CR4® - The Engineer's Place for News and Discussion®


Previous in Forum: Current Unbalance in Three-phase Induction Motor Under No Load   Next in Forum: Lifetime/Duration of Circuit Breaker After Short Circuit
Close
Close
Close
16 comments
Member

Join Date: Jun 2018
Posts: 5

Temperature Measurement Circuit Design Without Microcontroller

06/11/2018 11:09 PM

Hello everyone . i'm having a project about : Use thermorcouple to measure temperature without microcontroller(0 - 200 oC) . I researched and i think : " use opamp and icl7107 for display " . But i have 2 problem :
- I can't amplify exactly signal from thermorcouple (because it's very small ~uV) .
- I don't know how to display icl7107 with single ended input . ( i think i designed as datasheet) .

So i hope tobe given some specific guidance from you.

Sorry , 'cause my writing isn't good enough.

Register to Reply
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Comments rated to be Good Answers:

These comments received enough positive ratings to make them "good answers".

Comments rated to be "almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, rate them!
Guru

Join Date: Mar 2007
Location: at the beach in Florida
Posts: 20652
Good Answers: 1203
#1

Re: Temperature measurement circuit design without microcontroller

06/12/2018 12:11 AM
__________________
Life is like riding a bicycle. To keep your balance you must keep moving. A.E.
Register to Reply Score 1 for Good Answer
2
Guru
Engineering Fields - Electrical Engineering - Been there, done that, still doing it. Engineering Fields - Control Engineering - New Member

Join Date: Dec 2008
Location: Long Island NY
Posts: 12782
Good Answers: 798
#2

Re: Temperature Measurement Circuit Design Without Microcontroller

06/12/2018 9:37 AM

First, it looks like you are trying to accomplish all of your gain in one non-inverting amplifier without any frequency control in the amplifier stage or common mode noise rejection. Your single pole passive filter network component values are unclear to me. Also a thermocouple is a very low impedance. I expect the dominant signal you will be reading with this circuit is offset voltage of the operational amplifier and not the thermocouple signal itself.

Look up the topology for a differential amplifier circuit. Better yet look up the topology of an instrumentation amplifier circuit. Many instrumentation amplifiers already come with precision trimmed resistors built in for precise differential amplifier gain and well controlled common mode rejection. With such a low source impedance that a thermocouple has you may still have oscillation concerns, so the mantra of maximum gain in the first stage for low noise may make added problems.

I would still follow the first stage amplifier circuit with an active low pass circuit. My preference would be one of the multiple op-amp topologies (Akerberg-Mossberg, Tow-Thomas) with a corner frequency of maybe 10 HZ. This way any power line signal that still couples in will be attenuated.

__________________
"Don't disturb my circles." translation of Archimedes last words
Register to Reply Good Answer (Score 2)
2
Guru
Engineering Fields - Electrical Engineering - Been there, done that, still doing it. Engineering Fields - Control Engineering - New Member

Join Date: Dec 2008
Location: Long Island NY
Posts: 12782
Good Answers: 798
#3

Re: Temperature Measurement Circuit Design Without Microcontroller

06/12/2018 10:02 AM

Oh, some might consider the icl7107 chip as a fixed program microcontroller. Other than this concern it is a very good part selection.

Also, since a thermocouple actually measures a difference in temperatures by the Seebeck effect, the temperature at one end of the thermocouple wires must be known or a more complicating cold junction compensation circuit can be added. The traditional solution is to put one end of the thermocouple wires in an ice water bath making for a known temperature of 0°C.

__________________
"Don't disturb my circles." translation of Archimedes last words
Register to Reply Good Answer (Score 2)
Guru

Join Date: Apr 2010
Posts: 6299
Good Answers: 665
#4

Re: Temperature Measurement Circuit Design Without Microcontroller

06/12/2018 10:21 AM
Register to Reply Score 1 for Good Answer
Member

Join Date: Jun 2018
Posts: 5
#10
In reply to #4

Re: Temperature Measurement Circuit Design Without Microcontroller

06/14/2018 2:36 AM

thank for the help . i have one question . what op amp we can choose for this amplifier ?.
this is the first time i do it , sorry if my question make u uncomfortable .

Register to Reply
Guru
Engineering Fields - Electrical Engineering - Been there, done that, still doing it. Engineering Fields - Control Engineering - New Member

Join Date: Dec 2008
Location: Long Island NY
Posts: 12782
Good Answers: 798
#12
In reply to #10

Re: Temperature Measurement Circuit Design Without Microcontroller

06/14/2018 9:26 AM

I cannot make your choice, only you can make your choice. (Just a little semantics banter to lighten the mood. )

Since a thermocouple makes two low impedance voltage sources I see no advantage in choosing a FET input op-amp. You don't require the very high input impedance. If you follow my earlier suggestion of building a multiple op-amp circuit in making either an instrumentation amplifier or an active filter then a quad amplifier package will be a good idea because they will likely drift (with their internal temperature changes) together and thus cancel each other out.

__________________
"Don't disturb my circles." translation of Archimedes last words
Register to Reply
Guru

Join Date: Dec 2010
Posts: 1306
Good Answers: 94
#13
In reply to #10

Re: Temperature Measurement Circuit Design Without Microcontroller

06/14/2018 7:16 PM

The type depends on your budget & accuracy requirement e.g.

TLC271 Op amp, typical drift 2 μV/'C but max could be 20. Long term drift 0.1 μV - about $1 each.

INA126 is an instrument amplifier with 3 μV/'C max drift - about $4 each, but balanced differential input & gain set with one resistor.

AD620 similar INA126 has 1 μV/'C drift - about $8 each

OP177 precision OP amp has 0.3 μV/'C drift, also long term drift spec - about $8 each

AD595 has CJC for type K built-in & overall error < 1 'C - about $50 each

Note that 3 μV/'C drift - means +/- 10 'C ambient change effect, 30 μV, is equivalent to 0.75 'C error for type K, 3 'C for platinum.

So TLC271 is OK for showing you can design an opamp & CJC circuit, but OP177 will give real performance if all resistors & volt reference are high stability.

Register to Reply
Guru

Join Date: Jul 2010
Posts: 655
Good Answers: 172
#5

Re: Temperature Measurement Circuit Design Without Microcontroller

06/12/2018 12:35 PM

I can't make out any details of the original diagram, but this sounds like a student's assignment to force recognition that a key concept in thermocouple temperature measurement is that it takes two measurements to get the final result.

The first measurement is the obvious one, the process temperature as measured by the thermocouple. The 2nd measurement is the cold junction or ice point measurement at the junction point where the thermocouple connects to the measuring circuit. This is basic thermocouple thermometry.

Ice point or cold junction (CJ) are typically called 'compensation' because the value compensates for the lack of a 2nd thermocouple in an ice bath.

A CJ measurement is typically done with a thermistor or RTD and the CJ temperature value (above or below the freezing point of water) is added to the thermocouple temperature measurement to get the process temperature.

Register to Reply Score 1 for Good Answer
Guru

Join Date: Dec 2010
Posts: 1306
Good Answers: 94
#6

Re: Temperature Measurement Circuit Design Without Microcontroller

06/12/2018 1:50 PM

This is a thermocouple transmitter from the National Semiconductor data sheet for the LM10 [obsolete] - I send it because I knew it had T/C amp.

There are many operational amps to choose - TLC271 is more modern, single end output -some changes to connections. Cold junction temp compensation will need adjustment according to thermocouple type as will gain according to T/C required temp range and output volts to drive Analog-digital/display.

The LM10 provides a precision supply. Beware of using supply straight from LM117/LM317 and related stabilisers used for op-amp supplies - long term drift is very poor (read the data sheet!).

www.tc.co.uk has good data on thermocouples.

Register to Reply Score 1 for Good Answer
Guru

Join Date: Dec 2010
Posts: 1306
Good Answers: 94
#7
In reply to #6

Re: Temperature Measurement Circuit Design Without Microcontroller

06/12/2018 4:21 PM

A more simple T/C amp from same source.

Connect pin 7 direct to positive supply, not pin 6 output as shown -then a standard op-amp like TLC271 can be used, with pins 1 & 5 unconnected. TLC271 pin 8 can be used to set its current consumption/gain-bandwidth, connecting pin 8 to + supply (as pin 7) will probably suit your application with high gain & low bandwidth - but low output current capacity from pin 6.

Pin 1 gives a precision 2V supply from the 200 mV reference in the LM10 [R5 & R6 give a 10:1 ratio]. You can replace this with a suitable 2.000 V reference or alter resistor values for 2.500 V/5V/10V reference, which are "standard catalog options".

Register to Reply Score 1 for Good Answer
Member

Join Date: Jun 2018
Posts: 5
#15
In reply to #7

Re: Temperature Measurement Circuit Design Without Microcontroller

06/15/2018 12:06 AM

thank you for helping me a lot. Do I have less contact information? I would like to discuss something with you about this circuit.

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1306
Good Answers: 94
#16
In reply to #15

Re: Temperature Measurement Circuit Design Without Microcontroller

06/15/2018 6:47 AM

Thank you for "coming back" and saying replies were usefull. Mostly on CR4, one never hears if the problem was helped or solved!

If you click on the "Member : vukhan... text" at top right of CR4 screen, you will get a menu with "Mailbox" which allows mail direct to other members.

67model

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1306
Good Answers: 94
#8

Re: Temperature Measurement Circuit Design Without Microcontroller

06/13/2018 3:57 AM

On your second question...

What do you understand by "display ICL7107 with single-ended input"?

Do you mean without differential input facility used? Or single supply not + and - supplies?

What does "display" mean to you?

So far as I can see in data sheet, the polarity indication (+)/- is fixed in the IC, also 1999 max display. You cannot get 0-3999 display with single polarity.

Register to Reply Score 1 for Good Answer
Member

Join Date: Jun 2018
Posts: 5
#9
In reply to #8

Re: Temperature Measurement Circuit Design Without Microcontroller

06/14/2018 1:13 AM

i read icl7107's datasheet. i know the max display is 1999 . my problem is : when i connected single ended input to Vin+ , ground to Vin- , the simulation didn't run . But when i use Baterry ( no use thermocouple ) , it run . I don't know where i fail .
sorry for my english .

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1306
Good Answers: 94
#11
In reply to #9

Re: Temperature Measurement Circuit Design Without Microcontroller

06/14/2018 8:49 AM

I suppose you connected a DC source between Vin+ and Vin- with Vin- grounded.

It seems elementary, pardon me! but a T/C will not give a voltage if it is on bench at same temperature as amplifier. Warming T/C in your hand will give about 800 microvolt.

"Simulation" can mean a totally mathematical model on a computer, but I assume you mean a real "hook-up"!!!

You should post a readable circuit of what you are doing and identify supplies & test voltage sources used. Hand sketches can be scanned and posted - I find conversion from .bmp as scanned to black & white type .png files cuts file size drastically.

This may be a case where real equipment does not put its features onto the circuit diagram for you.

DC source may be grounded by connection to a metal case, which is wired to power socket earth. But supply to ICL7107 etc may come from lab power supply which is "floating", insulated, it may carry AC mains voltage due to capacitance coupling.

Lab p.s.u. will certainly have transformer isolation, but transfos usually have mains winding & LV winding one over the other with high C coupling. But if mains neutral (commonly earthed) is connected to "near" end of HV winding there is just a few volts to earth instead of 240V. Many countries have reversible mains plugs so interference can stop/start due to how plug is inserted.

When dealing with very small voltages, using battery powered sources, AC and DC, avoids a lot of such problems. It is also necessary to use screened cables and metal enclosures to keep out unwanted pickup with long runs from thermocouples - even from yourself. When you get the chance, touch the signal input probe of a high Z oscilloscope with a finger - you will usually get 10s of volts AC from mains pickup.

Sometimes, mains pick up can swamp the wanted signal - even rectify on the hidden diodes in integrated circuits (e.g. input to common) and produce a DC error.

Inputs + and minus usually have limits (common mode) on their mean voltage as well as the + to - difference; see data sheet. A typical case is op amps designed to use + and - power supplies where connection of input to either supply cause malfunction - even fatal damage.

Finally,some circuits I have made and used for T/C simulation...

The first is a variable battery source, which can feed a resistive divider to get millivolts. There are plenty of 1.22V precision regulators similar to ZNRE125.

Second is a cold junction simulator, very necessary with a millivolt source to simulate a thermocouple or to make measurement with thermocouples - all tables give e.m.f. relative to 0 Celsius cold junction. This is good enough in room near ambient temperature 20'C.

For accurate result you need close tolerance resistors (+/- 0.01%) and accurate voltmeters and much care (e.g. soldering the 0.01% resistors can cause 0.01% change, so get wire-end types you can clamp in bridge terminals or solder with copper heat shunt pliars and long leads).

You ask what amplifier IC to use. It depends on T/C type and how accurate you want to be. If you look-up the accuracy of cheap T/C type K (also called chromel-alumel or NiCr-NiAl - positive material always first) you will find it is about +/- 3 Celsius at ambient (assuming your cold junction is good) worse at 100'C.

It is not worth making a precision indicator with such a T/C so a " type K" amp with temperature drift of 20 microvolt for use in 20 +/- 5 'C room - about 0.5'C error - is OK. Lots of better multimeters can read T/C.

If you have a platinum T/C, this is expensive but accurate and stable over time. But it has a smaller voltage, 1/4 type K. Usually they are sub-standards to check/calibrate other T/C. So any amp used with it must be very accurate & stable too - or accuracy of platinum is "thrown away".

Of coarse, that means more costly - the amp in the precision platinum T/C transmitter I gave circuit for - LM121A costs $30 or so compared to <1$ for many available (but it is in a hermetic seal metal can to ensure stability - not cheap plastic). It has very low offset voltage and drift - so accurate that its offset voltage is a linear function of temperature and can be used, in that circuit, as cold-junction compensation! For accuracy, everything else in the installation has to be much better too - this may include the material of terminals and extension leads matching the thermocouple.

If you look for LM221 data sheet (lower temperature range version), this is easier to find.

I guess Rixter will give you a different reply, but also usefull.

Register to Reply Score 1 for Good Answer
Member

Join Date: Jun 2018
Posts: 5
#14

Re: Temperature Measurement Circuit Design Without Microcontroller

06/15/2018 12:05 AM

It is hard to find good people, as enthusiastic as you are. The knowledge you teach me is very useful. Thank you very much

Register to Reply
Register to Reply 16 comments
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Comments rated to be Good Answers:

These comments received enough positive ratings to make them "good answers".

Comments rated to be "almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, rate them!
Copy to Clipboard

Users who posted comments:

67model (6); Iris (1); redfred (3); Rixter (1); SolarEagle (1); vukhanh97na (4)

Previous in Forum: Current Unbalance in Three-phase Induction Motor Under No Load   Next in Forum: Lifetime/Duration of Circuit Breaker After Short Circuit

Advertisement