question on uVrms value shown in the GUI

KisiokKisiok
edited July 2018 in OpenBCI_GUI
Hello to everyone. I'm working on a project to study the sensibility of the sensors and i've got stuck in a problem. Once i give a known input (Sine: 1 Hz, 4milliVpp, 0offset) i find in the GUI a sinusoidal signal that respect the frequency value and shows a constant value of 1001microVrms.
The problem is that :
Vmax=sqrt(2)*Vrms=1415 microV
Vpp=2xVmax= 2831 microV = 2.8milliV which is not equal to the input of 4 milliV.
Does anyone know why?

P.S. i am using Cyton.

Comments

  • wjcroftwjcroft Mount Shasta, CA
    The uVrms figures shown on the GUI time series plots are only meant as an approximation. EEG amplitudes are are always recorded as actual microvolts values. Take a look at your recording file and you should see plus and minus 2000 microvolt swings.

    If you are recording from the scalp, there will be a DC offset you need to remove, see,

  • wjcroftwjcroft Mount Shasta, CA
    You can also use a two resistor voltage divider network, if you want to get in the more normal range for EEG, such as say 100 microvolts.
  • KisiokKisiok
    edited July 2018
    THANK YOU VERY MUCH, IT WORKS!!!

    The situation is a little more complex, let me explain. 
    The aim is to study the response of the sensor, given inputs in the range of micro V, both in DC and AC. 
    Firstly i started with DC input, i filtered all the undesired frequency components than plotted them in Matlab, but the values didn't match with the input i gave (in average they were lower). 
    Than i passed to something that i thought easier to understand and work through (Sine: 1 Hz, 4milliVpp, 0offset), and we know were it ended. 
    After your suggestion i tried to give as input this one (Sine: 1 Hz, 9.9microVpp, 0offset), than after filtering i obtained this response (there's even a zoom) :



    that confirmed what i hoped to find.

    I'am going to try the other inputs (DC & AC).

    Thank you again.
  • KisiokKisiok
    normal
    zoom
  • KisiokKisiok
    Here again. 
    I plotted the values but the result is not good at the end. 
    Cyton followed the input only for small values (like the first two). 

  • KisiokKisiok
    in the image : RILEVATI means "recorded"
  • wjcroftwjcroft Mount Shasta, CA
    Are you sure that your voltage driver is calibrated correctly? There may also be issues with proper impedance matching. The ADS1299 is expecting impedances in the typical skin scalp region, around ~ 5K to say 100K ohms impedance.

    Jeremy Frey found the Cyton to match calibration with other research quality EEG amplifiers.

  • KisiokKisiok
    To generate the input i am using a function generator connected to the inverting configuration(Rf= & Rin=) of an op amp(TL081). I will try with a scope to see if the values match. 

    Thank you again.
  • KisiokKisiok
    Rf =10 KΩ the one that connect IN & OUT
    Rin = 6.8 MΩ  
  • KisiokKisiok
    Hi, wjcroft.
    Maybe i got it right but i want to be sure. So, to say: if i use use a voltage divider (consider the resistances roles as in the picture:  Voltage_Divider). 
    R2 should be around ~ 5K to 100K ohms?
  • KisiokKisiok
    Maybe you guys can't answer right now, hope to read from you soon.
    In the meanwhile i'd like to pose you another question, so that when you will answer, i'll be sure and ready on what to do. Do i have to connect the sensors this way: the + signal coming out from the voltage divider connected to the channel pin, and the - signal coming out from the generator (used as reference even by R2) connected to both SRB2 and BIAS?
    (Took inspiration form here: http://openbci.com/forum/index.php?p=/discussion/1322/not-detecting-a-100uv-dc-signal-generator)

    Thank you for your time.
  • wjcroftwjcroft Mount Shasta, CA
    You will need connections on the specific channel, SRB2, and Bias/Ground. I think you can try combining the SRB2 and Bias, since you really only have two outputs from your generator. Suggestion: try contacting the person who posted on that other thread. Click on their name, then on the Message button.

    Regards,

  • KisiokKisiok
    Many thanks, i'll try both. Let me ask one more thing, what about R2's value(2 messages up)?
Sign In or Register to comment.

© Vanilla Keystone Theme 2026