MCP1640 - Negative ramp sawtooth 125mV/38ms Vout?

[paw]

Hi all,

Is this normal/expected? At 125mV (if I'm reading it right) it seems huge. I see the same powering from 1xAA, Cr2032 and power supply anywhere between 0.8-3.0V with no change. As probably expected changes in capacitance (decoupling) will affect the ramp length.

I see the same across 3 boards and 2 layouts (1 closely resembling the one from the datasheet). With 5mA and without any load. Looking at it it seems something is getting charged and then discharges over 38ms (depending on the load presumably).

My questions:
- could this be right or is there something obvious wrong?

Many thanks for any suggestions!