ADC Input Problem

[nipri]

I am using four ADC inputs to the 9160.

I am driving each input from an op amp buffer as in the attached schematic.

ADC channels 0, 2, and 3 work as expected. As for channel 1, I measure 0 volts to ground at the node of R1, C7 and ain1 but pin 7 of the opamp has the expected voltage on it to ground.

I thought, at first, that  this node was shorted to ground but this is not the case. If I short across R10, the ADC input works as expected. Also, if I remove R10 and apply a voltage from 0 - 3.3v directly to the ADC input, that will also work.

This board is part of a prototype run of which we did two separate builds, one in June of last year, the other in September. This problem only exists on the 2nd batch of boards and not on the first. There were no hardware design changes between the two runs and there are no measurable electrical differences on the nets or components in question  between the two runs.

The buffer / low pass circuit shown below is one I typically use with SAR ADCs with little trouble.

I believe the 9160 modules are from the same batch between the two builds but am working to verify this.

Thanks in advance

[Rob Garner]

Try removing R10, then measuring for continuity between the one pad of R10 and the output of the opamp (pin 7). Try the same for the other resistor pad to the top pad of C7. Depending on the CAD tool you used for the layout, traces are sometimes not connected 100%, so there could be a hairline gap in the trace somewhere that creates an open circuit.

[nipri]

Thanks Rob! I will measure the continuity as you suggested. The voltages I measured were on the end caps of R10. Also, the circuit works when I remove R10 and replace it with a jumper soldered across the pads in its place. Will report my findings tomorrow. Using Altium BTW.

[Rob Garner]

Also, have you verified that all four ports of the ADC are configured as anaolog inputs (so no pullups or pulldowns enabled, etc).