Re: electrical problems with Galvanic Skin Response circuit ( GSR SCL )

Stefan Jung <[email protected]>
Newsgroups gmane.comp.science.openeeg.general
Message-ID <CAC3TAzMZbdzE+sQdjFijmNMK8Oxv7LiqU399ODykbwbh7P=Lrg@mail.gmail.com>
Hi Steffen

I don't have a clue why your circuit doesn't work.  That's the fun of
electronics, even a simple circuit could have 40+ reasons why it doesn't
work as advertised.  The fact is, a hobbyist or newbie can sometimes make
elementary mistakes that an EE will just not imagine or think off.

There are many, many opamps that are also rail-to-rail.  You cannot use an
opamp in place of a INA.  For one, the INA has practically infinite input
impedance.  In the SCL circuit an INA is a handy device to produce (1) an
easy adjustable gain, (2) subtracts the DC excitation and (3) can adjust
the DC offset, should that be necessary.  The result being a nice 0...5V
output swing which is handy when reading with a microprocessor's internal
ADC.  Such internal ADC is just about always 10 bits, so you will want to
use the full input voltage swing for reasonable signal resolution.  I would
nearly always consider it a bad design if only a limited input range is
used for an ADC, as you simply waste the ADCs dynamic range/ resources.

In the case of the SCL circuit you CAN replace it with a rail-to-rail
opamp, but it would be a redesign of the circuit.  The fact is you would
need to add quite a few resistors with specific calculated values.  You
would need to work out the transfer function of that circuit from first
principles so as to calculate the various resistors.  Your circuit would
need to fulfill points (1) and (2).  (The DC offset (3) is not required in
the SCL circuit.)  If you want to adjust the working point, you would
likely need to change ALL resistors.  The INA just makes life much easier.

You could get another INA such as a AD620, INA128, INA129, etc. and run
that from say 10V.  I doubt if it would work well down to zero volt.  You
might have to run it from a split +/- rail.  You could build this for
testing and comparing with the AD623.  I would not use that for the proper
circuit. It would complicate the power supply and you could damage the ADCs
input should things go wrong.  So you would have to add more components to
clamp the INAs output to always stay within the ADCs input voltage.

Regards
Stefan

------------------------------------------------------------------------------
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.