Inrush Current Ripple

[fpmkh0]

Hi,
I have a opamp based PGA. When I look at the inrush current (enabling PGA, and looking at the supply current), I see ripples in the current and I can't justify it. During startup, the capacitors of the nodes are acting as short so I expect the peaking, but then I expect an exponential decay only. Where does these ripple come from? Does anyone have experience about this?

 

[PlanarMetamaterials]

Is this simulation, or measurement?

If this is just a simulation, I would check the upstream devices for possible inductance. You could also have an inductive feedback loop somewhere. Post your schematic so that we can comment further.

In real life (i.e., measurement), there are a variety of possible sources for this behavior. Capacitors unfortunately don't act as shorts; they have equivalent series resistance and inductance that can cause delays or oscillation. Check the capacitors data sheets for more information. Any other devices or even the PCB layout can cause unwanted parasitic reactances.

 

[fpmkh0]

This is simulation. At this point, there is no extracted view and I don't have any actual or parasitic inductance available in the schematic. Although, I understand sometimes impedances can become inductive even when there is no inductance. The feedback network consists of capacitor and resistor (to form a LPF).

 

[D.A.(Tony)Stewart]

Resonance in simulations without inductance may occur when the risetime approaches the sample interval, causing errors on the verge of convergence errors.

 

[KlausST]

Hi,

another possible issue: Amplifier input stage overdrive. Maybe due to too high input dV/dt.

Without seeing schematic and test conditions it´s all just wid guessing...



Klaus

 

[fpmkh0]

Hi,

another possible issue: Amplifier input stage overdrive. Maybe due to too high input dV/dt.

Without seeing schematic and test conditions it´s all just wid guessing...



Klaus

Thank you for your response. I don't apply any input in this simulation. Can you please elaborate what exactly you mean by input overdrive and how it can cause ripple in supply inrush current?
Regarding schematic, unfortunately, I'm not allowed to share. It's just a non inverting amplifier with RC feedback to make a PGA LPF.

 

[FvM]

Without a schematic it's just guessing. LPF feedback might reduce phase margin and cause oscillations.

 

[KlausST]

Hi,

I don't apply any input in this simulation.

You don´t apply any input? This means the input is floating. --> you can´t expect any meaningful result.
The input is random, so the behaviour will be random.

Klaus

 

[fpmkh0]

Hi,

You don´t apply any input? This means the input is floating. --> you can´t expect any meaningful result.
The input is random, so the behaviour will be random.

Klaus

Sir, input is just biased to the common mode. But I haven't applied any input signal. Whether I apply input or not, I see the same behavior. I don't apply input because I wanted to make sure the ripples are not from signal coupled there.

 

[dick_freebird]

Might just be the reference circuitry startup and the amplifier regular
bias loop branches being on simultaneously, or "trying too hard"
turns into "sticking around too long". That apply seems pretty abrupt,
most supplies ramp slowly. Are you being realistic, or making your
own problems?

Maybe some of the excess is blowing through to the load. Maybe,
on account of abrupt / unstable supply, itself less than stable close-in.

 

[FvM]

The advantage of a simulation is that you can look at any node voltage and branch current. Thus you can see how the oscillation is brought up.

 

[danadakk]

This is simulation. At this point, there is no extracted view and I don't have any actual or parasitic inductance available in the schematic. Although, I understand sometimes impedances can become inductive even when there is no inductance. The feedback network consists of capacitor and resistor (to form a LPF).

That is true, one can take an OpAmp and C and create L, good example is
gyrator :

If you do a simple s plane analysis of a single pole OpAmp, inverting amp, you
will see L behavior in input and output Z of the OpAmp.


Regards, Dana.