A CMOS Fabrication Report
I used to believe that the fabrication of CMOS was only marginally more difficult than the fabrication of a PMOS. And I was so totally wrong. And it took me 10 gruelling weeks to realize that. Thanks to Kazi, we pulled it off.
[pdf] - The report
(top)

DOE Design
This project comprises of the the design of a Diffracto Optical Element(DOE) for channeling lignt emanating from particular laser sources into into particular photo-detectors. The specs requirement may be downloaded from here. The report and the presentation is present below. Unfortunately, we only got 2 weeks to do the project. I wish we had more time.
[pdf] - The report.
(top)

Building a PMOS at the SMFL
This was one of the most memorable projects that I had undertaken. Its a report on how we fabricated a PMOS at RIT. Of ourse it was not something really fancy - well it was never meant to be. I also wanted to take the next course on thin films, but then we can only take so many courses. Too bad.
[pdf] - The report, the silvaco file and some data.
(top)

A pedagogical discourse on EHD Pumps
I did this for the last course that I passed at RIT. It was a tribute the dream that I always had – to teach.
[pdf] - The report
(top)

Design of IC Inductors
The first of a series of lab exercises that I had to complete for the RF IC design course. I never thought till this day that the design of passives could be so complicated. This was a lot of fun though. There is a program (Asitic) that allowed us to calculate the inductance parameters, and we extracted large volumes of data and visualized it. It was great fun.
[pdf] - The pre-lab report
[pdf] - The final report
(top)

Design on a RF LNA
Another of the lab exercises that I had to complete for the RF IC design course. This was fun.And then later I came to realize that the LNA was one of the only circuits that had any theoretical design principles in the world of RF IC design. Sad isn’t it? One of the reasons why I wanted to become a device physics guy. How are people going to come up with coherent design specs for RF ICs if people didn’t even have analytical equations that represented the true picture of the devices at high frequencies?
[pdf] - The pre-lab report
[pdf] - The final report
(top)

Design of a RF Mixer
This was the last of the Lab exercises that I had to complete for the RF IC design course. Had to complete this in a hurry, and without any specs to meet. Think this was a little bit of a disaster because there was very little to expect.
(top)

Self consistent solution to the Schrodinger-Poisson Equation of a AlGaAs/GaAs heterojunction
This wasn’t exactly self-consistant. Common, after all I had only a couple of weeks to do it.
(top)

Implementation of the CORDIC Algorithmon an FPGA
That’s something I had to do for my VHDL course. Extremely difficult. We had to implement it by synchronous, asynchronous and wave-pipelined architecture. Most of the difficulty lied in the tool itself (Altera). Half the commands would not work in that version of the tool we were using.
(top)

Design of a Comparator
This I had to do for my Analog IC class. Fun, and a lot of hard work. And I had to do it twice because my initial design was too ‘big’. Means I had to start all over from scratch after my final layout was done. That’s the time I learnt how little we know about the world around us, and how limited math is.
[pdf] - The report (The annexures were images of whichthe layout is provided...)
[gif] - Layout
(top)

Numerical solution of charge buildup on a MOS device with its comparison
This was done in the second quarter. I did a numerical integration for the charge buildup and compared it with the analytical result. Kinda re-inventing the wheel maybe but I just wanted to check it out for myself.
[pdf] - The report
[pdf] - Annexure A
[pdf] - Annexure B-E
(top)

Robotic Navigation: The Khepera solving a maize
This was the first project that I did at RIT and perhaps the most memorable. In this project, a microbot called the Khepera was programmed to traverse a closed maze and then come up with a three-dimensional view of the maize. The project was implemented using KDevelop the development environment of KDE. Graphics routines of Qt were used but the 3 d routines were hand coded. The code for serial-port access was borrowed from the internet (unfortunately the site from which I had downloaded the serial port program does not seem to be functional any more). It was the most exciting project of all.
The project can be gotten here
(top)