I cannot figure out how to amplify a very small voltage within the range of 0 to 5 mv and need some guidance. I am trying to read this voltage via the Arduino's analog input. I have a couple of op amps but do not know which one to use or how to use it. The op amps I have are the: LM324,741, TL082,LM339. I do not know how to set the gain of these op amps either and will need some help on choosing which resistors to use. I am pretty new to the Arduino and don't know much of the lingo. Any help would be a awesome, thanks!
Read here
http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/opampvar.html
Want a gain of 100 to 1000.
Ok thank you, but I'm still not understanding how to set the gain. What type of op amp am I going to need to use, noninverting?
You'll need a noninverting Op Amp circuit, as you are looking to turn 0-5mV into 0-5V. More than one, as a single Op Amp stage with a gain of 1000 is going to have a variety of problems.
However... that is a tall order, not really suited to old Op Amps like the LM324. It would also be wise to use a negative supply voltage for the Op Amps, because there is not really any such thing as a true rail-to-rail Op Amp, not when you are talking about inputs in the range of mV.
What is this signal coming from? Perhaps this is something that would benefit from being in a Wheatstone Bridge configuration.
Two free and really good books on Op Amps and how to use them:
Op Amps for Everyone Design Guide Rev B
www.ti.com/lit/an/slod006b/slod006b.pdf?
Handbook Of Operation Amplifier Applications
Operational amplifiers (op amps) | TI.com?
Yes, they are long, but using Op Amps with very small signals is not so simple. There are things like input offset voltage, input bias current, etc.
When amplifying a very small voltage the first step (or stage) of the amplification is not amplification at all but input offset adjustment because you don't want an offset to be amplified:
http://www.play-hookey.com/analog_experiments/basic_circuits/balance_offset.html
The 741 has offset adjustment pins so you could use that for the input stage.
You want the smallest gain in your offset stage so if you use a voltage follower configuration then the gain = 1 (no gain; output=input). see following:
This does two things:
Input offset adjustment
Voltage follower has high input impedance, low output impedance.
Personally I would go with a TL081 because it is a high impedance fet input with offset adjustment but you could get by with a
742 for now.
After input stage is your amplifier stage; (ie: Non-Inverting Op Amp configuration.
To amplify 5mV to 500 mV you need a gain of 100.
If you want to amplify 5 mV to 1.5V you need a gain of 300. Since the gain of a Non-inverting amplifier is 1 + Rf/R2 , you are talking about gain = (300-1)=299
If R2 = 1k , then Rf = (1k)(299)=299k
R2= 1k
Rf= 299k
Since 299k is not a standard value you could use 1k & 300k, 1k & 330k, 10k & 3.3M (this would yield a gain of 1+330=331.
You don't want R2 less than 1k . R2 = 10k is preferable but then Rf will need to be (10k299) or (10k*300).
Once you amplify it to 1.5Vdc you can measure it directly with the arduino.
Just for the record. You don't have to have more than one stage, but it is better if the input offset stage has a low gain, like 10 to 20 and the second stage has the larger gain , like 100, 200, 300 etc. If you wanted to limit the parts count you could skip the voltage follower and use just the one amplifier with a gain of approx. 300. Also, FYI, If you use an 741, unlike the TL082, you can run it off a single ended supply (no negative supply) , which you can't do with the TL082, so you could power it with +5V dc and amplify it to something in the mid to upper range of that (not to exceed 4.5 Vdc).
The LM741 will require a negative supply, it was not designed to be used with a single supply.
http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/a741p.html#c1
OOPS, yes , you are absolutely correct. Thanks Steve for correcting me. I was thinking of something else. I have to stop relying on my memory. Anyway, here are some precision single supply op amps. TLC27L2 , and LT1215. Both are precision, low noise and low input offset. The TLC27L2 will work with supply voltages as low as 3.3V and 5V up to 18V. This one was designed to work at 5V.The LT1215 will work with supply voltages as low as 3V and 5V up to 36V. If these are too exotic (pricey) you can use an LM324.
LT1215.pdf (339 KB)
TLC27L2.pdf (1.44 MB)
LM324A-186021.pdf (174 KB)
LM324-D.PDF (185 KB)
The LM324 is a quad op-amp, so you can use up to 4 stages.
Since you want to start out to just try to get something to try this out for the weekend…….
Connect 5v to Vcc (pin 4 of the DIP)
And ground to pin 11 (check data sheet.)
You will want to use two (or more) resistors to set a gain.
The op-amp will go into saturation if there is any non-zero voltage on the input. So, what you are doing is creating a voltage divider as the feedback. As the output increases, it sends a part of that to the other input.
Your signal goes to the (+) input
Your voltage divider is from the output of that stage to ground.
The op-amp output is the feedback in this circuit the center of the voltage divider goes to the (-) input pin.
Since you probably just want to see ‘something’ you can set the gain by using 1k from the (-) to ground and 300k from the output of that stage to the (-) of that stage.
As raschemmel said, for a gain of 300, you want to have a 299k resistor from the output of that stage, to your 1k resistor. You can use whatever you have to get 300k to 400k to test this to see what happens.
Once you get the output, you can try this out today.
Alas, as raschemmel also stated, that is not the end-all. Op-aps are sensitive and you will see odd changes, so you need to stabilize your input, maybe you might need to filter your input.
Or, if your sensor reads -50degess to 250 degrees, but you only ever want to see between 0 and 125, you can use the other stages to set where you want your zero and your maximum values to be.
Just to offer a couple things that will offer you the excitement of learning and that WTF is going on moment…. Power supply changes and temperature will offer you output changes that are un-related to your input. Running traces too close to PWM traces could effect the output.
Oh, you HAVE to do something with the un-used pins. Simply jump the output of every un-used stage to the (-) input of that stage. Then create a voltage divider of any value and put that into the (+) of un-used stages. One voltage divider for all the unused stages is fine.
As Polymorph pointed out, older op amps are problematic. Seriously, my recommendation for what it is worth is when it comes to precision, low noise, single supply op amps, you are much better off to pay for the latest and greatest to take advantage of the advancements in technology. The TLC27L2 is ideal for your purpose because it is designed for 5V operation. Look at the datasheet.Look at these specs. Someone who knows op amps can't help but say "WOW!" when they read this:
Trimmed Offset Voltage: TLC27L7 . . . 500 µV Max at 25°C, VDD= 5 V
Input Offset Voltage Drift . . . Typically 0.1 µV/Month, Including the First 30 Days
Wide Range of Supply Voltages Over Specified Temperature Range:
0°C to 70°C...3 V to 16 V ?40°C to 85°C...4 V to 16 V ?55°C to 125°C...4 V to 16 V
Single-Supply Operation
Common-Mode Input Voltage Range Extends Below the Negative Rail (C-Suffix, I-Suffix Types)
Ultra-Low Power ...Typically 95 µW at 25°C, VDD= 5 V
Output Voltage Range Includes Negative Rail
High Input Impedance ...10 , ESD-Protection Circuitry
Designed-In Latch-Up immunityThe TLC27L2 and TLC27L7 dual operational amplifiers combine a wide range of input offset voltage grades with low offset voltage drift, high input impedance, extremely low power, and high
It specifically states Vdd=5V at the top of the datasheet. That is very unusual, in the sense that even though it can operate up to 18V , the designers wanted you to know that it was designed with a 5V supply operation in mind. On top of that , it can take a DIFFERENTIAL (+/-) INPUT (LIKE FROM A WHEATSTONE BRIDGE) of up to +/- Vdd (+/- 18V IF Vdd=18V)
Input Offset Voltage Drift . . . Typically 0.1 µV/Month,Including the First 30 Days (That's insane ! , what more could you ask for ?)
So how much do you have to pay for all this ? Try $0.99 !
http://www.mouser.com/ProductDetail/Texas-Instruments/TLC27L2CP/?qs=sGAEpiMZZMtCHixnSjNA6NVXWo9AXLOeAY908%2Fuj8bY%3D
dave-in-nj,
I think the OP'S sensor (that's probably what it is, and yes HE SHOULD HAVE TOLD US THAT) is positive voltage output.
Oh, and Polymorph,
Great suggestion on the books !
Unfortunately we don't have the time or the room to talk about all the things you can do with op amps, but a little trivia on how they came about. The Allied countries had a secret project (like the Manhatten project, but even more secret) after WWI because they learned that they couldn't hit a moving ship accurately from another moving ship miles away. (It is , after all like Rocket science) so they put together a team of scientists to develop a real-time fire-control system for naval ships before the next war broke out (which turned out to be WWII). They realized the technology didn't exist so they designed the op amp (using vacuum tubes) and used them to build a real-time analogue computer to control the guns. I once met a man who was part of that project.
1941
Stewart Miller publishes an article with techniques for high and stable gain with response to dc, introducing “cathode compensation.” Testing of prototype gun director system called the
T10 using feedback amplifiers. This later leads to the M9, a weapon system instrumental in winning WWII.
Patent fi led by Karl D. Swartzel Jr. of Bell Labs for a “Summing Amplifi er,” with a design that could well be the genesis of op amps. Patent not issued until 1946. More : OPAMP History Developing history of Opamps, Vacuum Tube opamps, Solid-State Modular Hybrid Op Amps ic opamps information - Opamps Elektropage.com http://www.elektropage.com/default.asp?tid=176#ixzz2w38ri7Fd
There is a Spice model for the LM358. From the example:
It's interesting to know that the models from National
Semiconductor for the LM158, LM158A, LM258, LM358, LM358A,
LM324 and LM324A have all exactly the same subcircuit.
The LM358 is a dual Op Amp, the LM158 is a single Op Amp.
You've still never told us what is the signal source? What is its impedance? Is that 0-5mV riding on another voltage? What frequency, or is this only slowly changing?
raschemmel is right, there are newer Op Amps with better specs. Use them.
You've still never told us what is the signal source? What is its impedance? Is that 0-5mV riding on another voltage? What frequency, or is this only slowly changing?
raschemmel is right, there are newer Op Amps with better specs. Use them.
I would laugh if it was a 4-20ma transmitter. : )
Thanks for all of the input! I am trying to measure the voltage drop across a 0.01 ohm resistor. I am testing a narrowband o2 sensor. Their are 2 parts to the sensor, a heater element and a sensor element, and their are 4 wires that come out of the sensor. Two wires are the + and - for the heater and the other two wires are for the sensor. Right now I am trying to make sure the heater is working properly by measuring the voltage drop across the .01 ohm resistor. The heater is powered by 13.5 Volts also.
Use the LM386. You set the gain by connecting pins 1 and 8 with a capacitor. Check out RimstarOrg on youtube to see how to build an amp like this. Hope this helps!!!
No, the LM386 is an audio amplifier and is not suited to this at all.
Why should he use an audio amp ? This is an instrumentation application not audio.
That's a Differential Amplifier configuration application , with a subsequent Instrumentation amplifier as a second stage. Look at the links for the op amp books for Differential amplifier. Your inputs will go at either end of the resistor and and + input will go to the + side of the resistor. It is not good to build amps with gains greater than 300, so you will have to cascade any combination of gains with as many stages as necessary to get to your target voltage. You should use one of the two exotic op amps I mentioned. Before you actually build your circuit you need some training. Build some amplifiers using a Potentiometer to control the input voltage and try different combinations of resistors to obtain different gains. Wire up the offset adjust circuit designed for op amps that don't have offset adjust pins because the two op amps I told you about don't have them. Don't worry about offset unless you need it but be prepared to do it if you do. When you can build a two or three stage amplifier with an overall gain of 300 then your ready.
polymorph:
No, the LM386 is an audio amplifier and is not suited to this at all.
raschemmel:
Why should he use an audio amp ? This is an instrumentation application not audio.
It is an op-amp. 
Actually, the LM386 has some characteristics which can be useful - it is designed for input voltages of ±0.4V and automatically sets its output voltage at half of its supply - which could be quite useful for feeding the ADC which is also referenced to Vcc.
Its gain however is pre-set to 20 or 200. As offset can be simply corrected by software, it could quite possibly serve for the application specified as a simple, single-chip solution with no additional components other than bypass capacitors. A gain of 200 across a 0.01 ohm resistor would register 2 volts per Ampere with a resolution of approximately 2.5 mA.
Actually, the LM386 has some characteristics which can be useful
Would it work with dc with no coupling capacitors ?
I don't think it is suitable for this application at all.
I vote for the TLC27L2 or the LM1215