reading EMG signal with Open Ephys

[Iryna Yavorska]

Hi,

I'm trying to use EMG signal to measure animal's startle response. First I tried using analog input to record it with my neural data but I found that if low cut off on my AM systems amplifier is less than 10 Hz, it's difficult to rectify it since it wanders a lot and gets cut off below 0 (I'm still not sure what EMG signal should look like but I do see a response in less than 10 Hz freqs, just have hard time measuring it to confirm). I was thinking maybe if I can hijack one of the electrodes on the headstage I might be able to avoid this issue my recording it as an LFP signal. I'm using 32 channel headstage with silicon probes. Does anyone have experience with it? I'm a bit hesitant to start modifying the headstage, since I risk messing it up. I did notice that the headstage has easy access to auxiliary channels but I haven't figured out how I can modify them to help me solve this issue. 

Thank you,

Iryna

[Jakob Voigts]

hi, 

 

I'm trying to use EMG signal to measure animal's startle response. First I tried using analog input to record it with my neural data but I found that if low cut off on my AM systems amplifier is less than 10 Hz, it's difficult to rectify it since it wanders a lot and gets cut off below 0 (I'm still not sure what EMG signal should look like but I do see a response in less than 10 Hz freqs, just have hard time measuring it to confirm).

You can set the analog input to range from around -5 to 5V (this is not 100% precise, so calibrate with a known voltage if you end up using this) by setting the jumpers on the board. That might solve your issue.

 

I was thinking maybe if I can hijack one of the electrodes on the headstage I might be able to avoid this issue my recording it as an LFP signal. I'm using 32 channel headstage with silicon probes. Does anyone have experience with it? I'm a bit hesitant to start modifying the headstage, since I risk messing it up. I did notice that the headstage has easy access to auxiliary channels but I haven't figured out how I can modify them to help me solve this issue. 

Yea thats definitely possible but will be a bit tricky - if i had to do this i'd make an adapter by soldering two omnetics connectors end-to-end to make a straight passtrough and remove a few channels by bending the pins out and then add mill-max pins to these for hijacked aux. in. channels.

jakob