OPA388: Low flicker noise operational amplifiers / Is the OPA388 completely 1/f noise free?

dg.almeida

Part Number: OPA388
Other Parts Discussed in Thread: OPA2388, OPA376, OPA333, INA826, OPA320, INA821, OPA2380, OPA145, OPA1612

Hi everyone,

I'm developing an ECG test-bed for analog output front-ends as an academic project. The ECG signal is generated by a DAC and is attenuated with a resistive devider, at its output it presents a very low amplitude (on the orders of 100 uV to 5 mV). The divider's output must be buffered before entering the ECG front-end. To avoid a major decrease in SNR, this application requires opamps with very low noise figure (under 1 uVpp).

The ECG signal is mainly composed of very low frequency harmonics (from 0.5 Hz to 30 Hz). For the lowest possible noise figure (between 0.5 and 30 Hz) it seems that the OPA388 is the ideal operational amplifier.
The questions that I'm asking are:

The OPA388 is really 1/f noise free? In the datasheet there are no traces of flicker noise back to 1 Hz, does it occur at lower frequencies or does not occur at all?

Are there other operational amplifier options from TI which offer better noise figures, for this bandwidth (0.5-30 Hz)? While being RRIO or wide common-mode voltage input range for single supply operations.

Which are the models currently offered by TI which have no flicker noise (or very low figures) ?

I'm sorry for making all these questions, but it is difficult to find low flicker noise operational amplifiers, since there are no search filters for input voltage noise (between 0.1 Hz and 10 Hz), or low flicker noise, only input voltage noise density (at f = 1 kHz).

Kind regards,

Daniel Almeida

over 3 years ago

0 Michael Steffes over 3 years ago

Guru 47535 points

Hello Daniel,

Yes, that 1kHz metric comes from the audio world - odd, but pervasive

The OPA388 is a zero drift part - those have a sampling front end that elminate 1/f noise and popcorn -so narrow your search to those

There a few very lower input voltage noise audio amps that have very low 1/f corner - they might actually be lower than the 7nV than the OPA388 in the 1Hz region. OPA16xx series I recall - they will have current noise that most zero drift parts do not, so you need to watch the source impedance effects.

0 Raymond Zhang1 over 3 years ago

TI__Mastermind 48226 points

Hi Daniel,

I am enclosing our ECG/EKG application note and reference design.

https://www.ti.com/solution/electrocardiogram-ecg?keyMatch=ECG%20APPLICATION%20NOTE&tisearch=Search-EN-everything

https://www.ti.com/lit/ug/tidudo6b/tidudo6b.pdf?ts=1605641897823&ref_url=https%253A%252F%252Fwww.ti.com%252Ftool%252FTIDA-01580

https://www.ti.com/lit/ug/sbau181b/sbau181b.pdf?ts=1605659856883&ref_url=https%253A%252F%252Fwww.ti.com%252Fsitesearch%252Fdocs%252Funiversalsearch.tsp%253FsearchTerm%253Decg%2Bapplication%2Bnote

The OPA388 is really 1/f noise free?

Figure 6-17 is shown that voltage noise spectral density in 1/f noise at 1Hz to 1kHz range is approx. 6-7 nV/sqrt(Hz). It is not zero, but it is significantly lower than a typical 1/f noise spectral density.

Please also check OPA376, OPA333, OPA2388, OPA2380, INA826, INA821 OPA320 and OPA145 for ECG applications.

Please try our search tools below.

https://www.ti.com/amplifier-circuit/op-amps/cross-reference.html?keyMatch=OP%20AMP%20CROSS%20REFERENCE%20TOOLS&tisearch=Search-EN-everything

If you have additional questions, please let us know.

Best,

Raymond

0 dg.almeida over 3 years ago in reply to Michael Steffes

Prodigy 130 points

Hi Michael,

Thank you very much for your help,

It appears that there are several zero drift operational amplifiers from TI which may meet the design requirements. There are lots of options there.

What do you mean with "a sampling front end that eliminate 1/f and popcorn noise"? Is it something that is part of the opamp, or some external circuit configuration that is required? Is it some kind of data converter, does it require a reconstruction and/or anti-aliasing filter to eliminate spurious harmonics?

The OPA16xx series, at least the ones I have in mind, like the OPA1612, for example, which is a great opamp for audio, doesn't seem to fit the requirements, because, I forgot to mention, the system's supply voltage is 4.5 V.

Best regards,

Daniel Almeida

0 dg.almeida over 3 years ago in reply to Raymond Zhang1

Prodigy 130 points

Hi Raymond,

Thank you very much for your help.

I will follow your links and check the suggested operational amplifiers, I'm sure that there is at least one op amp that fits the design requirements.

The ECG application notes will be very helpful, thanks again.

Raymond Zhang1 said:

The OPA388 is really 1/f noise free?

Figure 6-17 is shown that voltage noise spectral density in 1/f noise at 1Hz to 1kHz range is approx. 6-7 nV/sqrt(Hz). It is not zero, but it is significantly lower than a typical 1/f noise spectral density.

I don't mean to be inconvinient, and I'm not correcting you, I don't have much experience, but that is not mainly from thermal + shot noise contribution? Rather than 1/f (flicker noise)?

UPDATE:

After a thorough analysis I concluded that the OPA388 is the more adequate operational amplifier for the application in cause.

Kind regards,

Daniel Almeida

0 Raymond Zhang1 over 3 years ago in reply to dg.almeida

TI__Mastermind 48226 points

Hi Daniel,

Enclosed is our op amp noise videos. Please take a look if you are interested how different noises are calculated. In addition, the voltage and current noises are modeled in the PSpice simulation in OPA388.

https://www.ti.com/product/OPA388?utm_source=google&utm_medium=cpc&utm_campaign=asc-amps-null-prodfolderdynamic-cpc-pf-google-wwe&utm_content=prodfolddynamic&ds_k=DYNAMIC+SEARCH+ADS&DCM=yes&gclid=CjwKCAiAzNj9BRBDEiwAPsL0dzROhwei1xwGKPeq68kL3ii6ZOj1kMew4gfJfRYsz0a0aZAv6TGuJBoC29IQAvD_BwE&gclsrc=aw.ds#design-development##design-tools-simulation

https://training.ti.com/ti-precision-labs-op-amps-noise-calculating-rms-noise?context=1139747-1139745-14685-1138803-13233

https://www.ti.com/tool/TINA-TI?keyMatch=TINA%20SOFTWARE%20DOWNLOAD&tisearch=Search-EN-everything

Best,

Raymond

0 dg.almeida over 3 years ago in reply to Raymond Zhang1

Prodigy 130 points

Hi Raymond,

Very interesting and helpful information, thank you very much. I've run some simulations, the noise is well modeled for the OPA388. The simulation and datsheet curves of the input voltagem noise spectral desity vs frequemcy are very similar in terms of average value. It provides a good foresight of the circuit's noise characteristics

Kind Regards,

Daniel Almeida

Amplifiers

Amplifiers forum

OPA388: Low flicker noise operational amplifiers / Is the OPA388 completely 1/f noise free?