Here is a document explaining how to use the software, and the latest version of the software in Labview 8.0. I hope I have done everything correct this time. With this information you should be able to tune the software, and measure skin conductance. Good luck. I am leaving soon but back in 3 weeks. Then I will post the full software. I found out it is actually quite similar to the software made by Mindware Technologies. It do almost the same analysis, but the Mindware software has more functions i guess.
Here is link http://www.limef.com/productImages/skinbig_600.jpg
I use standard ECG electrodes pre-gelled silver/silver chloride type. I know someone who supply me with expired electrodes for free. As mentioned in my document homemade variants is not recommended. They have very poor performance
This will be my last posting for a while , think. I am going on vacation soon. If anyone was speculating in what this system measure, I post this explanation
It is important to emphasize the fact that this setup only will measure changes in the conductance part of the skin admittance. Changes in the susceptance part will not be measured. In order to perform such measurement we need a second phase-sensitive detector, which is fed with the same signal input, but which is driven by a reference signal that is phase-shifted by 90 degrees. This second detector is followed by a second output filter will representing the susceptance part. Then we will have
Y is the admittance, measured in Siemens i.e. mho (inverse of ohm).
G is the conductance, measured in Siemens i.e. mho (inverse of ohm).
B is the susceptance, measured in Siemens i.e. mho (inverse of ohm).
What would be the best operational amplifier to go with in this project?
Is there an easy way to provide for the millivolt oscillator, Vref, noted in pdf file skin conductance?
Yes the soundcard In my posting 09.jun.2009 16:47 "SC system without osc.pdf " The soundcard is the sine wave source, and the voltage is attunated with a pot to the correct level. By the way did you got my private message. I have used some types opamps. but if you need a good one you can use in all places. Use LT1057/LT1058
I think these Opamps should work. But it is only a guess. I have used the components codes from my posting 09.jun.2009 16:47"SC system without osc.pdf "
TL051/052/054 (all opamps)
opa2277/4277 (U1, U3B). All opamps with low offset and say a 1Mhz bandwidth should do well here
LF411 (all opamps)
OPA606 (U3A,U2 My be used all places but this is overkill)
LT1057/LT1058 (all opamps)
LT1014 work as U2, But It has a low Gain BandWidth factor so it is not the best in this setting
For the U2 I have tested some types but I always go back to the LT1115. For U3A I prefer to use a FET type opamp, at least I have always done it.
My selection is much based on my local electronic store assortment and my IC collection I have to admit. Plus the fact that I do not have SMD solder equipment
Also remember that you has to use my software or a Lock-in amplifier if you are going to use this circuit. You want get much out of it by connecting it to plain datalogger
If I were to tackle this project I would use my NI USB-6211 in place of the sound card.
I will cross reference the parts, LT1057/LT1058, with Digi-Key.
If the oscillator frequency is low, I will probably bread board this project.
You may use your USB-6211 in place of the sound card. I used a sound card because I wanted to keep the budget low. If you are using your USB-6211 as the AC source I recommend that use most of the +/- 10 volt range and attenuate the voltage to a suitable voltage before the input filter. I also think you have make some small changes in the gain setting some places. but I guess you can figure it it out. I once used an old DAQCard 516(PCMCIA) Nidaq 6.9 and labview 8.5 in a similar application worked very well measuring data during a surgical procedure.
In case you have any question/problem I will giude you as good as I can
By the way are you affiliated with Mindware Tech?
Edit: You may use a frequency in the range say 40 to 200 Hz. I will explain this later. Stay away from frequencies close the the mains power frequencies or multiples. Test and see which frequency what work best for you. I often use 89 Hz. Scan through some frequencies in 1 Hz steps and see which frequency that will give the signal to noise ratio. Then using the calibration setup