Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Filter design for thermocouples

Status
Not open for further replies.

simon_12

New Member
Hi all



I want to design a 2nd order low pass active filter circuit for thermocouples , can anyone has a design of this which you have already tried and works well with thermocouples? What resistor and capacitor values you used?

I have studied the NI tutorials for signal conditioning of thermocouples.It is written that the frequency of thermocouple signal is near about 4 Hz. What is the cutoff frequency that you set ?



Please send me the circuit or a link of this. Any help will be greatly appreciated.
 

ke5frf

New Member
The "frequency" of a thermocouple signal???
Thermocouples are not AC devices, purely DC and millivolts at that.
Are you familiar with the Seebek affect concerning dissimilar metals? The closest thing to AC in a thermocouple signal that you might see would be if a measured temperature is fluctuating with some regular frequency.

Having said that, any variety of noise could be introduced to a thermocouple signal. This is especially true in an industrial environment with significant electromagnetic fields from machinery, motors, transformers, etc.

A lot of your noise filtering approach is dependent on the associated circuits. For instance, a recent poster asked about common-mode rejection in an amplifier circuit for an EKG design. I suggested using a common mode choke to help filter noise that would be present at the amplifier inputs. A thermocouple would typically be using a similar differential amplifier circuit. So a similar noise rejection technique might work here. Also, a simple capacitor to ground from the (+) input of your amplifier, maybe 100 uF, would help eliminate noise at or near mains frequencies.

Can you expand further on the 4 Hz frequency that you have read about? Are you confusing EMI with thermocouple behavior?
 

MikeMl

Well-Known Member
Most Helpful Member
The rate of change of a thermocouple (time constant) has to be several seconds, especially when considering the thermal mass of what the thermocouple is attached to. A LPF with a cut-off of 1Hz or so is entirely appropriate.

Download TI's free Filter-Pro cad program and cobble up an active LPF and try it...
 
Last edited:

simon_12

New Member
Thanks Mike for the software

But if anyone has tried the low pass filter for thermocouples practically. Please send me the circuit or any link in which values of R and C are mentioned.:)
 

MikeMl

Well-Known Member
Most Helpful Member
What could be simpler. Put it between the instrumentation amp and the ADC input.

Another example from free software from MicroChip
 

Attachments

  • ChLPF.png
    ChLPF.png
    35.4 KB · Views: 751
  • ChLPFa.png
    ChLPFa.png
    20.3 KB · Views: 515
  • ChLPFb.png
    ChLPFb.png
    29.9 KB · Views: 512
Last edited:

simon_12

New Member
Hi

Which topology of low pass filters do you think is suitable for filtering of thermocouples? either Sallen-Key topology or Multiple feedback topology.
 

MikeMl

Well-Known Member
Most Helpful Member
Would you rather buy two resistors, or three resistors?
 

crutschow

Well-Known Member
Most Helpful Member
Hi

Which topology of low pass filters do you think is suitable for filtering of thermocouples? either Sallen-Key topology or Multiple feedback topology.
For your application either would be fine. The differences are subtle. For a discussion of these see Design and Dimensioning of Active Filters.

Since the Sallen-Key uses an op amp with a gain of 1, an emitter-follower or source-follower transistor can be substituted for the op amp (with some loss in filter performance) if you want to avoid op amps.
 

MikeMl

Well-Known Member
Most Helpful Member
Cool implementation of a three-pole filter using only one opamp...
 

Roff

Well-Known Member
For your application either would be fine. The differences are subtle. For a discussion of these see Design and Dimensioning of Active Filters.

Since the Sallen-Key uses an op amp with a gain of 1, an emitter-follower or source-follower transistor can be substituted for the op amp (with some loss in filter performance) if you want to avoid op amps.
Keep in mind that an emitter follower introduces ≈0.6 - 0.7 volts of offset, which changes about 2mV per degree C.
 

crutschow

Well-Known Member
Most Helpful Member
Keep in mind that an emitter follower introduces ≈0.6 - 0.7 volts of offset, which changes about 2mV per degree C.
Good observation. I forgot that he was filtering a thermocouple signal and DC offset level is critical.
 

simon_12

New Member
Hi Mike

Please tell me what is the meaning of Seed Resistor Value in the Filter pro V2.0 software and how can i select the desired Op-amps?
 

Roff

Well-Known Member
Hi Mike

Please tell me what is the meaning of Seed Resistor Value in the Filter pro V2.0 software and how can i select the desired Op-amps?
From the help file:
R1Seed Value

The Seed Resistor setting allows you to scale the computer-selected resistor values to match the application. Move the cursor to the Seed Resistor field and enter your seed resistor value. The default value of 10 k ? is suggested for most applications.

When the circuit is in a power sensitive environment (battery power, solar power, etc.) the value can be increased to decrease power consumption. Some high speed op amps require lower feedback resistance, so their seed resistor value should be decreased.

Higher resistor values, e.g., 100 k Ohm, can be used with FET- input op amps. At temperatures below about 70°C, DC errors and excess noise due to op amp input bias current are small. Remember, however, that noise due to the resistors is increased by nwhere n is the resistor increase multiplier. Lower resistor values, e.g., 50 Ohm, are a better match for high frequency filters using a wide range of high-speed amplifiers from Texas Instruments.

The seed resistor value is ignored when a capacitor value is entered.
 
Last edited:
Status
Not open for further replies.

Latest threads

Top