PPK2 -> strange Source-mode current readings after sudden change in range

[WhiteFang]

Hi,

EDITED: This appears to NOT be related to any open circuit, the SAME behavior comes even with abrupt changes to the current scale, without any interruptions. Additional pictures at bottom of post to show this as well.  So then that seems like something goes wrong with the range selection logic internally in the PPK2 then?

I am using a PPK2 and doing some simple testing with an LED and two resistors in parallel. I have one resistor running via a momentary switch, and I have Vout from PPK2 feeding the circuit via a second momentary switch.
I have it set to source mode, 3000mV.

If I press and hold the Vout button, then turn on the output and start sampling, I see values as expected around 1.4mA, if I press the second switch, the current increase to a bout 2.3mA. If I momentarily let go of the Vout button the readings go haywire and show 0-20mA "all over the place" (see snippet below).  Once in this mode, the only way to reliably recover is to turn source off and then on again, and make sure the Vout is connected before turning it on. 

I suspect this has something to do with the PPK2 entering a bad state when the circuit is "broken" but I am very surprised that it is unable to recover once the circuit is completed again?!

It would imagine there would be a number of ways the DSP on the PPK could detect this type of condition and avoid getting "stuck" in some weirdo mode?

Has anyone else seen this issue?

Of course one could argue that this is a corner case, and one would be correct.  This is not how the device would be expected to be used in most cases, but imagine you want to measure some relay operated load for example, then this type of behavior could happen?

I did a PPK2 power-cycle and repeated the experiment so I could include a "clean" sampling metric output from the shell:

New snippets showing the same behavior without any interruption in the circuit.  In the below I temporarily connect a 100Ohm in parallel to suddenly increase the current draw of the circuit. Upon removing the 100Ohm I would expect to fall back to the initial low 1.x mA, instead it goes into "freak mode".

[Sigurd Hellesvik]

Alright, so i connected the LED on the 1k resistor instead.
And the button on the 100R:

The LED does not light now, as it is an 1k resistor, so too low voltage over the LED.
However, from the below graph, we can see that it consumes a bit of current when I plug in the LED:

No, this shows only ~1.5uA with 1k + LED only.

Then when I press the button to connet the 100R resistor, I get the following:

This jumps from ~1.5uA to ~35mA, which should be more than two "ranges" , right?

Here is also one test wiht 100,000 samples per second. I also put the supply voltage at the same that you got:

EDIT:
And for good measure, here is how the circuit looks: