Tool/software:
Hi,
Using AMC3311 to sense 0-300VDC.
Resistor divider with 1Mohm + 10kohm + 1Mohm. AMC3311 measures over the 10kOhm resistor.
At full 300V there is about 1,5V into the AMC3311 which then goes into a MCP3421 18-bit ADC on the other side.
Two problems:
1. At 0V input (over the whole resistor divider) there is still 5-10 mV present on the AMC3311 input pins (measuring with a Fluke V-meter and also the ADC measures this). There seems to be some interference. Moving the 10uF capacitor closer to the AMC3311 makes the problem worse. Removing it also does not help. Putting the 100pF capacitor directly on the input pins of the AMC3311 also makes it worse. Removing (shorting) the ferrites does not help. Shorting the pins on the AMC3311 with a small screwdriver makes the reading 0V. Shorting the 10k resistor also makes reding 0V. Removing the screwdriver again, the reading goes up to 5-10 mV again into the AMC3311 input pins and also into the ADC on the other side.
2. Holding my finger on the AMC3311 IC package the reading goes up about 2.5mV after 10 seconds. Is this within spec? Offset drift should be 10 microvolts/degree.
I think I have to redesign the PCB and pay closer attention to your PCB reference design this time. I thought the ground plane (blue color) was a good thing but maybe not.
Components are not the exact manufacturer article numbers that you specify. Could also be a problem but I am doubtful.
There is three AMC3311 on the PCB. Maybe they can interfere with each other?
Any other suggestions?
