While checking w/ test panels I was able to detect a drop when touching both leads from a channel.  Any thoughts?

Thanks very much,
Jim

[Fox farmer]

Kyle,

I double checked these channels using the MAX task I had set up.  I was able to see some fluctuation from 5.10xxx to 5.11xx-5.12xx when attaching the battery to the leads.  Also, when I attached the battery to ch0, all the following channels seemed to be affected.  However, when I checked ch10, channel 0 was unaffected but ch 10 and 19 showed a slight jump (to 5.11xxx or so.)  As I moved up the channels the previous channels seemed to be unaffected as I would check the battery. 

Thought more info might help troubleshoot my issue.

Thanks again,
Jim

[Kyle_A.]

Hi Jim,

Here is some information from the SCXI-1303 Installation Guide and Specifications.  When the thermocouple channel is open, there is circuitry in place to detect that channel to be open, and a pull up resistor to +5V rail causes the channel to saturate at 5V.  You can reference this on page 2 Open Thermocouple Detection.  So the open circuit voltage means that the 1303 is properly working.

Do you have any other terminal blocks with you?  Since the 1303 is meant for thermocouples, it might be doing something with the signal to attenuate it or make it rail at 1V since thermocouple measurements are generally low voltage.  It might be a better test to use a terminal block that doesn't account for thermocouples.  Did you specify ±10V input signal range in your tasks? 

I can't seem to find an explanation for seeing 1V off a 9V battery.  Maybe try a different voltage (1.5V battery, power supply, etc.) to have two sets of data to compare to.  If you still see bad results with a different voltage value, and no conclusions can be drawn from that data, then there is a good chance the hardware is broken.  How long has this worked before?  Any ideas what changed to make this card not work?

[Fox farmer]

Kyle,

I'll double check my MAX task configuration to make sure I've set it up correctly.

As far as the hardware goes, this is all new equipment being set up for the first time.  1001 chassis, 2-1102 modules (1303 & 1308 term blocks), 2-1126 modules w/ 1320 term blocks and a 1167 switch controller.  I'm trying to get it set up for the first time so I'm not sure what "should" work and what shouldn't...very frustrating.

I'll keep trying.  Does NI have 24hr call in help?  I'm working on this project in the evenings/weekends and its difficult for me to call in and talk to someone during the day.

Thanks for the help,

Jim