Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
thermocouple noise
mhtl
Member
03-01-2004 04:15 PM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
I am having trouble with noise measuring with a thermocouple.
I have a SCXI-1000 with an 1100 module and a 1303 add-on using Labview 7.0 and MAX 3.0.2. I have no trouble getting a 'reasonably' stable signal in the test panel. However, when I use a simple program running at 500 hz and one channel, I get huge amounts of noise if I touch the thermocouple tip to the frame of the machine everything is mounted on(This happens in Max as well - the average temperature is ok, but the fluctuations are +/- hundreds of degrees C). The machine frame is grounded, and this happens even when the power to the machine is completely off. Since I want to measure the temperature of the frame, this is a bit of a problem.
I have followed the instructions in the SCXI-1
100 manual to wire the negative of the channel I am using to the SCXI ground pin... doesn't seem to help at all.
Is there something in MAX that I should be setting? How does the system know I am using a floating reference and not a ground reference thermocouple?
I have a SCXI-1000 with an 1100 module and a 1303 add-on using Labview 7.0 and MAX 3.0.2. I have no trouble getting a 'reasonably' stable signal in the test panel. However, when I use a simple program running at 500 hz and one channel, I get huge amounts of noise if I touch the thermocouple tip to the frame of the machine everything is mounted on(This happens in Max as well - the average temperature is ok, but the fluctuations are +/- hundreds of degrees C). The machine frame is grounded, and this happens even when the power to the machine is completely off. Since I want to measure the temperature of the frame, this is a bit of a problem.
I have followed the instructions in the SCXI-1
100 manual to wire the negative of the channel I am using to the SCXI ground pin... doesn't seem to help at all.
Is there something in MAX that I should be setting? How does the system know I am using a floating reference and not a ground reference thermocouple?
KyleBryson
Member
03-02-2004 01:44 PM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
Dear mhtl,
I believe that your problem is grounding the signal twice. When you attach the thermocouple to the grounded machine frame, you are grounding one side of the thermocouple through the frame. You are correct in making sure that your thermocouple is grounded, but if it is grounded more than once, a ground loop can introduce noise. Even a few mV of noise can cause tens or hundreds of degrees difference, as you are seeing.
Make sure that you are not shorting both leads of the thermocouple when you attach it to the frame. If you are attaching the thermocouple such that one of the leads electrically connects to ground through the frame, consider the
thermocouple to be grounded, and do not connect the SCXI-1100 CH� to ground in the terminal block (or use the internal W1 jumper).
If this does not solve your problem, please let me know the results from trying without the ground connection in the terminal block and we can continue troubleshooting from there.
Thank you,
Kyle Bryson
National Instruments
I believe that your problem is grounding the signal twice. When you attach the thermocouple to the grounded machine frame, you are grounding one side of the thermocouple through the frame. You are correct in making sure that your thermocouple is grounded, but if it is grounded more than once, a ground loop can introduce noise. Even a few mV of noise can cause tens or hundreds of degrees difference, as you are seeing.
Make sure that you are not shorting both leads of the thermocouple when you attach it to the frame. If you are attaching the thermocouple such that one of the leads electrically connects to ground through the frame, consider the
thermocouple to be grounded, and do not connect the SCXI-1100 CH� to ground in the terminal block (or use the internal W1 jumper).
If this does not solve your problem, please let me know the results from trying without the ground connection in the terminal block and we can continue troubleshooting from there.
Thank you,
Kyle Bryson
National Instruments
03-03-2004 11:52 AM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
Kyle,
Actually, I had already been at that point before... I had previously not been grounding the thermocouples in the terminal block and was having problems with noise, so I first tried using the W1 internal jumper, and then added the ground in the terminal block.
It is true that our thermocouple bead is completely bare and unshielded, so we are definitely making contact with the machine frame at the bead as well.
If I 'shield' the thermocouople with a piece of teflon tape, the noise goes away. However, I wanted to mount the bare thermocouple bead directly on some of the steel parts of the machine frame - which will be grounded - in order to improve the thermal response time of the thermocouple since I am measuring a high speed process.
Actually, I had already been at that point before... I had previously not been grounding the thermocouples in the terminal block and was having problems with noise, so I first tried using the W1 internal jumper, and then added the ground in the terminal block.
It is true that our thermocouple bead is completely bare and unshielded, so we are definitely making contact with the machine frame at the bead as well.
If I 'shield' the thermocouople with a piece of teflon tape, the noise goes away. However, I wanted to mount the bare thermocouple bead directly on some of the steel parts of the machine frame - which will be grounded - in order to improve the thermal response time of the thermocouple since I am measuring a high speed process.
KyleBryson
Member
03-03-2004 01:53 PM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
Dear mhtl,
It seems possible that the machine frame is actually conducting the noise to the thermocouple. Have you tried measuring the potential of the frame with an oscilloscope? It's possible that the frame is picking up noise from other devices or the atmosphere, and conducting that to the thermocouple when they are connected.
If there is actual noise present (instead of being a measurement problem), try using the filtering on the 1100. The 4 Hz filtering will eliminate most noise sources, esp. the 60 Hz noise that is a likely problem on a grounded frame. Of course, if you are trying to scan multiple channels with an 1100, the 4 Hz filter will limit your scanning speed pretty significantly. Averaging a number of points is also a common
method used in thermocouple measurement systems. Both filtering and averaging may be a problem depending on how fast your process actually is, so this may not be too helpful.
The thermocouple bead should be made of two metals connected together; make sure that only one (preferably the negative side) contacts the frame. If you are constantly or intermittently shorting both sides to the frame, that would certainly cause very inaccurate measurements.
Let me know if there is still a problem with the measurement, or if you need help with setting up filtering or anything.
Thank you,
Kyle Bryson
National Instruments
It seems possible that the machine frame is actually conducting the noise to the thermocouple. Have you tried measuring the potential of the frame with an oscilloscope? It's possible that the frame is picking up noise from other devices or the atmosphere, and conducting that to the thermocouple when they are connected.
If there is actual noise present (instead of being a measurement problem), try using the filtering on the 1100. The 4 Hz filtering will eliminate most noise sources, esp. the 60 Hz noise that is a likely problem on a grounded frame. Of course, if you are trying to scan multiple channels with an 1100, the 4 Hz filter will limit your scanning speed pretty significantly. Averaging a number of points is also a common
method used in thermocouple measurement systems. Both filtering and averaging may be a problem depending on how fast your process actually is, so this may not be too helpful.
The thermocouple bead should be made of two metals connected together; make sure that only one (preferably the negative side) contacts the frame. If you are constantly or intermittently shorting both sides to the frame, that would certainly cause very inaccurate measurements.
Let me know if there is still a problem with the measurement, or if you need help with setting up filtering or anything.
Thank you,
Kyle Bryson
National Instruments
