Go Down

Topic: Question about LM358 Dual OP-AMP (Read 279 times) previous topic - next topic

Peet79

Hi, I'm about to buy a sensor which has an output of 7-13mV, with steps of 6uV I believe. Is the LM358 sensitive enough to amplify this voltage? Can I amplify this to the range of arduino, 0-5V? Any tips on how to do this?
My thinking is that I can only amplify it by a certain amount, say 200 times, which would give me a range 1.4-2.6V, meaning I lose resolution? Correct me if i don't understand how amplification works...

Thanks!

jremington

#1
Oct 19, 2018, 05:59 pm Last Edit: Oct 19, 2018, 06:01 pm by jremington
Yes, you can use an op amp to amplify a voltage.

But post a link to the sensor, as a better solution may be possible. It is very difficult for a beginner to build a circuit with microvolt sensitivity.

Peet79

Thanks jremington, it's the attached sensor. It should have a resolution of 0.1% O2.

JohnRob

I think the LM358 is a marginal choice for use with this transducer.  It depends on how precise you would like your measurement.

If I were to use this sensor I would use this board:
Low Noise ADC

For a number of reasons:

1) Is is easy to use compared to building a circuit.

2) This type of conversion  (delta-Sigma) is FAR superior to the Arduino analog input for low noise applications.   It is much slower so it's not as good for higher speed applications.


Hope this helps.

John
Please do not PM me with thread based messages.  If your thoughts are worth responding,  the group should benefit from your insight.

jremington

#4
Oct 19, 2018, 06:54 pm Last Edit: Oct 19, 2018, 06:55 pm by jremington
I don't think you can use that sensor with Arduino, without a lot more information.

It is intended to be used with the companion display/driver unit.

Peet79

Oh damn... So it's not as simple as making the signal arduino-compatible? I mean, it's just a variation in voltage that is being transmitted, it doesn't sound impossible with the right electronics.

And could the Low Noise ADC work? I really don't know what to look for regarding that one, like does it work fine at such low voltages?

jremington

Quote
it's just a variation in voltage that is being transmitted, it doesn't sound impossible with the right electronics.
Correct. What would be the "right electronics"?

According to the data sheet, the sensor includes an NTC thermistor. That means the sensor requires a power source.

Peet79

That's what I'm trying to figure out, I was just replying to your answer that it might be impossible to use with arduino.


JohnRob


[/quote]
And could the Low Noise ADC work? I really don't know what to look for regarding that one, like does it work fine at such low voltages?
Yes it will (but**)

The board I suggested is based on the ADS1232 device which can have an internal gain of up to 128.

So    0.013 * 128 = 1.6V
and  0.007 * 128  = 0.896

The ADS1232 has 24 bit resolution which divides the measurable range into 16 million steps.  You would probably only get 4 million signification steps due to various errors.  Could easily make the measurement per the transducer spec.


** Reading the transducer specification there is no mention of input power and the connector is specified as a two pin device.   So either the transducer needs no input power (like most O2 sensors.)  Or we collectively don't understand the operation of the transducer.


I suggest you contact the company, tell them you are doing a student project and see if they can shed some more light on the interface requirements.





Please do not PM me with thread based messages.  If your thoughts are worth responding,  the group should benefit from your insight.

jremington

#9
Oct 19, 2018, 10:23 pm Last Edit: Oct 19, 2018, 10:24 pm by jremington
Quote
we collectively don't understand the operation of the transducer.
Correct, but the data sheet claims there is a thermistor, which absolutely requires a power source to make a measurement.

Regardless, this is not a project for a beginner.

Peet79

Thanks guys for the help. I think it's time for me to give up, I will have to go with my other solution, the much less accurate one.

hammy

#11
Oct 19, 2018, 11:45 pm Last Edit: Oct 19, 2018, 11:46 pm by hammy
I think the device produces a voltage proportional to partial
Pressure of oxygen 0-21% giving 0-13mV ( oxygen powered battery) is how I read it , NTC for temp compensation ???. The data sheet is not too clear .

If might help to know the project , hopefully not a safety application , there may be other methods .  An op amp should do it , but needs careful selection to get good performance and circuit layout is important too . As said some expertise and/or research needed

Peet79

Thanks hammy for the help, but I think it's true I won't be able to make it work. I'll go with the Grove sensor, I hope I can manage that since there are libraries for it and everything.

MarkT

#13
Oct 20, 2018, 11:50 am Last Edit: Oct 20, 2018, 11:52 am by MarkT
Correct, but the data sheet claims there is a thermistor, which absolutely requires a power source to make a measurement.

Regardless, this is not a project for a beginner.
The power source may be the oxygen it is measuring - ie its like a fuel cell.  There is not enough
information to tell if this is the case, what the biasing and impedance requirements are...

[ The LM358 is not a precision opamp, for these low voltages something with 100uV offset or
so might be a better choice. ]
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

wvmarle

Hi, I'm about to buy a sensor which has an output of 7-13mV, with steps of 6uV I believe. Is the LM358 sensitive enough to amplify this voltage? Can I amplify this to the range of arduino, 0-5V? Any tips on how to do this?
I don't think the LM358 is a suitable OpAmp for this. Get a precision OpAmp, designed for this. You may also have to look at the output impedance of your sensor, it may be very high requiring an OpAmp with suitably high input impedance.

For amplification: don't go for 0-5V, instead you better go for 0-1.1V and use the internal reference. Much more stability, and less amplification needed. It's an absolute reference, which is much better for what I assume is also an absolute output. So total amplification needed is about 80 times.

For extra resolution, you may consider a two-stage amplification, applying an offset to the second state. E.g. a 20x first stage for a 140-260 mV signal, then a 130 mV offset on the second (leaving a bit of margin) and another 7x amplification for 70-910 mV resulting range. That way the 10-bit ADC of the Arduino gets you close to the desired 0.1%.

Another option: use an external ADC like the ADS1115, 256 mV scale, and an 18x or so amplification. That'll definitely give you the 0.1% you're looking for (assuming you can keep the noise under control).
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

Go Up