Change Analog Input to Thermocouple

Hello everyone.
I have 8 sets(two load cells paralleled per set -Totalcomp TS50's) connected to a signal amplifier and inputting to a 8 channel input card. We are constantly having to adjust the amplifiers. Does anyone know of an issue replacing the above setup with a 8 channel thermocouple input card and doing away with the signal amplifiers?

What PLC?

For load cells I've used a Beckhoff ethernet interface and load cell module. Beckhoff also has an excitation power supply allowing you to assemble a fully integrated solution on about 7" of din rail.

I've installed over 150 of them and they work great. They are self calibrating and don't need an amplifier.

(note that Beckhoff calls its units kilograms on its configuration register but just make it whatever units you really want, eg, if your load cell is 50 pounds then putting 50000 in the register will get you 0.001 pounds per count.)

I have a SLC 5/04 64k in a 10 slot rack. The current program is moving the current analog value into N7:x where x = 0 through 7 for the 8 sets . I would like to just simply change the card and configure the slot(9) it's in to a 1746-NT8.
Jim

I have no knowledge of the specifics of the cards involved but most thermocouple cards can be configured as mV input. You just have to make sure that the thermocouple cold junction compensation (CJC) is disabled when that happens, otherwise CJ voltage-equivalent will be added to the signal mV.

I would hope that most, if not all manufacturers would disable the CJC when the card is used in mV mode, but with hundreds of models and variations of card models, who knows?

danw said:

I have no knowledge of the specifics of the cards involved but most thermocouple cards can be configured as mV input. You just have to make sure that the thermocouple cold junction compensation (CJC) is disabled when that happens, otherwise CJ voltage-equivalent will be added to the signal mV.

I would hope that most, if not all manufacturers would disable the CJC when the card is used in mV mode, but with hundreds of models and variations of card models, who knows?

Click to expand...

Great advice. Thank you

just something else to thnk about (if you haven't already done so) ...

I might be needless complicating this – and you've probably already considered this angle - but –

we are constantly having to adjust the amplifiers.

Click to expand...

that leads to the question: Why? ...now if the amplifiers themselves are "drifting" and constantly needing readjusting, that's one thing –

but – on the other hand,

if it's your load cells that are responsible for the drift in calibration, then eliminating the amplifiers probably (indeed, almost certainly) will not give you the improved results that you're looking for ...

plus ...

in many industrial situations, a "current" instrumentation signal (4 to 20 mA, etc.) is USUALLY less susceptible to picking up "noise/interference" than a "millivolt" signal ...

so ...

please don't let me talk you out of trying the thermocouple module idea – but personally, I wouldn't rip out and discard the old amplifiers and their wiring, etc., until after the new method had been installed and working correctly for a while ...

I suppose that the big question in my mind (based on what you've posted so far) would be: "What's really causing the problem here?" ... maybe (just MAYBE) you're having to constantly readjust the amplifiers (bless their little hearts) to compensate for a change in some OTHER components in your system ...

so ...

have you fully investigated and found that the load cells, and their mounting, and their power supplies, etc. are all OK – and that it is indeed just the amplifiers that are causing you to have to constantly readjust them? ...

DISCLAIMER: this post is all just guesswork based on what you've posted so far ... we only know what you've told us – and I'm doing a lot of "reading between the lines" here ... feel free to ignore all of this if it doesn't make sense in your particular situation ...

Ron Beaufort said:

I might be needless complicating this – and you've probably already considered this angle - but –



that leads to the question: Why? ...now if the amplifiers themselves are "drifting" and constantly needing readjusting, that's one thing –

but – on the other hand,

if it's your load cells that are responsible for the drift in calibration, then eliminating the amplifiers probably (indeed, almost certainly) will not give you the improved results that you're looking for ...

plus ...

in many industrial situations, a "current" instrumentation signal (4 to 20 mA, etc.) is USUALLY less susceptible to picking up "noise/interference" than a "millivolt" signal ...

so ...

please don't let me talk you out of trying the thermocouple module idea – but personally, I wouldn't rip out and discard the old amplifiers and their wiring, etc., until after the new method had been installed and working correctly for a while ...

I suppose that the big question in my mind (based on what you've posted so far) would be: "What's really causing the problem here?" ... maybe (just MAYBE) you're having to constantly readjust the amplifiers (bless their little hearts) to compensate for a change in some OTHER components in your system ...

so ...

have you fully investigated and found that the load cells, and their mounting, and their power supplies, etc. are all OK – and that it is indeed just the amplifiers that are causing you to have to constantly readjust them? ...

DISCLAIMER: this post is all just guesswork based on what you've posted so far ... we only know what you've told us – and I'm doing a lot of "reading between the lines" here ... feel free to ignore all of this if it doesn't make sense in your particular situation ...

Click to expand...

The power supply for these amplifiers I have not verified as stable. Good thinking on this. I will see if I can establish that it is. After reading your post - I would guess it's not because there are a couple of the amplifiers that we have to adjust less than the others but still from time to time we have to those as well. Sounds like the supply may definitely be at least part of the problem with all of them requiring frequent adjustment.
All of the equipment is old so the amplifiers themselves may be part of it but your thoughts here on the power supply could be a huge time and money saver. Excellent! I will let you know
Jim

All of the equipment is old

Click to expand...

how old are the load cells themselves? ...

how clean/dirty/corrosive is the environment where the load cells are installed? ...

no offense intended - but do you know how a load cell is actually constructed "under the hood"? ... in most models, the strain gauge - the actual working part - is an extremely thin film of metal which gets stretched when the load cell gets bent by the load ... this stretching changes the film's electrical resistance - and it's this tiny change in resistance which actually gets measured and converted into a reading of "pounds" or so on ...

basic idea:

most load cells "feel" pretty hefty and sturdy when you hold them in your hand – but the part (under the hood) that actually makes it work is really pretty flimsy – and it can be easily damaged by corrosion, etc. ...

here's some light reading from the Omega Technical Reference material on Strain Gauges that might be helpful ...

http://www.omega.com/techref/strain-gage.html

I'm going out on a limb here - but everything else being equal (which is seldom the case) smart money says that your strain gauges are causing your problem ... BUT – by all means - check the cheaper and easier-to-change parts first ...

please let us know how this turns out ... it's an interesting problem whose final solution could help others in the future ...

Ron Beaufort said:

how old are the load cells themselves? ...

how clean/dirty/corrosive is the environment where the load cells are installed? ...

no offense intended - but do you know how a load cell is actually constructed "under the hood"? ... in most models, the strain gauge - the actual working part - is an extremely thin film of metal which gets stretched when the load cell gets bent by the load ... this stretching changes the film's electrical resistance - and it's this tiny change in resistance which actually gets measured and converted into a reading of "pounds" or so on ...

basic idea:

most load cells "feel" pretty hefty and sturdy when you hold them in your hand – but the part (under the hood) that actually makes it work is really pretty flimsy – and it can be easily damaged by corrosion, etc. ...

here's some light reading from the Omega Technical Reference material on Strain Gauges that might be helpful ...

http://www.omega.com/techref/strain-gage.html

I'm going out on a limb here - but everything else being equal (which is seldom the case) smart money says that your strain gauges are causing your problem ... BUT – by all means - check the cheaper and easier-to-change parts first ...

please let us know how this turns out ... it's an interesting problem whose final solution could help others in the future ...

Click to expand...

The equipment are fiber feeders that all feed onto a blend line conveyor in the Non-woven textile industry for furniture. All of them are at least 20 plus years old. The Load Cells themselves range from 2 yrs to unknown age. Corrosives is minimal(water sprayed onto the fiber just beneath the weigh pans) but dust is quite a bit. There is also quite a bit of dust in the enclosure where the 8 channel power supply for the analog amplifiers is located.

Temperature can have a surprisingly large effect on load cells. The strain gauge will react to thermal expansion and contraction of the load cell in much the same way it reacts to tiny deflections of the load cell due to an applied weight.

Is your temperature constant?

Have you plotted the need to adjust your load cells to seasonal temperature changes?

Are the ones you need to adjust less often in a more temperature stable location?

Instead of adjusting the amps can you tare the load cell readings electronically (by automatically adjusting gains in your PLC program) daily or between shifts?

TConnolly said:

Temperature can have a surprisingly large effect on load cells. The strain gauge will react to thermal expansion and contraction of the load cell in much the same way it reacts to tiny deflections of the load cell due to an applied weight.

Is your temperature constant?

Have you plotted the need to adjust your load cells to seasonal temperature changes?

Are the ones you need to adjust less often in a more temperature stable location?

Instead of adjusting the amps can you tare the load cell readings electronically (by automatically adjusting gains in your PLC program) daily or between shifts?

Click to expand...

Temperature is pretty constant. It can be quite a bit warmer in summer but our room temp is typically between 65 to 95 degrees Fahrenheit.
The machines are all side by side so they are subject to the same room temperature.

We can calibrate the load cells via the HMI screen and do so daily (plus as needed). This is a bit of a problem for us because the HMI is a PanelView 1000 and I don't have the software to communicate with it yet. The actual values are not being displayed on the screen and the only way we know if it is off is to catch a sample and weigh it. If the sample weight is different than our set point then we check the voltage of the amplifier output and find that it is not in the range we require more times than not. I do not know the manufacturer name of the amplifier cards but there are three points to check.
1st point should read between 1.7 and 4.0 volts and is changed by a small rotary micro switch and is finely tuned by a small screw adjustment resistor(I assume that's what it is)
2nd point should be between .25 and .40 volts finely tuned by a small screw adjustment resistor
3rd is the output voltage of the card with 2 lbs added to the pan and adjusted to 9 volts finely tuned by a small screw adjustment resistor.

This gives us basically a range from 1 to 9 volts for just 32 ounces of material. Our set points are typically around 10 ounces.

I do not have the ability to communicate with the 5/04 yet being DH+ communications but I will soon I hope. When I do I will adjust the program to automatically adjust the gain if empty weight is not achieved after the dump cycle completes if not in range.

I suspect that if and when I change the card to a Thermocoupler then I may need to change the load cells to a 25 lb unit to get a decent resolution in mV change. Our weigh pans do not weigh more than 15 lbs or so.

I suspect that if and when I change the card to a Thermocoupler then I may need to change the load cells to a 25 lb unit to get a decent resolution in mV change.

Click to expand...

If your typical weights are 0 to 32 ounces, it seems you would be decreasing your resolution if you go to 25 lb load cells. Normally you would want your maximum weight to be close to the full resolution of your analog input. If 32 ounce load cells do not produce the max milivolts to your thermocouple module, then that module is not going to give you the best results.

a load cell needs to be zeroed very often,as temperature is big factor.
also calibration is needed like once a day with half of max load
parrallel operation is bad practice
TC card uses cold junction and can not be used her either.

You're using 50lb load cells now. Is the reason for choosing 25lb load cells because the manufacture makes the same size/model cell in a 25lb rating?

There are other load cell manufacturers and models that are a lot closer to what you need.

Why does the holding pan for 10 ounces of material need to be as heavy as 15 lbs? It seems that taring 15 lbs to measure 10 ounces looses you some accuracy of measurement.