• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 47 subscribers
  • Views 2901 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Problem with INA122P

rodrigojager
Prodigy 20 points
Other Parts Discussed in Thread: INA122

Hi guys, I've tried to test the circuit in the following picture, and even though it works flawlessly in the simulator, the amplification stage using a INA122P doesn't work as intended whilst mounted in a protoboard.



The gain is not constant, it is falling as the input voltage increase, as you can see from the measured values I've take with a voltmeter:

Input (mV) Output (mV) Gain
2.2 830 377
4.7 1640 348.9
7.2 2520 350
11.4 3340 292.9
15.7 4215 268.5

:

 

 

 

 

The problem is: I need to amplify a diferential signal from 0 mV to 19.5mV to something near 0V to 5V (cause it will pass through and A/D conversion, but it doesn't need to be this exactly range, it can be a lower one). The power suply comes from USB, so I have to use a single suply (5V). I am using a precision resistor of 750 ohm as RG to set the gain (I know that theorically it makes the amplifier to saturate achieving 4.9V while the input is lower than 19.5mV, but since in the simulation it didn't, I trusted that, I can change that later if needed).


I am using a 22pF capacitor between GND and VCC from the USB to make it more stable, and besides that, everything is exactly like in the provided picture (including capacitors and resistors values). The battery represents a voltage source setted in 20mV.

 

I've tried to put the bias input only in positive input, in both inputs, I've tried to pull down both inputs (resistor to GND), but in none of the tries it had a constant gain.

 

Does anyone can help me?

PS: Sorry for bad grammar, english is not my first language.

  • 0
    Guru 21975 points

    Rodrigo,

    The transfer function data that you have provided gives the appearance of instability or oscillation in the amplifier. You say that you have a 22pF capacitor on the Vcc supply. This is much too small. Could you possibly have meant 22uF? This would be one-million times larger value and would be a more appropriate value for this capacitor. I recommend that you use a 0.1uF ceramic capacitor in addition to the 22uF. This ceramic capacitor should be located close to the pins of the INA122.

    If you have further problems, I recommend that you use an oscilloscope to view the output signal. Look for high frequency oscillations. Another possible source of oscillations could be the connections to R8. Be sure this resistor is located close to the INA122.

    I hope this helps,

    Regards, Bruce.

  • 0 in reply to Bruce Trump
    Prodigy 20 points

    Well, this is it Bruce, thanks for your time and help, it is working now, u saved me!

    I've added a 100uF eletrolitic capacitor in parallel with a 100nF polliester one, and the problem is now solved as you can see in the following table (the output instability now is very small and I rather think it's a problem in my enviroment now, like lights, high fans, measurement instrument cable, and of course it is amplified by the gain factor. Anyway it is reasonable now).

    But my final circuit is also a little different from what I've presented you (which I've tried before and wasn't not working without the capacitors - just saying cause if someone else had the same problem, can now discover the solution using the search function). If I take out the input offset, it will stop working again:

    Anyway I've tried this setup before and wasn't not working, and it only works now after this capacitor hint you gave me...

    It is such a simple issue and I've neglected it. I've thought that my power suply was stable enough since it comes from USB and also that the capacitor on the VUSB pin usually needs to be higher than 220pF to that CI even work, but with 22pF the USB of the device is working flawlessly for a long time. I was deceived by this stability.

    Just a last question, how did you got the transfer function from my data? Was using some tool like Matlab's Ident? Cause I've never thought that with this few data we could get a nice transfer function.

    Input (mV)
    		 Output (mV) Gain
    3,7 1000 270,2703
    5,6 1520 271,4286
    7 1900 271,4286
    9,4 2560 272,3404
    10,8 2950 273,1481
    12,9 3490 270,5426
    13 3560 273,8462
    15,1 4140 274,1722
    16,6 4540 273,494
  • 0 in reply to rodrigojager
    Guru 21975 points

    Rodrigo,

    I'm pleased that you are getting good results now. Congratulations.

    I plotted the input and output data points you provided in your first posting in Excel, just making an simple x-y plot. I saw that the points were monotonic (continuously increasing) which indicates that the amplifier was not running into input or output limitations. The points did not make a straight line, or as you stated, the gain is varying. This varying gain (or nonlinearity) is a strong indicator of oscillations.

    Best regards, Bruce.

  • 0 in reply to rodrigojager
    Genius 5430 points

    Rodrigo;

    I would advise against using polyester film capacitors as they are not suitable for bypassing applications at high frequency. Their internal inductance together with the capacitance forms a series- resonant circuit; above resonance, that capacitor no longer looks like a capacitor! A ceramic capacitor has far better characteristics and, as Bruce pointed out, a parallel combination of a large electrolytic and a ceramic capacitor makes a very effective bypass capacitor over a wide frequency range.

    Regards, Neil  ex-Burr-Brown