A powerbank has 3.7V lithium ion battery inside which is buck boost converted to 5V. What caused buck boost converter to produce these noise?




Since a differential amplifier cancel all common mode noise. Can all such noise in the Powerbank be canceled when used in a differential amplifier? if not, what noise would remain?

 

In general, no, you can't eliminate all noise from a system.

You give no indication of what either the vertical or horizontal scales of your plot are, which makes it pretty useless. You show two vertical divisions. Is the scale is 5 µV/div, you already have a very low noise system. If it is 5 V/div, you have an incredibly noisy system. We aren't mind readers, so there's no way for us to guess.

Switching converters are noisy. Switching anything causes noise. Careful design can reduce it, filtering can reduce it further, other techniques can reduce it even more. The question comes down to how low does the noise have to be before it is acceptable and how much time, effort, and money are you willing to devote to getting there.

As for a differential amplifier, no, it does not cancel all common mode noise, it simply does a pretty good (not perfect) job of rejecting it, within limits.

But how are you going to use this powerbank in a differential amplifier and at what point are you concerned about the noise?

You REALLY need to provide more details of what you are trying to achieve and how you are attempting to do it.

 

Voltage ripple due to switching will be correlated to the switching frequency and can be reduced to very low levels with appropriate design techniques. The plot above does not look like typically observed voltage ripple, but without some sense of magnitude, what is being measured, and the time base being used it is hard to draw appropriate conclusions.

 

In general, no, you can't eliminate all noise from a system.

You give no indication of what either the vertical or horizontal scales of your plot are, which makes it pretty useless. You show two vertical divisions. Is the scale is 5 µV/div, you already have a very low noise system. If it is 5 V/div, you have an incredibly noisy system. We aren't mind readers, so there's no way for us to guess.

Switching converters are noisy. Switching anything causes noise. Careful design can reduce it, filtering can reduce it further, other techniques can reduce it even more. The question comes down to how low does the noise have to be before it is acceptable and how much time, effort, and money are you willing to devote to getting there.

As for a differential amplifier, no, it does not cancel all common mode noise, it simply does a pretty good (not perfect) job of rejecting it, within limits.

But how are you going to use this powerbank in a differential amplifier and at what point are you concerned about the noise?

You REALLY need to provide more details of what you are trying to achieve and how you are attempting to do it.

Above is the complete scale. It is from a Xiaomi 20000 powerbank I got a year ago because someone said it had very low noise.. I didn't show the scale because my question is in general. And I just ordered the Anker Powercore Essential 20000 PD a few hours ago to see if the same noises are present. Note the 2 channel noises in the Xiaomi is similar in 2 electrodes, ch 15 and ch 16.. about 50uV. And they can be canceled when I deduct ch 15 - ch 16 resulting in 1uV noise. So when you guys said not all noises can be canceled. What does it mean? And what exactly noises are retained that I can't see? below is the scale legend.

 

The above is output from the EEG screen with the electrodes not connected to skin. It doesn't measure the voltage of the powerbank like in oscilloscope. But the noise is cancelled. However, since it is not direct noise from the powerbank, in what way can the powerbank 3.7V to 5V buck boost converter ripples affect the EEG overall like does it make the final differential output ripples? can you give an example of the ripples that can affect the output that differential amplifiers can't eliminate?

 

So, originally, you said this noise came from a boost converter, and now you tell us it comes from an EEG circuit with the electrodes floating. Were you purposely misleading us, or do you just not know any better?

 

So, originally, you said this noise came from a boost converter, and now you tell us it comes from an EEG circuit with the electrodes floating. Were you purposely misleading us, or do you just not know any better?

Both are true. I used Anker Powercore Essential 20000 PD (Anker 525 powerbank) to power the g.USBamp with the g.GAMMAbox active electrode driver box. The active electrodes are floating, but it produced the above noise. However, when I deduct channel 1 - channel 2. the noise is gone. That why I was asking whether all buck booster noise in powerbank can be cancelled.. or somehow the ripples in the powerbank can still somehow affect it (how?).

 

In post #2, WBahn is correct on every point!!
Is this another medical researcher with a lack of technical insight struggling to duplicate claimed results from others research, and not having much success??
Certainly most switching supply noise can be filtered out, BUT that is usually a multiple section filter, not simple nor cheap. AND, there is always some DC power loss, and so efficiency suffers a bit.

 

I was asking whether all buck booster noise in powerbank can be cancelled.

The degree of cancellation depends at least in part on the Power Supply Rejection Ratio of the amplifier or other circuitry.

 

It is not likely that any practical filter can remove 100% of all noise! Filters are generally rated by the amount of noise reduction that they can provide, often including a frequency range for the noise spectrum removed. In addition, most amplifier circuits provide power supply noise rejection intrinsic to the amplification process. (PSRR, Power Supply Rejection Ratio).
The ultimate question is what level of noise in the amplifier output will have no effect??