LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

How does the SCB-68 read a thermocouple?

Question: I am trying to use the SCB-68 to read thermocouples directly with no signal conditioning. I am aware this is not ideal, but i have done it in the past with excellent results.
 
My problem now is environment, i have several k type thermocouples installed on a very high power induction motor. Its source is 480v and is being fed by a very powerfull and noisy pwm square wave drive.
 
I am trying to use 8 differential channels, The  TC's work fine in air, they also work fine when touching the ground potential of the scb-68. When the motor is powered, the TC's go to an extremely low reading (-30000)
 
I have tried the TC's with a ground wire grounding the rig to AIgnd, AIsense, and the chassis of the scb 68. nothing has any effect.
 
Obviously the TC's are grounded to the motor, it is what they are connected to.
 
I am guessing the ground potential changes once the motor is turned on causing havoc with the delicate micro volt temperature readings.
 
Is there anything i can do to fix this issue without external signal conditioning?
 
Also how does the SCB-68 actually read the tc? is it applying a slight poitive voltage to ALL input terminals and expecting current to flow back thru ground? Or is the positive terminal of each tc positive and the current flows thru the loop of the TC itself without any ground at all?

Message Edited by Vr6Fidelity on 01-12-2006 12:47 PM

0 Kudos
Message 1 of 5
(3,580 Views)
1. The SCB-68 is just a terminal block; it does not measure anything. The measurement is made on the board connected to the SCB-68.

2. Computer-based data acquisistion devices and 480 V motors with solid state speed controls do not "play nice," as the kids say. The connections you are trying to make may well end up with a fried data acquisition board, fried computer, and MOST IMPORTANTLY a fried USER!

Ground loop problems may produce noisy measurements or may destroy all the low current devices (like the computer) in the system.

Signal conditioners are not only to reduce noise but also to provide safe isolation between the high voltage and high power parts of the system and the more sensitive parts.

A thermocouple is a voltage generating device. The measurement system measures that voltage (typically millivolts) and independently measures the temperature of the "cold junction" of the thermocouple. The temperature at the hot junction of the thermocouple is calculated as Temperature(TC voltage difference) - Temperature(cold junction). If currents flow through the TC wires, due to ground potential differences between the motor frame and the computer ground for example, the voltage drop from Ohm's law is added to the TC voltage and will produce temperature errors. TC wire may have significantly higher resistance than wires used for most elelctrical signals so the interfering signal may be greater than the desired signal.

Lynn
Message 2 of 5
(3,560 Views)

30mv is the max that goes to the terminal block. shouldnt smoke anything but my patience.

Im well aware the SCB-68 is just a terminal block.

Any advice from the pros out there how to do this?

I have tucked tail and returned to the trusty chart recorder for now. but data reduction is qualatative not quantatative as i would like. 

0 Kudos
Message 3 of 5
(3,544 Views)
That 480V makes me a little nervous.  What type of thermocouples are you using?  I know that the temperature range for type-K thermocouples is ~30mV.  Have you stuck a voltmeter on the TC leads while the motor was on?  TC wire is still a wire and it will conduct any old voltage it comes across.  I have used TC wire as a temporary patch wire when I could not find anything else.

You may need a little bias resistance.  My test rig was giving me fits before I broke down and rewired/soldered everything. 

0 Kudos
Message 4 of 5
(3,532 Views)
I think this is a common mode voltage problem.  Although your thermocouple reads 30mV, that 30mV may be riding on top of some other higher voltage.  In noisy environments the only way to accurately measure a thermocouple is to use isolated inputs.  That means that the inputs are truly isolated from each other, and in most all DAQ cards, they are not.
0 Kudos
Message 5 of 5
(3,519 Views)