Hello,

Here is a list of rail to rail opamps from TI.
Some opamps even will work down to 2.2 Volts.

Bertus

 

My computer and I have never seen an ods file type before. There are some converting to pdf programs available but they are a nuisance. Bertus, please can you post the list of opamps as a pdf file?

 

hi agu,
This is in PDF format.
The free Libre Office opens 'ods' files, also it has an export option to PDF.
E

 

Here it is the improved version of the circuit.
I just have to wait for the parts to arrive and I'll give it a try.

 

hi LL,
Have you simulated the circuit.?
E

 

hi LL,
Have you simulated the circuit.?
E

Yes... Right now... and it doesn't work anymore at 4.8V... it only works at 3.6V...
The simulator gives an error of convergence.
I am depressed now.

 

hi,
I have created the LTspice sim version, I will try that for you.
Will post the results
Why are you depressed, it will not help you solve a problem.

E

 

hi LL,
This is as your circuit, please check it for accuracy.
I will modify it to suit the required specification, later.
E

 

hi LL,
This is as your circuit, please check it for accuracy.
I will modify it to suit the required specification, later.
E

The first op amp is connected to the temp sensor (1V at 50 degrees Celsius)
Thanks

 

hi LL,
Post the details of the temperature sensor, ie: it's voltage range and the mV/C temp change

E

 

hi,
This shows the Temp function , it lights the LED at 50C.
Also note the Vbty is changing from 4.8v down to 3.6V and the 50C point is still OK.
E

Will look now at the Vbty function..

 

The simulator gives an error of convergence.
I am depressed now.

In real life, all circuits converge!

 

hi LL,
This shows the Low voltage part R7 should be a resistor and pot to enable a trim setting for 3.6v.
On this sim the Vbty falls from 4.8v to 3.5v to show the switch on LED at 3.6v.
E
R5 driving the TL 431 Vref is too low, you dont need 10mA, go for say 2mA, save your battery power.

 

hi LL,
This sim shows the full circuit operation plots.
E

 

I haven't understood what is the problem with my simulation.
Thanks for your simulation.

 

I haven't understood what is the problem with my simulation.

Hi LL,
Look at R7 value, change R7 into series a 75K fixed and a 50k Trim pot, you can trim to the exact 3.6Vbty setting.
E

For convergence, try a different type of low power OPA in your simulation


Update;
Check this link, others have convergence problems with that OPA.
https://www.google.com/search?client=firefox-b-d&q=mcp602+ltspice+model

 

Hi LL,
Look at R7 value, change R7 into series a 75K fixed and a 50k Trim pot, you can trim to the exact 3.6Vbty setting.
E

For convergence, try a different type of low power OPA in your simulation


Update;
Check this link, others have convergence problems with that OPA.
https://www.google.com/search?client=firefox-b-d&q=mcp602+ltspice+model

Interesting!
Any suggestion for an alternative rail to rail op amp easy to find for this little project?
Using LM6132 the simulation works correctly but I've notice that the price of that IC is too high.
Maybe mcp602 in real life will work just fine.

 

Maybe mcp602 in real life will work just fine.

It likely will.
An LMC6484A should also work.

In your simulation, trying changing some of the simulation parameters to see if that eliminates the non-convergence problem.
Try using The Skip Initial operating point solution option (below):



Or try using the Gear Integration method instead of modified trap (below):