Solved: Re: Switching time for mux on Opamp - STMicroelectronics Community

Various STM32 products have opamps that have a mux in front of it.Ex: STM32G4, STM32F3, STM32H7 series, etc.

If I change the mux input setting using the control status register (OPAMPx_CSR::VP_SEL or VN_SEL), how long does it take for the mux to switch to the new setting?

The datasheet doesn't have this information.

1 ACCEPTED SOLUTION

Accepted Solutions

Hi @Singh.Harjit ,

Sorry for delay to provide an answer, I was checking for answer with our Experts.

Unfortunately, We don’t have exact values to answer your question "how long does it take for the mux to switch to the new setting?". But you may consider that the switching time of the switch itself is few 10th of nsec to 100th nsec.

However, as the feedback loop of the OPAMP is broken by its nature, it needs some time to become a stable output (close the feedback loop).

This will be limited by the slew rate of OPAMP and condition of the circuit (for example capacitance in the feedback loop).

For example, if one of the GPIOs is 0.1V and another one is 2.9V, OPAMP is configured as voltage follower so Output must jump from 0.1V to 2.9V.

In the particular case of G4 OPAMP in normal mode the rate is 2.5V/us (minimum spec). So to move from 0.1V to 2.9V, the delta is 2.8V: 2.8 / 2.5 = 1.12 => it takes 1.12usec.

==> So switching time is slow enough compared to the OPAMP stabilization time.

Does this answer your question? Or may you still have other related ones?

-Amel

To give better visibility on the answered topics, please click on Accept as Solution on the reply which solved your issue or answered your question.

View solution in original post

4 REPLIES 4

Hi @Singh.Harjit ,

Sorry for delay to provide an answer, I was checking for answer with our Experts.

Unfortunately, We don’t have exact values to answer your question "how long does it take for the mux to switch to the new setting?". But you may consider that the switching time of the switch itself is few 10th of nsec to 100th nsec.

However, as the feedback loop of the OPAMP is broken by its nature, it needs some time to become a stable output (close the feedback loop).

This will be limited by the slew rate of OPAMP and condition of the circuit (for example capacitance in the feedback loop).

For example, if one of the GPIOs is 0.1V and another one is 2.9V, OPAMP is configured as voltage follower so Output must jump from 0.1V to 2.9V.

In the particular case of G4 OPAMP in normal mode the rate is 2.5V/us (minimum spec). So to move from 0.1V to 2.9V, the delta is 2.8V: 2.8 / 2.5 = 1.12 => it takes 1.12usec.

==> So switching time is slow enough compared to the OPAMP stabilization time.

Does this answer your question? Or may you still have other related ones?

-Amel

To give better visibility on the answered topics, please click on Accept as Solution on the reply which solved your issue or answered your question.

@Amel NASRI Thank you for the information. It answers my question.

No worries about the time it took to get an answer. I figured that since it wasn't documented, you must have reached out to the design or verification team to see if anyone knew the information and were waiting for a response. Appreciate that!

The information you provided is sufficient - that the switching time is less than 100ns and will be dominated by the opamp's slew rate/gain bandwidth product.

I have a project using stm32G474 nucleo board. The idea is "upconvter" sampling rate of the internal 4 fast - 15 msps DAC using OPAs with input MUX switching. So, 1-st OPA alternatively selects outputs of two DAC's - everything is driven synchronously by Timers chain network. 2-nd OPA - another pair of DACs. And last OPA selects output opa-1 or opa-2 . Switching glitches stop me from developing this into practical design.

Here is a picture:

Awesome data! Thank you for making time to share it and explain what you did and the issue.

In my application, the opamp is going into the ADC. So, I can adjust the sampling time so that the glitch has settled out.

In your application, I'm guessing you want a sample and hold so that you can use that to sample and hold the opamp output, switch the opamp's mux input, wait for the opamp to settle and then reconnect the sample and hold to pass through the new input.