I am in the fifth year at a Electronics Engineering Computer Science
faculty. I have skills in C/C++ programming, microcontroller use, some
knowledge of DSP algorithms. My initial intention was to combine these
3, but the guiding teacher didn't like that too much. I guess that he
wanted to keep it simple, avoid any complications. And I wasn't too
happy about he rejecting my ideea. I was planning to do some
elementary speech recognition, but using a powerfull uC, not a DSP,
for instance dsPICs. All my colleagues that are into "hardware" stuff,
make electronic thermometers, or multimeters, and other simple things.
I really don't like such things, because they are too simple, and you
really don't learn much by doing them. For instance I made a working
remotelly controlled light dimmer, powered directly form power line,
using a single PIC12F629 uC, and a few extra components, from
ground-up, using only the datasheets for the components, in about 1
week. Some thought that was tough...
I'm not a genius or anything but I don't like routine stuff, I like to
lear something new. I know that this kind of project requires some
work, but I'm not afraid of it. I'm just afraid that some teachers
want to see a PIC that blinks a LED, and that makes them happy.
I thought once about making a remotelly controlled FM radio, with
frequency display, using for instance, Philips' TDA7000 and a PIC uC.
But that's again a problem. I actually had a quarrel with a teacher
about the use of analog electronics in digital designs. He is against
anything that's analog, and he's also a member of the evaluation
comission That plainly sucks.
The perfect project for me would be one that incorporates a uC, maybe
more than one for some parallel processing, some signal processing,
some PC link, IRDA, USB, maybe a remote controller, data acquisition,
some elementary analog electronics... I guess that's too much for just
one project.
Since you are looking for a project. Here is a couple of
projects that interest me.
How about building a DC welder out of old computer
power supplies, say 10 or more in parallel. OK that is analog,
but how about controlling the current in 1 amp increments
from 5 to 100 amps. That is a job for a micro controller.
A stick or tig welder power supply usually has an open circuit
voltage around 80 volts but 50V to 100V is still usable. Below
50V is getting a bit low, above 100V is fine but becomes a bit
hazardous to the operator. So aim for roughly 75V no load.
The transformer secondary windings in the computer power
supplies would have to be changed and the control circuits
changed to be a constant current supply. Some method of
current sharing is needed also.
Dial up the desired current on the digital meter before welding.
Make the digital display read the actual current while under
load or welding and if it is within 10 percent or better that would
be wonderful.
---------------------------------------------
How about a handheld electrical joint tester.
A very low reading ohm meter with a digital readout in.
micro ohms, milli ohms or suitable units of conductance.
Say you want to check how good the connections are
between some bolts and a busbar or some heavy duty
diodes and a heatsink you could pass a current of 10 amps
to 100 amps or whatever is appropriate and measure the
voltage across the connection and use your micro controller
to come up with the resistance reading. This might be tricky
if the battery that is supplying the heavy current starts to die
while supplying the current. Maybe you could make the
micro controller take care of the drooping test current
problem.
There you are, a couple of projects that may be of
some use to you, after being submitted to your teacher.
Cheerio,
John Crighton
Sydney
