piranha32
-
Content Count
8 -
Joined
-
Last visited
Reputation Activity
-
-
piranha32 got a reaction from oPossum in Hello from Washington
The project I'd like build using the launchpad is a digital rlc meter. The c2k family has two features which make it ideally suited for such job: dual channel simultaneous sampling AD converter and lots of computational power. The simultaneous sampling is important in measuring the phase shift between voltage and current. An A/D with sequential sampling would make the job much more difficult. Many such projects just look at the min/max sample values to determine the amplitude of the measured voltages and look for zero-crossing to estimate the phase. While simple to implement, such methods require high degree of oversampling to be accurate. I want to fit sinusoidal functions to the measured values and use them to estimate the parameters of the voltages and current. It requires solving of several non-linear optimization problems, but can work well with only 4-5 samples per period. That's where the high MIPS power comes in handy.
I have a working proof-of-concept implementation in octave, now I need to implement it on the microcontroller. I'll keep you posted on the status of the project, but don't hold your breath. It may take a while
-
piranha32 got a reaction from msptest6 in Hello from Washington
The project I'd like build using the launchpad is a digital rlc meter. The c2k family has two features which make it ideally suited for such job: dual channel simultaneous sampling AD converter and lots of computational power. The simultaneous sampling is important in measuring the phase shift between voltage and current. An A/D with sequential sampling would make the job much more difficult. Many such projects just look at the min/max sample values to determine the amplitude of the measured voltages and look for zero-crossing to estimate the phase. While simple to implement, such methods require high degree of oversampling to be accurate. I want to fit sinusoidal functions to the measured values and use them to estimate the parameters of the voltages and current. It requires solving of several non-linear optimization problems, but can work well with only 4-5 samples per period. That's where the high MIPS power comes in handy.
I have a working proof-of-concept implementation in octave, now I need to implement it on the microcontroller. I'll keep you posted on the status of the project, but don't hold your breath. It may take a while