Hi, I hope I'm posting in right section. I'm new to arduino and electronics in general, so don't be too harsh if I miss some important info or ask stupid question
I have a copper coil with lot of turns and magnet over it. When magnet moves above it, the coil generates up to 20mV. I want to amplify that voltage so I can read coil voltage oscillations on my arduino analog input. The final idea is that arduino detects when magnet is moving - and that's done by detecting oscillating voltage when magnet moves.
Problem: The voltage from the coil goes through TL061 amplifier, but on it's way out (pin 6) it constantly gives me about 2.5V. I've measured the signal with oscilloscope and before amp it oscillates in range of 20mV, and on pin6(out) of TL061 it's just flat line.
The tutorial from where I took the scheme is using OP27 amplifier, but it says that TL061 should also work.
If anyone can shed some light to why I'm getting constant voltage out of amplifier (instead of osccilating one), I would really appreciate it!
Oh, few more infos: board I'm using is Arduino Uno, and as for the coil, I've tried couple of them and it happens with all of them.
Don't bother with Fritzing -- hand drawn schematics are much better and more informative. To the extent yours can be trusted, the wiring is wrong.
Check the pinout of the op amp by consulting its data sheet and make sure the wiring to it is correct. The pin identification and function need to be shown on a schematic. Also, numeric component values are needed.
You need a 100 nF (0.1 uF) ceramic decoupling capacitor, close to and across the power leads of any IC.
There needs to be a capacitor from the voltage divider to ground (1 uF should work).
The coil should not be connected to ground. It should be connected to the center of the voltage divider.
but on it's way out (pin 6) it constantly gives me about 2.5V.
You (or somebody) may have copied an audio-input amplifier design. It's pretty-standard to "bias" AC signals (such as audio) so you can read the negative half of the AC waveform with the Arduino. Bias also helps with signals that are near-zero when you have an op-amp that's powered from a single supply. (There are "rail-to-rail" op-amps but I'm not sure they go all the way down to exactly-zero without a negative power supply.)
Anyway... 2.5V should correspond to an ADC reading about 512 with no signal and you can simply subtract that out in software.
Or, you may be able to read the voltage without amplification... With the default 5V reference 20mV should read about 4 or with the optional 1.1V reference 20mV should read about 18 or 19. (Of course you can't use the 1.1V reference with the 2.5V biased input.)
Also, Arduino does not accept negative signals at inputs ... if you want to use an opamp (a common-rail one for the single power supply polarity) for amplify the signal, be sure to cut out the negative part ... or you can just merge amplification and rectification, google for "precision rectifier" ... and this app note can give you some more info too ...
That don't mean it's right ... means only that who drawn it was unable to draw a proper schematic ...
You can use EAGLE (the old version 7.7 from cadsoftusa, NOT the actual bullshit that is become after autodesk ruined it totally filling it of bugs and unhelpful things) ... the free version let you draw schematics with almost no limits and make PCB with 2 layers and 80x100mm of component space ...
Of course it isn't right.
The author uses an OP27, which doesn't have rail2rail inputs, on a single 5volt supply,
and seems to tap the signal from the other input of the opamp, resulting in a gain of one, if it works.
One wire of the coil to the analogue input, and the other wire to a 10k:10k voltage divider would work better.
Leo..
As Leo says, the signal from the coil will change polarity; so you need to provide a "mid-point" voltage reference for the signal as shown here. (schematic drawn with TinyCAD)
Adjust the gain by choosing an appropriate value for R4.
Also the OP27 is designed for +-15V rails and will not run from a single 5V supply. I'd recommend the MCP6002 - its a dual op amp so if you need more gain just add another stage.
So for fritzing schematic, I've read in beginner topic not to use it but it was the only one I had, and as I'm new to electronics I wasn't brave enough to draw another one so I don't mess something up haha.
Out of curiosity, why are fritzing schematics bad?
jremington:
Check the pinout of the op amp by consulting its data sheet and make sure the wiring to it is correct. The pin identification and function need to be shown on a schematic. Also, numeric component values are needed.
You need a 100 nF (0.1 uF) ceramic decoupling capacitor, close to and across the power leads of any IC.
There needs to be a capacitor from the voltage divider to ground (1 uF should work).
The coil should not be connected to ground. It should be connected to the center of the voltage divider.
Thanks! I will try to make a schematic with these elements in it and I'll put it here to see if I placed everything correctly.
johnerrington:
As Leo says, the signal from the coil will change polarity; so you need to provide a "mid-point" voltage reference for the signal as shown here. (schematic drawn with TinyCAD)
Adjust the gain by choosing an appropriate value for R4.
Also the OP27 is designed for +-15V rails and will not run from a single 5V supply. I'd recommend the MCP6002 - its a dual op amp so if you need more gain just add another stage.
Thank you for the picture, this really helps a lot!
Quick question, would it work if I put TL061 instead of MCP6002 (if I connect pins correctly)? If I provide too low voltage for amp (for example 5V for 15V amp) what results can I expect on its >>out<< pin?
Don't leave out the two diodes, they are important to protect the op amp!
This is the preferred design, used in lots of amateur-built seismometers (linked in reply #2). Any "rail to rail" op amp is fine, only very low audio frequencies are amplified. I also suggest to increase C5 to at least 4.7 nF as the signals of interest are 0.1 to 10 Hz.
polynome312:
Out of curiosity, why are fritzing schematics bad?
Those images may make good wall art, they're not schematics.
Often the parts used are wrong ("it doesn't have the correct power so I added a 9V battery to power that massive motor which is represented as something totally different"), which causes confusion at best. Chips are represented as package, meaning you have to hunt down the part and its pinout to understand what it's supposed to do. Just a few of the issues...
If I provide too low voltage for amp (for example 5V for 15V amp) what results can I expect on its >>out<< pin?
Voltage on the out pin is always between the highest and lowest supply voltage, so with 0V and 5V to the power pins, the output is between 0V and 5V.
