I'm debugging a simple power circuit involving a TPS61202. It is boosting a 3.7v input from a battery to 5v. For reasons I cannot explain, it only sporadically operates when a 0.1uF capacitor between Vaux and ground is present. Vout is 0v most of the time, but (seemingly at random) about %10 of the time it will output 5v. I am finding the success and failure conditions very hard to reliably reproduce. However, if I remove the 0.1uF capacitor at Vaux, the device seems to work fine.
I found mention of problems with UVLO triggering due to high inrush current, and a soft-start circuit was suggested to fix this, as described in this datasheet:
http://www.ti.com/lit/an/slva307a/slva307a.pdf
I installed this extra circuitry, and while it doesn't appear to hinder the operation of the circuit when missing the Vaux cap, it also doesn't fix the problem when the Vaux cap is present. I had to substitute the following diode instead of a 1N4148, however:
http://www.comchiptech.com/cms/UserFiles/CGRM4001-G%20Thru.%20CGRM4007-G%20RevD.pdf
Any ideas about what might be going on? I've done my best to rule out cold solder joints, bad discrete components, and mislabeled/incorrectly installed parts.