• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 38 subscribers
  • Views 1121 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Noise Analysis EMG Amplification Circuit "S/N Amplitude?"

Chad Knutsen
Prodigy 140 points

Hello,

I'm a mechanical engineering student so take it easy on me.  This is more of a personal project, than school related although I have been working with a couple of my professors.  My grandma suffered a stroke recently and lost the use of her right arm.  All of her rehabilitation has been passive, and so I decided to design a myoelectric prosthesis to give her a more active rehabilitation.  I finished the circuit and it works as expected.  I tried the noise analysis in TINA and I'm not sure if I'm setting it up right.  The frequeny of interest is between 20 and 500Hz, then I chose about 500 samples, and then for the "S/N Amplitude" I entered "1.5m".  I used 1.5m because I designed this circuit to amplify a 1.5mV potential across the muscle.  I inserted several probes after each stage in the circuit and I noticed in the signal to noise graph that in the last stage, the recification stage, introduced a lot of noise.  The SNR is about .7 at 150Hz which means the noise at this point is overpowering the signal.  Is that right?  But then I change the "S/N Amplitude" to 4.5 with the assumption that the voltage at that point will have been amplified to about 4.5V.  The SNR is very high which is good.  Please let me know if I'm setting this up right, or if you need more information.

Click the image to see the circuit and test results.

EMG Amplification Circuit

Thank you, Chad K

  

  • 0
    TI__Mastermind 40295 points

    Chad,

    I'd be glad to take a look at your circuit.

    Can you upload he .TSC file?

    If this approach is okay, some comments/guidance on your overal goals for the circuit as well as the individual blocks would be helpful as well.

    Regards,

    John

  • 0 in reply to John Miller - Sensing
    Prodigy 140 points

    Here you go. My goal is to amplify a 1.5mV potential about 3000 times so it can be usable by the input of an arduino microcontroller.  I want to filter out the RFI before the differential amplification.  I set that at about 2kHz the I want to cut the low frequency motion artifacts at about 10Hz and then LPF again at 500Hz cut off.  Then I want to rectifiy the signal because the arduino can't have negative voltage.  From there the arduino will be programmed to activate a motor located at the elbow joint of the prosthetic brace when the input signal passes a high threshold.  For self calibration, if after say, 30 seconds this event doesn't happen, the threshold is lowered, and it waits another 30 seconds, and repeats until the event happens at which point the current threshold value is used.  The idea is that this is going to be used for rehabilitation, not daily living.  So there should be atttempts to move the arm within those timeframes.  I'm going to use a coreless DC motor to eliminate additional EMI.  Thank you

    RLD.TSCDifferential Amplification Gain 30.TSCEMG Amplifier Circuit Noise Analysis.TSCEMG Amplifier Circuit 2.1.TSCFull Wave Rectify Gain 100.TSCLow Pass Filter.TSC

  • 0 in reply to Chad Knutsen
    Prodigy 140 points

    I'm most interested in knowing if I setup the noise analysis right and what the correct entry would be for the S/N Amplitude.  Anything else that you notice that needs work would be appreciated as well.  Thank you in advance for your help.

  • 0 in reply to Chad Knutsen
    TI__Mastermind 40295 points

    Chad,

    Your circuit looks pretty good.
    The only change I can suggest for your circuit is in the final stage. In its current configuration it isn't amplifying the rectifier output symmetrically, so I added a 100k voltage divider to the "+" terminal of the op amp. This is intended to mirror the effective voltage divider that is formed by the feedback path (R3 & R6 in your circuit); this configuration is frequently referred to as a difference amplifer. In this configuration the output goes from 0V to 6.1V max. The modified TINA schematic is attached.

    It looks like the high output noise in the rectifier circuit is simply due to its high gain.
    Its basically taking the output noise from the LPF and amplifying it.
    According to the TINA noise analysis of your circuit the output noise spectrum of the LPF has about ~100nV/Hz1/2 of output noise at 1Hz.
    The output noise spectrum from the following rectifier and gain stage is about 3.3uV/Hz1/2 of output noise at 1Hz.
    This is equivalent to a noise gain of about ~30x at 1Hz - and this seems to extend to the higher frequencies as well.

    I hope your grandmother's recovery goes well.

    Please let me know if you have any more questions.

    Regards,
    John

     

     

    EMG Amplifier Circuit Noise Analysis_mod.TSC
  • 0 in reply to John Miller - Sensing
    Prodigy 140 points

    Would you be able to tell me what type of diodes I should use?

    That 100k voltage divider made a big difference!  Thank you for that suggestion.   

    Chad K

  • 0 in reply to Chad Knutsen
    TI__Mastermind 40295 points

    Chad,

    I do not know which diodes would be best for your application.

    The apps team that monitors the E2E/Precision Amplifer forum should be able to address that question.

    I see you submitted your inquiry to that forum; they should be able to provdie the information you are looking for.

    Regards.

    John