Thread: [help] home weather monitoring miniproject

The output from some analogue sensors is not enough (or is in the wrong range) for an analogue input. Opamps can be used to invert a signal or shift it into a suitable range for the analogue to digital converter.

Looking at the configuration of the opamp, I think it might just be acting as a buffer. Someone with more experience will be able to correct me on this.

hi,
As the input to the LM358 does not have any type of pre-regulation, the LM358 is not required.
RV1 and R3 will do the same action, ie: give +2.5V to Vref

For greater accuracy and repeatability the Vref should be regulated.
Voltage references for 2.5v are not expensive.

The +2.5Vref sets the SPAN of the adc,, that means as the adc has 8 bit resolution, an LM35 input signal of 2.5V will give a adc value of 255 [FFh].

Here's a 2.5v reference I've used before with analogue temperature sensors. It requires only one capacitor (between Vout and GND) and costs almost nothing. It is in a tiny package too, making it ideal.

MCP1525

Eric, why was the LM358 included in the first place? Perhaps that part of the circuit was stolen from somewhere that wasn't using a uC input?

If you use an 8 bit ADC, you will always get a value in the range 0-255 (the size of an 8 bit register).

If you use a 2.5v reference, 0 will be 0v and 255 will be 2.5v. This is why sensors that "scale" to 2.5v and precision voltage references are so useful.

There are also 4.096v references, which work in an identical way for a 10 bit ADC.

Originally Posted by alkoko111

eric ....
will the ADC 8-bit always has the value of 255. or the value depend on the formula I wrote ?

No the ADC will give a value proprtional to the voltage input from the LM35.

The LM35 in the way that you have it wired will only measure from +2C thru +150C deg.
As the LM35 output is not amplified in your circuit, the LM35 will output from +20mV at +2C thru to +1.5V at +150C

So the very maximum adc conversion [with a 2.5Vref] will be [1.5/2.5] *255 =153 decimal.
I would suggest that the LM358 is used as an amplifier for the LM35 output.

Do you follow OK.?

Originally Posted by alkoko111

i need more explanation .... and thank you for your respond

hi,
Its difficult to know which part you require an explanation for,
if you ask a specific question/s, we can answer.

If you mean the formula: its [+V input to the ADC0804 divided by the Vref voltage] multiplied by the ADC0804 max bit resolution in decimal.
As its 8bit resolution, that a max numeric value of 255

So thats: [LM35/2.5] * 255

At:
2Cdeg[ +20mV],,,, [0.02/2.5]/255 = 2
20Cdeg[ +200mV],,,, [0.2/2.5]/255 = 20
100Cdeg[ +1V],,,, [1/2.5]/255 = 102

hi,
This is your original circuit, edited to show the LM358 as an amplifier for the LM35 output.

Set the gain of the LM358 to give the max output voltage that the ADC will accept. With a Vref of 2.5V that will be 2.5Vmax

Say your max temperature is going to be 50Cdeg, which is 0.5V outfrom the LM35, if you set the gain of the LM358 to *5, that will give a voltage input to the ADC of 2.5V at 50Cdeg.

By using the amplifier it will increase the temperature resolution of the circuit.

Effectively the LM35 will be now 50mV/degC, which is a ADC change of 5 counts,
that means a resolution of of about +/-0.2C per ADC bit.

-About the operational amplifier, on the first configuration it was not working as an amplifier as ERIC said. i search about that with my team and we figured out that, it is working as an insulator buffer
between the input voltage coming from the RV1 and the output voltage of the opamp to the pin9(Verf/2).
The reason is to guarantee that the value RV1 is stable it will not make Vref stable because sometimes, even the pin9 is an input pin and logically, it must not carry any value on it ( no voltage ) its Vref ,
but practically , if u measure the value of it, you will notice that carries some voltage, so this value will included to that coming from RV1which mean, the results will be an expected . Sorry, this part dosnt make any sense
WHAT DO YOU THINK GUYS ? CORRECT ME IF I've MISTAKEN

-Finally, i want to talk about the Vref I applied , as follows :

Vref=1.28 v then we will apply Vref/2=0.64 v on the pin 9 .
according to the formula [LM35/Vref]*255

WHAT WILL BE THE TEMPERATURE ? as eric calculated on 2.5 v

The adc value is [Vinp/Vref] * 255 .. Pin9 is the Vref pin and Pin6 is Vinp

If you set Vref at 1.28V and input 2.5V you will still get the max adc value output of 255.
BUT why would you want to do that.?

Originally Posted by alkoko111

Now ...i have confused .

Firstly, about the LM35. it gives 10mV/Centigrade, is that correct ? if it not so what is it ?


secondly, how could I know the value of Vin ?

Finally, you told of why i would do that ? about the Vref= 0.64v . is this wrong because i choose this value ? really i don't know .....one of my friends suggested that ...... tell me your theory , eric .

hi,
There is nothing to get confused about.

The LM35 outputs 10mV/Cdeg change.
The way you have the LM35 wired on your original circuit enables the LM35 to work over the range of +2C thru +150C.
Thats a Vin [ to the adc] of +20mV thru +1.5V.

I suggested a Vref of +2.5V, this means that a Vin to the adc of +2.5V would give an adc value of 255.

As your project is for a limited temperature range, say +2C thru +50C, it would make sense to amplify the LM35 output so that a tempr of +50C would give a voltage Vin to the adc of +2.5V,,, which would give the max resolution from the adc of 255.

So I suggest you use the LM358 in your original circuit as an amplifier for the LM35.

OK.?

EDIT:
Please tell me what actual temperature range you require and I will post a LM358 amp circuit.