[SOLVED] Op-amp stability in the current and voltage references

Hello

in the current or voltage reference source when we use the the op-amp (like in BGR), I have read that we must take care about the op-amp stability so we need to consider the compensation. My question is here, why do we need to consider the stability while the current or voltage reference circuits is working under DC condition, there is not such a frequency

Regards

Junus2012 said:

... why do we need to consider the stability while the current or voltage reference circuits is working under DC condition, there is not such a frequency

Click to expand...

Ciao Junus,
yes, there is: the stability of an opAmp doesn't depend on the (main) frequency where you operate it.
Moreover, you want noise separation from input, and PSR.

Hello erikl

you mean to say that the noise can make the opamp oscillate if we dont compensate it in the BGR ??

erikl said:

Ciao Junus,
yes, there is: the stability of an opAmp doesn't depend on the (main) frequency where you operate it.
Moreover, you want noise separation from input, and PSR.

Click to expand...

Yes noise can make your circuit unstable. As you must be knowing thermal noise is wideband. The high frequency noise components will excite your loop. If there is insufficient phase margin, then the loop will not behave properly. If the phase margin is negative, then your loop goes into positive feedback.

rakshitdatta said:

Yes noise can make your circuit unstable. As you must be knowing thermal noise is wideband. The high frequency noise components will excite your loop. If there is insufficient phase margin, then the loop will not behave properly. If the phase margin is negative, then your loop goes into positive feedback.

Click to expand...

No, it`s NOT the noise that makes a circuit unstable. Either a circuit is stable or it is not. That is (a) independent on noise and (b) also independent on the used signal frequency.
If a circuit is unstable it needs a small "kick" to leave the equilibrium - and in this context you often can read that the always present noise will provide this initial "kick".
But that is nonsense. The power supply switch-on transient (or a small disturbance within the circuit or on the power line) always is sufficient to cause this effect.
But - as a most important information you should know that a circuit with a negative phase margin goes into saturation or it oscillates - independent on the desired application frequency (even if it is dc).

Junus2012 said:

you mean to say that the noise can make the opamp oscillate if we dont compensate it in the BGR ??

Click to expand...

No, this is nonsense, as LvW already noted: either it is unstable - or not. But you might want to get rid of input voltage and power supply changes ("noise") up to a certain (required) frequency - so you need the BGR opAmp to have a sufficient bandwidth.

LvW...you are absolutely right.

Hello all

Erikl, As I understood from you all guys that even if the Op-amp is working in a DC circuit we must take care about the compensation.

but the think you said Erikl that the op-amp must have a sufficient bandwidth, I dont think so we really need a certain bandwidth unless if we consider the settling time behavioural for fast switching otherwise the circuit of BGR is under DC condition of zero frequency

erikl said:

No, this is nonsense, as LvW already noted: either it is unstable - or not. But you might want to get rid of input voltage and power supply changes ("noise") up to a certain (required) frequency - so you need the BGR opAmp to have a sufficient bandwidth.

Click to expand...

Junus2012 said:

... even if the Op-amp is working in a DC circuit we must take care about the compensation.

Click to expand...

We must take care about its stability (phase margin). For single stage amplifiers usually you don't have to take care about the compensation.

Junus2012 said:

... for fast switching ... otherwise the circuit of BGR is under DC condition of zero frequency

Click to expand...

Yes, if you can filter out these fast switching fluctuations behind the BGR.

thank you all

it was very helpful discussion for me

see you for sure with another post

Kind Regards