K type - Electronic Circuits Projects Diagrams Free

chandu13 · 27th March 2007, 09:16 AM

hai

Aim using K type thermocouple for temperature measurement its range is 0-1000degress.

0 Degree - 0.000milivolt
1 Degree - 0.039milivolt
500 Degree- 20.640milivolt
1000 Degree - 41.269milivolt

These values taken from K Type thermo couple data sheet.

How to convert this 0 – 41.269mv signal to 0 –5 v.

Please provide me signal condition circuit for K Type thermo couple.

Aim using 89C51 micro controller

Thank you

Regards

chandu

Leftyretro · 27th March 2007, 02:20 PM

Hi chandu;

A good low noise instrumentation op amp with the correct gain could scale the value to your desired range. However be aware that thermocouples are slightly non-linear in response and require compensation to correct their response curve. This is often done in a look up table. Also be aware that thermocouples require cold reference junction compensation to offset the error created by the 'cold' end of the thermocouple, you basically need to know the temp at both ends of the TC. There are special IC chips designed for TC processing, which included all of the above functions, check out Analog Devices web site for one.

Good Luck

gramo · 27th March 2007, 02:46 PM

Quote:

Originally Posted by chandu13

How to convert this 0 – 41.269mv signal to 0 –5 v.

Use an op-amp with a gain of 121.156, that will make an input of 41.269mA = 5V output

For an ideal op-amp, the non-inverting amplifier gain (Vout / Vin) is given by

gramo · 27th March 2007, 02:53 PM

Use a 1M pot for Rf and a 8.2K for R1, that should give you a good range for the gain

ericgibbs · 27th March 2007, 02:55 PM

hi chandu,

I would recommend that you avoid the use of the '-' symbol in your number values, it can be mis-read as a minus variable.

How to convert this 0 – 41.269mv signal to 0 –5 v.

I would suggest a modification to Gramo's 1Meg Resistor, make it two resistors, one fixed 910K and a 200K multiturn,
else the pot will be near end one for a gain of 121

Regards

chandu13 · 29th March 2007, 04:13 AM

hai

Thanks for the reply

Here aim using LM324. Its maximum gain is 100

But we required gain more then 100.

So how to improve the gain of the amplifier.

How to design the amplifier to reduce noise in the signal.

Regards

chnadu

gramo · 29th March 2007, 04:23 AM

Ahh yea. maybe use a dual op-amp, and set the gain to 10 on one 12.187 on the other, the gains would multiply

eg 41.269mV into the 1st amp(with a gain of 10) would give 410.269mV

410.269mV into the 2nd amp (with a gain of 12.187) would have an output of 5V

gramo · 29th March 2007, 04:26 AM

Your LM324 has 4 amps built into it according to the datasheet

Pommie · 29th March 2007, 05:09 AM

A thermocouple is not a battery. You cannot draw any current when taking the measurement. The only way to do this is to output a voltage until current flow is zero. The outputted voltage is your reading. Using an opamp as suggested will result in a very inaccurate reading (if any). Use a thermocouple amplifier IC such as the AD595.

Mike.

Sceadwian · 30th March 2007, 12:29 AM

Actually, a thermocouple is very like a battery, it's impedance is just extremely high, so it would be better said that the current flow must be extremely low. Though settling time might be an issue simply using a capacitor as a buffer for a circuit which requires a current pulse for a reading might be acceptable, though it would require a very stable capacitor.

Leftyretro · 30th March 2007, 01:51 AM

Quote:

Originally Posted by Sceadwian

Actually, a thermocouple is very like a battery, it's impedance is just extremely high, so it would be better said that the current flow must be extremely low. Though settling time might be an issue simply using a capacitor as a buffer for a circuit which requires a current pulse for a reading might be acceptable, though it would require a very stable capacitor.

TC impedeanace is a spec I hardly ever recalled reading. However having troubleshoot and repaired hundreds in a refinery over the several decades I can say that with a ohmmeter a good TC seldom reads over 200 ohms, depending on lead length of course, so I always considered a TC a low impedeance device

But I guess low and high are pretty subjective terms.

dknguyen · 30th March 2007, 02:27 AM

Have you taken into account the cold junction compensation of the thermocouple? A TC only measures the difference in temperature between two points so a voltage reading will only tell you the temperature difference between both ends. In order to get an absolute temperature reading you need to know the temperature at one end at the TC. THis is so called "cold-junction compensation" or you set one end to a known temperature.

Unless you are only interested in the temperature difference, you will need to use an absolute form of temperature measurement on your microcontroller to figure out the absolute temperature you are measuring. You could also use an IC like the max6675 that takes care of the amplification, cold junction compensation and look-up tables for you.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Latest
The type of connector	liupengjian	General Electronics Chat	8	22nd December 2004 06:52 AM
help on inductor design for buck type dc-to-dc converter	markking	General Electronics Chat	15	15th December 2004 03:37 AM
various type of capacitor	yuenkit	Electronic Projects Design/Ideas/Reviews	1	1st December 2004 04:16 PM
Electric motor as a generator, DC type?	Johnson777717	General Electronics Chat	5	23rd June 2004 09:13 PM
Cap Type for 555 timer	Screech	General Electronics Chat	4	12th January 2004 11:32 PM

27th March 2007, 09:16 AM
chandu13 Experienced Member	K type hai Aim using K type thermocouple for temperature measurement its range is 0-1000degress. 0 Degree - 0.000milivolt 1 Degree - 0.039milivolt 500 Degree- 20.640milivolt 1000 Degree - 41.269milivolt These values taken from K Type thermo couple data sheet. How to convert this 0 – 41.269mv signal to 0 –5 v. Please provide me signal condition circuit for K Type thermo couple. Aim using 89C51 micro controller Thank you Regards chandu

27th March 2007, 02:20 PM
Leftyretro Experienced Member	Hi chandu; A good low noise instrumentation op amp with the correct gain could scale the value to your desired range. However be aware that thermocouples are slightly non-linear in response and require compensation to correct their response curve. This is often done in a look up table. Also be aware that thermocouples require cold reference junction compensation to offset the error created by the 'cold' end of the thermocouple, you basically need to know the temp at both ends of the TC. There are special IC chips designed for TC processing, which included all of the above functions, check out Analog Devices web site for one. Good Luck

27th March 2007, 02:53 PM
gramo Experienced Member	Use a 1M pot for Rf and a 8.2K for R1, that should give you a good range for the gain __________________ Spency. How many PC programs are developed with assembler? Then why use it with PIC's? PIC's and interfacing with other devices - a PIC Basic Guide @ digital-diy.net

27th March 2007, 02:55 PM
ericgibbs Experienced Member	hi chandu, I would recommend that you avoid the use of the '-' symbol in your number values, it can be mis-read as a minus variable. How to convert this 0 – 41.269mv signal to 0 –5 v. I would suggest a modification to Gramo's 1Meg Resistor, make it two resistors, one fixed 910K and a 200K multiturn, else the pot will be near end one for a gain of 121 Regards __________________ Eric "Good enough is Perfect" PIC tutorials: Nigel's site: http://www.winpicprog.co.uk/ Gramo's site: http://www.digital-diy.net/ Last edited by ericgibbs : 27th March 2007 at 05:06 PM.

29th March 2007, 04:13 AM
chandu13 Experienced Member	gain hai Thanks for the reply Here aim using LM324. Its maximum gain is 100 But we required gain more then 100. So how to improve the gain of the amplifier. How to design the amplifier to reduce noise in the signal. Regards chnadu

29th March 2007, 04:23 AM
gramo Experienced Member	Ahh yea. maybe use a dual op-amp, and set the gain to 10 on one 12.187 on the other, the gains would multiply eg 41.269mV into the 1st amp(with a gain of 10) would give 410.269mV 410.269mV into the 2nd amp (with a gain of 12.187) would have an output of 5V __________________ Spency. How many PC programs are developed with assembler? Then why use it with PIC's? PIC's and interfacing with other devices - a PIC Basic Guide @ digital-diy.net

29th March 2007, 04:26 AM
gramo Experienced Member	Your LM324 has 4 amps built into it according to the datasheet __________________ Spency. How many PC programs are developed with assembler? Then why use it with PIC's? PIC's and interfacing with other devices - a PIC Basic Guide @ digital-diy.net

29th March 2007, 05:09 AM
Pommie Experienced Member	A thermocouple is not a battery. You cannot draw any current when taking the measurement. The only way to do this is to output a voltage until current flow is zero. The outputted voltage is your reading. Using an opamp as suggested will result in a very inaccurate reading (if any). Use a thermocouple amplifier IC such as the AD595. Mike.

30th March 2007, 12:29 AM
Sceadwian Experienced Member	Actually, a thermocouple is very like a battery, it's impedance is just extremely high, so it would be better said that the current flow must be extremely low. Though settling time might be an issue simply using a capacitor as a buffer for a circuit which requires a current pulse for a reading might be acceptable, though it would require a very stable capacitor. __________________ Curiosity killed the cat; That's why they have nine lives.

30th March 2007, 02:27 AM
dknguyen Experienced Member	Have you taken into account the cold junction compensation of the thermocouple? A TC only measures the difference in temperature between two points so a voltage reading will only tell you the temperature difference between both ends. In order to get an absolute temperature reading you need to know the temperature at one end at the TC. THis is so called "cold-junction compensation" or you set one end to a known temperature. Unless you are only interested in the temperature difference, you will need to use an absolute form of temperature measurement on your microcontroller to figure out the absolute temperature you are measuring. You could also use an IC like the max6675 that takes care of the amplification, cold junction compensation and look-up tables for you.