EE Senior Design Project

[Skillet49]

I am a senior EE major at Northern Illinois University and this semester I am enrolled in our equivalent of a Senior Design proposal class.

We've had most of the semester to come up with designs for a feasible project to build next semester. I along with a couple of other students who want to form a team came up with the idea to develop a self-calibrating surround sound system for a home theater system. We knew that they existed, but until we started really researching we did not realize how advanced and prevalent their are in the mainstream.

This leaves us kind of hanging loose. We need to have an idea finalized and presentable tomorrow, but and we've spent the better part of a week trying to figure out an idea that is practical and actually might be fun doing. Now we're almost to the point of desperation in trying to find a project.

I was wondering if anyone had any ideas of things that could be improved on, or a simple electronic device that would be worth designing.

At this point the more ideas the better.


Thanks

 

[RapidSnail]

I guess it's nice that you're asking for help, but aren't you supposed to think about this during the time leading to your senior design?

 

[Tiamat]

great ideas don't come this quickly unless you are drunk. Better get started! :beer:

 

[FleshLight]

http://www.youtube.com/watch?v=9WqlvukVq0U

 

[Skillet49]

Well we are encouraged to think of ideas and try to find projects to work on and we really thought that we had a good project going. It just turned out that at the last minute we realized that this was not going to work as well as we thought and we have been scrambling ever since to figure out another idea. I'm not looking necessarily for people to give me the project that is the most awesome innovation ever. I'm just trying to get ideas that might other people with other daily experiences might have.

 

[Deeko]

The title of this thread made me shudder thinking about my senior design project.

 

[bobsmith1492]

Do you have any requirements? Can you build an automated plant watering device (measures soil dampness electrically then turns on a solenoid to water it) or is that too simple? Could you build a dusk/dawn light simulator alarm clock that ramps the lights up and down at specified times? Can you make a Class D amplifier from scratch? How about a fully-featured stereo amp with preamp, digital volume control, and a DAC for SPDIF input (my current for-fun project)?

How much time do you have? What budget do you have and who is paying for it?


Good luck. I spent 7 months on our senior project...

Our school is nice enough to find sponsors from local companies who bring proposals to us. I can't even imagine trying to come up with an idea for a project.

 

[JohnCU]

for mine, we used a camera that tracked a ball through a maze. the maze tilted back and forth to navigate the ball from start to finish. it was neat. just two motors, camera, and matlab. completely closed loop so it could always fix itself.

also, we turned one of the lab tables into a bar at night. god i miss college.

 

[bobsmith1492]

Originally posted by: JohnCU
for mine, we used a camera that tracked a ball through a maze. the maze tilted back and forth to navigate the ball from start to finish. it was neat. just two motors, camera, and matlab. completely closed loop so it could always fix itself.

also, we turned one of the lab tables into a bar at night. god i miss college.

We made an automated loudspeaker test system.

It consisted of multiple, independent modules in a chassis connected to a network switch. Each module had a 1GB SD card to store test data and a test audio MP3 track. We made a PC GUI in C++ to control and view results of the tests but you could shut it off at any time since the modules did everything by themselves - our sponsor was paranoid about losing test data during a month-long test...

The modules play back an MP3 into an external amplifier. This drives the loudspeaker under test. The module measures and calculates RMS current, voltage, and real power (instantaneous I*V) at 96KHz sampling and measures temperature of the coil by checking its resistance. The control algorithm then adjusts the output power for the next run through the test track so you can test at a constant -temperature, -current, -voltage, -power, or -test with increasing power until the unit is destroyed.

Any MP3 can be uploaded as a test track and results can be exported into CSV files. The results are also stored on the SD card which can be removed to view the files in case a module fails (standard FAT32 formatting).

We wrote all the code from scratch (2x ARM7 microcontrollers and the C++ GUI, a FAT32 filesystem, 20 minutes voltage/current/power averaging capability, ethernet communications via sockets), designed all the circuits from scratch (no modules): 20A, 150VRMS measurement capability +/-~5%, power failure data protection, MP3 decoder chip and outputs, ethernet interface and socket control code, power supplies on-board, the works.

Then we built 6 channels and tested them all. All in all it was about $3000. The sponsor got a great deal, though - I put in >70 hours/week for at least two months during the peak. That's when I realized the engineering world is all about man-hours and equipment costs are largely irrelevant.

 

[Skillet49]

Originally posted by: bobsmith1492
Do you have any requirements? Can you build an automated plant watering device (measures soil dampness electrically then turns on a solenoid to water it) or is that too simple? Could you build a dusk/dawn light simulator alarm clock that ramps the lights up and down at specified times? Can you make a Class D amplifier from scratch? How about a fully-featured stereo amp with preamp, digital volume control, and a DAC for SPDIF input (my current for-fun project)?

How much time do you have? What budget do you have and who is paying for it?


Good luck. I spent 7 months on our senior project...

Our school is nice enough to find sponsors from local companies who bring proposals to us. I can't even imagine trying to come up with an idea for a project.

I think that an automated plant watering device might be a bit to simplistic for this sort of a project. However twists could be thought of for a system like that that would be possible.

Other than that we have to be able to actually finish the project, and our budget is fairly limited ($25 that the school provides).

Some companies do sponsor some projects, unfortunately not very many actually come and propose ideas for us to use.

Thanks for the ideas though, they might be useful

 

[JohnCU]

Originally posted by: bobsmith1492

Originally posted by: JohnCU
for mine, we used a camera that tracked a ball through a maze. the maze tilted back and forth to navigate the ball from start to finish. it was neat. just two motors, camera, and matlab. completely closed loop so it could always fix itself.

also, we turned one of the lab tables into a bar at night. god i miss college.

We made an automated loudspeaker test system.

It consisted of multiple, independent modules in a chassis connected to a network switch. Each module had a 1GB SD card to store test data and a test audio MP3 track. We made a PC GUI in C++ to control and view results of the tests but you could shut it off at any time since the modules did everything by themselves - our sponsor was paranoid about losing test data during a month-long test...

The modules play back an MP3 into an external amplifier. This drives the loudspeaker under test. The module measures and calculates RMS current, voltage, and real power (instantaneous I*V) at 96KHz sampling and measures temperature of the coil by checking its resistance. The control algorithm then adjusts the output power for the next run through the test track so you can test at a constant -temperature, -current, -voltage, -power, or -test with increasing power until the unit is destroyed.

Any MP3 can be uploaded as a test track and results can be exported into CSV files. The results are also stored on the SD card which can be removed to view the files in case a module fails (standard FAT32 formatting).

We wrote all the code from scratch (2x ARM7 microcontrollers and the C++ GUI), designed all the circuits from scratch (no modules): 20A, 150VRMS measurement capability +/-~5%, power failure data protection, MP3 decoder chip and outputs, ethernet interface and socket control code, power supplies on-board, the works.

Then we built 6 channels and tested them all. All in all it was about $3000. The sponsor got a great deal, though - I put in >70 hours/week for at least two months during the peak. That's when I realized the engineering world is all about man-hours and equipment costs are largely irrelevant.

good lord. you obviously went to a better engineering school than i did. cool project :thumbsup:

 

[CraKaJaX]

Originally posted by: bobsmith1492

Originally posted by: JohnCU
for mine, we used a camera that tracked a ball through a maze. the maze tilted back and forth to navigate the ball from start to finish. it was neat. just two motors, camera, and matlab. completely closed loop so it could always fix itself.

also, we turned one of the lab tables into a bar at night. god i miss college.

We made an automated loudspeaker test system.

It consisted of multiple, independent modules in a chassis connected to a network switch. Each module had a 1GB SD card to store test data and a test audio MP3 track. We made a PC GUI in C++ to control and view results of the tests but you could shut it off at any time since the modules did everything by themselves - our sponsor was paranoid about losing test data during a month-long test...

The modules play back an MP3 into an external amplifier. This drives the loudspeaker under test. The module measures and calculates RMS current, voltage, and real power (instantaneous I*V) at 96KHz sampling and measures temperature of the coil by checking its resistance. The control algorithm then adjusts the output power for the next run through the test track so you can test at a constant -temperature, -current, -voltage, -power, or -test with increasing power until the unit is destroyed.

Any MP3 can be uploaded as a test track and results can be exported into CSV files. The results are also stored on the SD card which can be removed to view the files in case a module fails (standard FAT32 formatting).

We wrote all the code from scratch (2x ARM7 microcontrollers and the C++ GUI, a FAT32 filesystem, 20 minutes voltage/current/power averaging capability, ethernet communications via sockets), designed all the circuits from scratch (no modules): 20A, 150VRMS measurement capability +/-~5%, power failure data protection, MP3 decoder chip and outputs, ethernet interface and socket control code, power supplies on-board, the works.

Then we built 6 channels and tested them all. All in all it was about $3000. The sponsor got a great deal, though - I put in >70 hours/week for at least two months during the peak. That's when I realized the engineering world is all about man-hours and equipment costs are largely irrelevant.

What school? U Mich? (Guessing from profile)

 

[bobsmith1492]

Originally posted by: CraKaJaX

Originally posted by: bobsmith1492

Originally posted by: JohnCU
for mine, we used a camera that tracked a ball through a maze. the maze tilted back and forth to navigate the ball from start to finish. it was neat. just two motors, camera, and matlab. completely closed loop so it could always fix itself.

also, we turned one of the lab tables into a bar at night. god i miss college.

We made an automated loudspeaker test system.

It consisted of multiple, independent modules in a chassis connected to a network switch. Each module had a 1GB SD card to store test data and a test audio MP3 track. We made a PC GUI in C++ to control and view results of the tests but you could shut it off at any time since the modules did everything by themselves - our sponsor was paranoid about losing test data during a month-long test...

The modules play back an MP3 into an external amplifier. This drives the loudspeaker under test. The module measures and calculates RMS current, voltage, and real power (instantaneous I*V) at 96KHz sampling and measures temperature of the coil by checking its resistance. The control algorithm then adjusts the output power for the next run through the test track so you can test at a constant -temperature, -current, -voltage, -power, or -test with increasing power until the unit is destroyed.

Any MP3 can be uploaded as a test track and results can be exported into CSV files. The results are also stored on the SD card which can be removed to view the files in case a module fails (standard FAT32 formatting).

We wrote all the code from scratch (2x ARM7 microcontrollers and the C++ GUI, a FAT32 filesystem, 20 minutes voltage/current/power averaging capability, ethernet communications via sockets), designed all the circuits from scratch (no modules): 20A, 150VRMS measurement capability +/-~5%, power failure data protection, MP3 decoder chip and outputs, ethernet interface and socket control code, power supplies on-board, the works.

Then we built 6 channels and tested them all. All in all it was about $3000. The sponsor got a great deal, though - I put in >70 hours/week for at least two months during the peak. That's when I realized the engineering world is all about man-hours and equipment costs are largely irrelevant.

What school? U Mich? (Guessing from profile)

Na, Grand Valley!