LM358 Gain

S

Suraj143

Active Member
I need to amplify K-type Thermocouple reading to 0-5V range.This is the range that I need to amplify.

K Type Output Readings
======================
0'C to 250'C

0'C =0.0 mV
1'C =0.039 mV
250'C =10.153 mV

Gain = 492.462

Please suggest me a circuit with commonly available values specially the resisters.
 
As above stated the gain needs to be 492.462.

Adding two op-amps, 1st Amp gain of 10 & 2nd Amp gain of 49.262 would it be possible?
 

Attachments

  • K Type - Gain.JPG
    20 KB · Views: 5,838
I think the above gain seems to be an odd value.What about using those two gains?

Amp-A=100,Amp-B=5 ?? overall gain = 500.

So the outputs comes like this.

1'C=19.5mV
250'C=5.076V

Note:
*Don't worry on cold junction temperature (I don't care & I don't know what it is)
*Don't worry on accuracy issues I just need 1'C resolution.
*Limit to traditional Op Amps LM358,CA3140 etc....Do Not use AD595,MAX6675 etc........
*Also I have seen many commercial thermo controllers they used ordinary OP-amps.
 
Last edited:
Hi Suraj143,

I suggest to use a rail to rail OpAmp for your project.

An LM358 requires a positive supply voltage of 6.5V for an output voltage of 5V not getting to 0V on the output at the lowest input voltage of 0V, which means the power supply must be symmetric.

Check out the MCP4061. That one will do the required amplification in a single stage and be precise with a supply voltage of 5V.


Boncuk
 
None of this makes any sense? For a type K thermocouple while 0 degrees C = 0.00 mV and 250 degrees C = 10.153 mV those numbers are based on the CJC (Cold Junction Compensation) that you have chosen to ignore. Then you mention 1 degree C resolution but why? Since you want to ignore CJC the induced error of CJC not being correctly refrenced will amount to whatever the CJC temperature actually is. That could easly be 20 to 25 degrees C so why worry about 1 degree C resolution? Also while a Type K T/C does have a somewhat linear output region the T/C like any T/C is non-linear. While you may not care about uncertainty the error on what you are proposing will be tremendous.

You can buy temperature transmitters that will do exactly what you are trying to do and provide outputs for a given span of 0 to 5 volts, 0 to 10 volts or even 4 to 20 mA. That or build using the very chips you mentioned not using. Why even bother being concerned with a gain like "Gain = 492.462" when it really won't matter at all?

<EDIT> and if you really want to do this, do as Han's suggest and use a single rail to rail operational amplifier. Han's posted while I was posting. That guy is quick! </EDIT>

Ron
 
Last edited:
Hi Boncuk & Ron thanks for your response.

I searched on net "MCP4061" but there are no device like that!! Is that number correct?

@ Ron

You are correct.My post doesn't make sense that is because I do not know about what is CJC (Cold Junction Compensation). But I read several documents reg CJC but I still couldn't understand what it is.

Is there any rail to rail op amps which is commonly available?

Can somebody tell whats the minimum input offset voltage in LM358?Note that I also have CA3140.
 
Now I slightly understand why the LM358 is not suited.That means it is not rail to rail.Of course my supply is 12V so LM358 output can go for 5V.The problem comes in the lower end of the inputs.That means due to the input offset voltage the LM358 will reject the temperatures until it passes the input offset voltage.

Am I correct?
 
Thanks Mr Sarma you got it.

I also found MCP6042 says it has 3mV input offset voltage.

Found another one
MC33202 it has 8mV input offset voltage.

From the above IC's do I still able to read lower end temperatures?
Because my lowest temperature is 1'C = .039mV.
 
Last edited:
The LM358 is old, inexpensive and has poor spec's today. Its maximum input offset voltage is shown on its datasheet as 7mV at room temperature so the error on the first opamp could be as high as 70mV and the error on the second opamp could be as high as 3.8V.
It is probably too noisy to amplify your very low levels.
 
Here is a little of what I was getting at with CJC. You mention a Type K Thermocouple. So lets look at a Type K T/C:

Nickel-Chromium
vs.
Nickel-Aluminum

LIMITS OF ERROR
(whichever is greater)
Standard: 2.2°C or 0.75% Above 0°C
2.2°C or 2.0% Below 0°C
Special: 1.1°C or 0.4%

We know the junction of the two dissimilar alloys comprise the thermocouple. The Seebeck Effect tells us there will be a millivolt output. However, when we connect the thermocouple alloys (Chromel & Alumel) to our measurement plane we create more junctions. These are called the Cold junctions. If you lay a Type K T/C down on a bench in a room they is 23 degrees C and use a DMM to measure the output you will not see anything. You would expect 919 uV but you will not see that. The cold junctions are subtracting from the hot junction. If the cold junction is maintained at 23 degrees C and you try to measure say 200 degrees C the minimum error you will have is 23 degrees C which is a rather large error.

Ron
 
Thanks Mr Ron now I understood about CJC.Thanks for the explanation.

Today I'll go for stores & check whether that dedicated chips or rail to rail opamps are available.

I'm going to check those.
*MCP6042
*AD595
*MC33202
 
LM358 is a terrible choice. t has about 8mV of input offset voltage and you are trying to resolve signals less than one mV. You need a precision op amp, sometimes called an instrumentation op amp. Analog Devices used to make high precision op amps.
 
Last edited:
Hi,

I have to agree with almost everything i have read here so far

The LM358 will be hard to use because of the input offset. It can be done, but why bother going through the offset adjustment pain in the neck.
The Microchip parts i have tested myself first hand, and found that i could get as little as 50uv input offset with no trouble at all (dead bug technique).
Also, as Ron said you need to add cold junction compensation which so far i have not come across a good explanation on the web anywhere that i did not write myself
You can find some good tables on the web however for K thermocouples.
 
Last edited:
Guys I bought MCP6042 from stores.That's my final solution.I need some guidance building the 500 gain amplifier.Can it amplify at single stage?I read the data sheet but I couldn't find the maximum amplification.
Does it need to configure like a normal OP-Amp?
 
Last edited:
perhaps you can work under instrumentation amplifier mode Fig 4-8 in the combined datasheet
 
Hi sarma it has V1 & V2 but I have only one input.
 

Attachments

  • Instrumental Amplifier.PNG
    27.2 KB · Views: 1,226
Hi Suraj

About a year ago I started this thread. I still have several of the temperature transmitters I mentioned in the thread. They are for type J T/C. I suggest you read through the thread. They cover a range of 1.67 deg C to 371.11 deg C ( 35 to 700 deg F.) and if you could run with Type J I would be happy to try to send you a few for free. The shipping is on me. They are small and I have sent them out in padded mailing envelopes. Never sent overseas other than Canada but this can't be that difficult I would guess. Anyway, if you want a few just send me a PM with a shipping address to send them to. They would simplify everything for you.

Ron
 
You must log in or register to reply here.

Similar threads

O
  • Question Question
PWM Current buffer to drive Proportional Solenoid from PLC
2
Replies
25
Views
8K
Oecist
O
S
unable to make a microphone amplifier circuit using LM358
Replies
12
Views
8K
dr pepper
S
Op-Amp appears to be stable when operating below the minimum gain specified on the datasheet?
Replies
3
Views
2K
ronsimpson
S
B562 Reduce Gain
Replies
4
Views
2K
ronsimpson
Design specific trial on Hearing Aid
2
Replies
35
Views
10K
MD MUBDIUL HASAN
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…