I’m currently using a Panda II to read acceleration values from a ADXL335. This is working so far, but the analog readings deviate 10% or more each reading (without moving the accelerometer, I also checked the voltage with a multimeter and it stays absolutely the same).
I then connected An3 and An5 directly and used An3 to output 0-1023 and An5 to read the value. Here I also got about 10% deviance between the value I write and the one I read. So it seems to be a general problem.
With my Arduino Uno, I never had those problems.
Is this normal or is there anything I can do about it? Any help is appreciated
Cheers, Dave
10% is unusual. Only enable analog input on the one pic you have connected to accelerometer. Make sure you have good power source and good decoupling caps.
I use the onboard 3.3v for powering the ADXL335, and the output seems to be stable (readings from the panda and sensor via an external multimeter area absolutely steady). As mentioned I get those errors even if I connect two analog pins (one as input, one as output) and comparing input to output.
GND will get 0, 3.3 will get 1023 but in between it is floating around 10%.
All other analog ports are not enabled.
I have no clue how to proceed from this point on. It is possible that my board is faulty?
Sample time instability of the USbizi in combination with .NETMF?
Comparing Arduino with C-code and FEZ with .NETMF is like comparing apples and pears.
Both are fruits, but the taste is completely different.
If I have to sample analog data in .NETMF, i use a dedicated chip like e.g. Microchip MCP3204 (4 channel)
or MCP3208 (8 channel). These chips are 12 bit and communicate through SPI with a processor and handle their AD conversion autarc from any processor.
Cu Wim.
P.S. maybe there are shield with these processors?