No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

<p>The first prototype version of the current source for electrical impedance tomography was studied. Based upon its performance, a second version of the current source was built, including various improvements designed to address the issues of this first version.</p> <p>The frequency range was extended from 1kHz to 125kHz up to 1kHz to 1 MHz, and additionally the new design allow the production of either sine, triangular or square waves.</p> <p>The DC offset from the DDS output voltage signal was removed and the voltage signal was amplified to ±10V (peak to peak) from ±325mV (peak to peak) in two stages. A fixed gain of 5 was implemented in the first stage and a variable gain (from 0 to 6) was implemented in the second stage to provide more control of the injected current.</p> <p>The single-ended voltage signal from the variable gain stage was translated into double-ended signals by implementing two unity gain buffer operational amplifiers, one inverting and the other non-inverting. This double-ended voltage signal then was applied to the voltage control current source.</p> <p>Three voltage control current source circuits; current pump, improved Howland and Bi-polar were implemented on a PCB and a wire board.</p> <p>The following five quality indicators were used to assess the performance of both versions of the current source: (1) The stability, (2) the balance of the current injection, (3) the output impedance, (4) the variation in the output current due to changes in the impedance at a fixed frequency and (5) the variation in the output current due to changes in frequency at a fixed load.</p> <p>All three circuits in the second version were stable; however, the first version was not stable. Only in the second version - the improved Howland voltage control current source was the current injection balanced over the full range of frequencies.</p> <p>The output impedances of the second version current source circuits were 25KΩ (bi-polar), 256KΩ (Howland) and 7KΩ (current pump) at 100 KHz as compared to the first version 8KΩ at 62.5 KHz. The output impedance for the second version Howland circuit was much higher than the other circuits.</p> <p>The maximum variations in the output current were 6.35% (bi-polar), 22.44% (current pump) and 0.77% (Howland) due to variations in the load (150Ω - 2150Ω) at a fixed frequency (100 KHz). This is in comparison to the first version, which was 10.55% due changes in load (480Ω -1180Ω) at a fixed frequency (62.5 KHz). Clearly the improved Howland circuit demonstrates the lowest current variation as a function of the load.</p> <p>The maximum variations of the output current due to changes in frequency (1kHz to 300kHz) was about 37% (bi-polar), 55% (Howland) and 89.83% (current pump) at a fixed maximum load 2150Ω for this second version of the McMaster EIT current source. This compares to the first version of the current source which was about 8.47% due changes in frequency range from 1 kHz to 62.5 kHz at maximum load of 1180Ω. In this case the bipolar current source demonstrated the lowest variation in output current with frequency.</p> <p>However based upon all of the test results, the improved Howland current source version II outperforms the other circuits and the first version.</p> / Master of Science (MS)