This question is to help me with my work in building a VCCS for the excitation signal of an Electrical Impedance Tomography measurement set-up to use with EIDORS. I have only basic knowledge in electrical engineering since I'm a computer science student.

I'm trying to build a similar system as described in this paper for my Electrical Impedance Tomography (EIT) system. In particular, I took this Figure from the paper as a reference for building my test system:

EIT System by Yang Zhang, Robert Xiao, Chris Harrison

I am generating a sinusoidal excitation signal (10kHz-100kHz) with the AD5930 DDS waveform generator that has a full-scale current output of 3-4mA and 0.56V peak-to-peak.

I want to build an AD8220-based VCCS to generate the EIT excitation signal. However, I don't know how to build a VCCS, especially with the AD8220 instrumentation amplifier, and was hoping someone could help me with that?

The constant AC current should be less than 1mA, for example, around 100uA p-p would be sufficient.

What I have tried so far: I used LTSpice to simulate the following circuit to try and build a VCCS:

LTSpice Simulation

However, I don't know how to set and calculate R2 given R_Load and if this really is a VCCS and the right approach to this.

  • \$\begingroup\$ You desperately need to talk to someone in your department who understands this stuff. The "circuit" for the VCCS is not actually a circuit, and your attempt to copy it will not work. Furthermore, why have you decided on an AD8220? Asking it to work at 100 kHz is asking for trouble. As for your basic approach, you have not specified whether the load being driven must be grounded or can be floating, and this is critical. Don't try to learn an entire new subject by asking strangers on the internet. Go find someone in your school and throw yourself on their mercy. \$\endgroup\$ – WhatRoughBeast May 27 '18 at 17:34
  • \$\begingroup\$ How ACCURATE must your currents be? and how well must they track? \$\endgroup\$ – analogsystemsrf May 28 '18 at 4:22
  • \$\begingroup\$ The current does not have to be super accurate I think. And I do understand most the stuff I just never really applied it and thought this would be my chance to experiment. For the AD8220, I see that it makes sense to use it as an ADC preamp but I do not see why and how it is used for building a VCCS? And the load being driven must be grounded. \$\endgroup\$ – gammaALpha May 28 '18 at 10:17
  • \$\begingroup\$ Take a look at analog.com/en/analog-dialogue/articles/… It covers VCCS and how to improve them. Some nice stuff. \$\endgroup\$ – GB - AE7OO Dec 6 '19 at 22:29

I'm going to suggest you abandon the approach you are taking because it looks like there will be too much of a learning curve. For a start you need a bipolar supply for a true AC delivery of current but I would just go for the Howland current source: -

enter image description here

The op-amp needs power rails of course and these should be typically bipolar in nature typically +/- 15 volts. Vin would be your voltage signal representative of a sine wave with no dc offset. Can be removed using an RC high pass filter.

You could buy one from Analog Devices if you didn't want to build it: -

enter image description here

It's good for well over 100 kHz.

  • \$\begingroup\$ would this circuit work if I built it with the AD8220 or do I need a regular op-amp? \$\endgroup\$ – gammaALpha May 27 '18 at 16:25
  • \$\begingroup\$ It works with a regular opamp. \$\endgroup\$ – Andy aka May 27 '18 at 17:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for?Browse other questions tagged or ask your own question.