Biosensor Array Galvanic Skin Response

From Pumping Station One
Jump to navigation Jump to search
This page is part of the Biosensor Array Project

Team Members

Primarily researched and prototyped by Drew Fustini after handover from Phil Strong. Please check with Drew for any clarification.

Prototype

Primary source of prototype is: Truth Wristband

Sean Montgomery, owner of the above site, has a GSR circuit that uses no micrcontroller and a low & high pass filter in the circuit to create a 0.48Hz to 4.8Hz band pass.

Schematic: http://produceconsumerobot.com/truth/content/GSR_schematic_04.png

This circuit was created by Sean for an earlier version of his Truth Wristband project. This circuit is used at as the biosensor array GSR prototype as it produces a simple analog output that is already filtered. This reduces the complexity of the Arduino code and the graphing display application on the computer.

The circuit is featured in Make Magazine Vol 26 (Spring 2011) in the biosensor article written by Sean.

Make Magazine is now selling a kit on Maker Shed for this circuit:

http://blog.makezine.com/archive/2011/04/new-in-the-maker-shed-galvanic-skin-response-kit.html

A video of the assembly and operation is on that page as well.


Truth Wristband

The GGHC biosensor GSR sensor is based on a earlier version of Sean's work. His latest GSR is the Truth Wristband v5. This circuit has just a low pass filter with the high pass filter done in software on the PIC.

Measuring Actual Skin Conductance

Some projects like the Cornell EDA project used a Wheatstone bridge and op-amp (configured as difference amp). With this circuit and microcontroller software, they were able to calculate the exact resistance of the skin and thus the skin conductance.

This would have been a good measurement for the biosensor array but Drew was not able to get their circuit to work correctly. Drew tried both with LM324 quad opamp using 2x voltage follower (to buffer) and 1x as the difference amplifier but saw no signal on the scope. Drew also tried using dual Micrcochip Op-Amp MCP6002 as voltage follower to buffer each input going into the TI INA114 instrumentation amp.

However, no appropriate GSR signal was seen on scope unlike the above Truth wristband circuits by Sean Montgomery. Further development may likely would have resulted in working Wheatstone bridge-based circuit however, time did not allow given and working circuit of a different type was already acheived.

The Cornell EDA project was based on this GSR sensor for MRI machine monitoring. The link on the Cornell EDA page is broken but it is still http://web.archive.org/*/http://physweb.mnstate.edu/shastri/EDABoxPaper.ppt

Note the circuit using a wheatstone bridge and amplifier is also referred to as a bridge transducer amplifier like in the TI INA114 datasheet on page 12: http://focus.ti.com/lit/ds/symlink/ina114.pdf

Related Links