bias_steps.take_bgmap(S,cfg) leaves a 0.29V bias on all bias lines

[samdayweiss]

Hi All, we've noticed on TSAT that bias_steps.take_bgmap() sometimes (meaning, not consistently, but it has happened 4-5 times to us now on different slots) leaves a 0.3V bias on the all bias lines after running. Is there no routine to check that all biases are set to 0?

Thanks!

[msilvafe]

It looks like this is the way the function is written this way, with a dc_voltage parameter that is default to 0.3V:

sodetlib/sodetlib/operations/bias_steps.py

Line 1060 in 56d04ce

S.set_tes_bias_bipolar(bg, dc_voltage)

The doc string does not actually described why that dc_voltage is needed but you could just try running that function but passing dc_voltage = 0 instead. We may consider modifying the function to set the dc_voltage to 0 at the end of the function anyway but maybe @jlashner can recall why we added that argument in the first place?

[jlashner]

Ah yes I forgot that I did that. I think the non-zero DC voltage makes the math of computing DC params easier, so you don't get infinities places like here:

sodetlib/sodetlib/operations/bias_steps.py

Lines 657 to 658 in 56d04ce

	Pj = I0 * R_sh * (Ib - I0)
	R0 = Pj / I0**2

[msilvafe]

In that case should we pr a change that zeros it out at the end?

[jlashner]

ya i think that's valid

[msilvafe]

Ok, addresses in PR #400.