- http://medusa.sdsu.edu/Robotics/Neuromuscular/Theses/Saksit/Saksit.htm -- master's thesis, i guess.
- used active reference = a sum of the common mode, inverted and driven to the skin. This is because he is using a computer (NI DAQ 6220 A/D card).
- not certain if this is needed with a battery voltage source
- nonetheless, it seems that he got rather good results, good recordings.
- http://fcrar.ucf.edu/papers/tp2_ahabdima_fiu.pdf
- used a 1st order HPF @ 20hz, 2nd order LPF @ 650hz -40db gain @ 3.0khz. In an A/D system, most likely need about the same to prevent aliasing @ the ADC.
- total gain of about 2000. this course document says about the same.
- David Groh's presentation recommends the same - eliminate anything below 20hz, as this is likely motion artifact.
- http://advan.physiology.org/cgi/content/full/31/2/242
- use a INA 128 instrumentation amplifier with a 2 100uf capacitors in the gain-setting feedback loop (so as to be nonpolar)
- use silver electrodes dipped in bleach to generate ions for good conductive contact with the skin.
- input ADC = sound card. :)
- however, this means that any faults in the computers ground connection will be reflected to the user - bad.
- evaluation and design of a small portable EMG amplifier with potential RMS output - 1999
- they designed a discrete instrumentation amplifier, with inherent high-pass filter.
- used a AD736 AC-to-RMS converter.
- EMG amplifier for phd thesis, made by a technician in the same lab
- for recording EMG from locusts! sweet. quote: "To implant the electrodes, Jeff put the locusts in a refrigerator for a while to put them sleep."
- users guide
- used 22MOhm resistors to bleed bias charge from the AD620's input
- student report on a EEG / EMG amplifier project
- a tad bit verbose, but well reported.
- total project budget < $1000 can.
- selected AD73360 16 bit ADC (digi-key: $12),
- 6 channels simultaneous sampling.
- 80mw power.
- 44 lead TQFP, 12x12mm outer dimensions.
|