LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

SCXI 1102 problems continue!!!!

Our temperatures are still reading 10 - 20 degrees F lower than what it should. But when I read 1 t/c it reads the 10 - 20 deg lower and when I add another t/c to the next channel the first channel drops 2 - 5 degrees and so on. It only does this when I try and read my bearing temps. I tried the same thing with just ambient t/c's but there was no change in temp and they read the correct temp. What could my problem be? I have four SCXI 1102 cards and 4 TBX 1303 terminal blocks all four are doing the same thing.
0 Kudos
Message 1 of 7
(4,103 Views)
Hi Joey,

Have you tried using MAX?

How fast are you scanning?

What range do you have the input amps set too (i.e. +- 10V or +-100mV)?

Ben

Ben
Retired Senior Automation Systems Architect with Data Science Automation LabVIEW Champion Knight of NI and Prepper LinkedIn Profile YouTube Channel
0 Kudos
Message 2 of 7
(4,103 Views)
Also, using the SCXI-1102 along with the MIO card, you should be able to set the input range to +-50mV or +-20mV which MAY be more appropriate if the temp measurement range is <~250 deg F.
(It may take more settling time, though.)

BUT I don't think input range (gain) is the fundamental problem here.

Dave
0 Kudos
Message 4 of 7
(4,103 Views)
You said in an earlier post that you are using type 'J' TCs. Are you using them for all of your TCs? or just some of them, along with some other type? Standard error for the type 'J' TC is about 4 deg F, so the error you are talking about indicates a real problem. SO the question is whether it is
1) in your physical setup, 2) broken hardware, or 3) coding error.

One thing that you have to think about is timing. Part of your problem could be a settling time thing. This can be an issue at any speed, but you sometimes have to be more careful about settling time with a fast IO card. One of the big things is settling time on the amplifier(s) as you change gains (or input ranges).
If you configure the one channel in the scan list and continuously scan it (without any other channels) does it start out at one value and then drift toward a (relatively) stabile value over a few seconds (or millisec)?

The first thing that I would do, however, is a reality check. Forget about the inputs as thermocouples for the moment and simply apply a known voltage (say 1 V DC) into channel 0, for example. Then go read the voltage input value at channel 0.
Do you get 1V reading for 1V in? repeat for all channels.
Read the CJC sensor continuously. Does it
give you a temp close to the environmental temp in the room? (It should be a little bit higher than the room it is located in. It may be notably higher if there is not a reasonable amount of air motion in the room.)

Sometimes a good thing to do is to have another person check your connections, e.g. your TC lead wires in the 1303 term block.

Also, look at all of your TC connections. The connections must be clean. The TC and the lead wires must all be in good conditions, without sharp bends or any kind of kinks.

Do you have any extension wire between the TC and the 1303 term block? If so, it is type 'J' thermocouple wire, right?? If

Beyond that, you should check that your VIs have the conversion of the CJC temp from temp to type 'J' volts before adding it to each of the type 'J' TC voltages. Then make sure that the conversion of volts-to-temp uses the type 'J' as well.

If you want to follow-up the answers to your questions right here by adding your reply as another 'answer'.

Dave
0 Kudos
Message 3 of 7
(4,103 Views)
Dave's method for troubleshooting is excellent. If I could emphasize one of
his points, settling time for very small signals can be important. On the
AI Config VI is an input for the interchannel delay. This is the delay
between reading each channel in the channel list. You might try increasing
the interchannel delay time on this VI.

Mike
0 Kudos
Message 6 of 7
(4,103 Views)
Joey,

You said in an earlier post that you are using type 'J' TCs. Are you using them for all of your TCs? or just
some of them, along with some other type? Standard error for the type 'J' TC is about 4 deg F,
so the error you are talking about indicates a real problem.
SO the question is whether it is:
1) in your physical setup, 2) broken hardware, or 3) coding error.

One thing that you have to think about is timing. Part of your problem could be a settling time thing.
This can be an issue at any speed, but you sometimes have to be more careful about settling time
with a fast IO card. One of the big things is settling time on the amplifier(s) as you change gains (or input ranges).
If you configure the one channel in the scan list and continuously scan it (without any other channels) does it start out at one value and then drift toward a (relatively) stabile value over a few seconds (or millisec)?

The first thing that I would do, however, is a reality check. Forget about the inputs as thermocouples for the
moment and simply apply a known voltage (say 1 V DC) into channel 0, for example. Then go read the
voltage input value at channel 0.
Do you get 1V reading for 1V in? repeat for all channels.
Read the CJC sensor continuously. Does it give you a temp close to the environmental temp in the room?
(It should be a little bit higher than the room it is located in. It may be notably higher if there is not a reasonable
amount of air motion in the room.)

Sometimes a good thing to do is to have another person check your connections, e.g. your
TC lead wires in the 1303 term block.

Also, look at all of your TC connections. The connections must be clean. The TC and the lead wires must all
be in good condition, without sharp bends or any kind of kinks.

Do you have any extension wire between the TC and the 1303 term block?
If so, it IS type 'J' thermocouple wire, right??

Beyond that, you should check that your VIs have the conversion of the CJC temp from temp-to-type 'J' volts
before adding it to each of the type 'J' TC voltages. Then make sure that the conversion of volts-to-temp uses
the type 'J' as well.

If you want to follow-up the answers to your questions right here by replying to this message(or any other including your own). Let us know what you've tried and what the results are, and we may be able to help further.

Dave
"Joey Wood" wrote in message news:50650000000800000027220000-991728092000@quiq.com...
> Our temperatures are still reading 10 - 20 degrees F lower than what
> it should. But when I read 1 t/c it reads the 10 - 20 deg lower and
> when I add another t/c to the next channel the first channel drops 2 -
> 5 degrees and so on. It only does this when I try and read my bearing
> temps. I tried the same thing with just ambient t/c's but there was no
> change in temp and they read the correct temp. What could my problem
> be? I have four SCXI 1102 cards and 4 TBX 1303 terminal blocks all
> four are doing the same thing.
0 Kudos
Message 5 of 7
(4,103 Views)
Hi Joey,

Did you get this working yet?

Ben
Retired Senior Automation Systems Architect with Data Science Automation LabVIEW Champion Knight of NI and Prepper LinkedIn Profile YouTube Channel
0 Kudos
Message 7 of 7
(4,103 Views)