[Eeglablist] Persistent 20 Hz Harmonic Line Noise in EEG Data

Wed May 13 14:48:35 PDT 2026

I would follow all these great suggestions.
One thing that hasn’t been asked unless I missed it. Is your system recording other modalities simultaneously (GSR, PPG, fNIRS, etc.)?

If so, many PPG modules use PWM-driven LEDs pulsing at or near 20 Hz, and electrical crosstalk through shared grounding or circuitry can inject this directly into the EEG signal.

GSR/EDA devices typically use low-frequency AC excitation, so high-frequency interference is less common, but I’ve seen it with the Biosemi system at 16 Hz.

fNIRS modulates its light sources at specific frequencies (sometimes 20-100 Hz) for lock-in detection. If the driving electronics share a power supply or chassis with the EEG amplifier, that modulation frequency can appear as a narrow spectral peak.

If that’s the case, the cleanest diagnostic is to record EEG with each co-sensor powered on vs. off and check whether the artifact peak appears only when a given sensor is active.

Cedric

Sent from Proton Mail for iOS.

-------- Original Message --------
On Tuesday, 05/12/26 at 17:06 Rawls, Eric via eeglablist <eeglablist at sccn.ucsd.edu> wrote:
Is the 20Hz interference visible in the power spectrum of the raw data or does it only show up after some preprocessing? I have seen something like this occur with specific downsampling/filtering orders in older versions of EEGLAB but I do not recall the exact order-of-operations required to recreate it.

Best, E

Eric Rawls, Ph.D. (he/him/his)

Assistant Professor | Department of Psychology
Director | Brain, Data, and Causality (BDC) Lab
College of Science and Engineering
University of North Carolina Wilmington

Adjunct Assistant Professor | Department of Psychiatry and Behavioral Sciences
College of Medicine
University of Minnesota Twin Cities

Land Acknowledgment: I acknowledge that UNCW is located on the ancestral and traditional homelands of Indigenous peoples, including the Waccamaw Siouan and other tribes with enduring ties to this region such as the Lumbee, Coharie, Haliwa-Saponi, and Cherokee. I honor the ongoing presence, history, and stewardship of Native communities and offer this acknowledgment in a spirit of respect and relationship.

NOTICE: Emails sent and received in the course of university business are subject to the North Carolina Public Records Act (N.C.G.S. §132-1 et seq.) and may be released to the public unless an exception applies.

[cid:ca52e587-5639-4806-9792-0befc92fb683]<https://urldefense.com/v3/__https://outlook.office.com/bookwithme/user/82c92b80a22d45b0a310ddc81d7cf83c@uncw.edu?anonymous&ep=signature&ismsaljsauthenabled__;!!Mih3wA!AIhiGfU3uPoKEyYZGULe3B4YLszx9mBhQGd07e2ZK9SnCFUO7t1H_VXRrWWbSCO6Sxru1LMboIkoQD5astO1_g$ >
Book time to meet with me<https://urldefense.com/v3/__https://outlook.office.com/bookwithme/user/82c92b80a22d45b0a310ddc81d7cf83c@uncw.edu?anonymous&ep=signature&ismsaljsauthenabled__;!!Mih3wA!AIhiGfU3uPoKEyYZGULe3B4YLszx9mBhQGd07e2ZK9SnCFUO7t1H_VXRrWWbSCO6Sxru1LMboIkoQD5astO1_g$ >
________________________________
From: eeglablist <eeglablist-bounces at sccn.ucsd.edu> on behalf of Scott Makeig via eeglablist <eeglablist at sccn.ucsd.edu>
Sent: Tuesday, May 12, 2026 10:32 AM
To: Евгений Машеров <emasherov at yandex.ru>
Cc: eeglablist at sccn.ucsd.edu <eeglablist at sccn.ucsd.edu>
Subject: Re: [Eeglablist] Persistent 20 Hz Harmonic Line Noise in EEG Data

[This email originated from outside of UNCW]

Eugen,

Yes, thanks for correcting me - I was indeed reading from the tutorial
example(@ 25 Hz)  rather than the actual data (@ 20 Hz) ... I believe you
are correct that some nearby electronic source must have been the cause. I
believe it would be worthwhile to apply an initial ICA to see whether its
scalp projection (scalp map) is stable enough to be largely (or totally)
accounted by a single independent component, which then could be removed
neatly from the data -- or ignored in further analysis of the (brain)
independent source distribution.

Scott

On Mon, May 11, 2026 at 3:19 PM Евгений Машеров <emasherov at yandex.ru> wrote:

> If you're referring to this spectrum,
> https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=https*3A*2F*2Furldefense.com*2Fv3*2F__https*3A*2F*2Fibb.co*2F5WbNqYCW__*3B!!Mih3wA!CULZcfFkieH_bUE1w0wwnlMTADY94DKwrrOAJL4cnmw9hjrWFGypYhZLJUl1ZVoht5v4sXALi-8AvrBSXjma*24&data=05*7C02*7Crawlse*40uncw.edu*7C5687d65a15aa4e64d03908deb033b1dc*7C2213678197534c75af2868a078871ebf*7C0*7C0*7C639141933252982464*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=sYXFOzhWyOH2ACDQyJCVC3IVpqjKiAJvcXasU*2F9nytw*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUlJSU!!Mih3wA!AIhiGfU3uPoKEyYZGULe3B4YLszx9mBhQGd07e2ZK9SnCFUO7t1H_VXRrWWbSCO6Sxru1LMboIkoQD6kn9dfOw$ <https://urldefense.com/v3/__https://ibb.co/5WbNqYCW__;!!Mih3wA!CULZcfFkieH_bUE1w0wwnlMTADY94DKwrrOAJL4cnmw9hjrWFGypYhZLJUl1ZVoht5v4sXALi-8AvrBSXjma$>
> I see a peak at 20 Hz (around 30 dB), 40 Hz (around 40 dB), and 60 Hz
> (around 30 dB). Possible sources include MRI (pulse frequency) or
> physiotherapy equipment. It could also be some industrial equipment. In one
> case, the interference was from a nearby airport's radar.
>
> > The plots here clearly show the noise is at 25 Hz and its harmonics ...
> >
> > On Sat, May 9, 2026 at 11:08 AM Евгений Машеров via eeglablist <
> eeglablist at sccn.ucsd.edu> wrote:
> >
> >> This isn't power line interference. It would be 50 or 60 Hz, and
> possibly harmonics at multiples of 100-150-200... or 120-180-240-...
> frequencies. Subharmonics wouldn't appear. It's undoubtedly a technical
> artifact, although I'm unsure of the source. Railways sometimes use other
> frequencies (16 2/3 Hz, others—20, 25, 45, and others no longer in use)—but
> not 20 Hz. A television screen will show 50 (PAL/SECAM) or 60 (NTSC) Hz,
> computer monitors even higher, and it could be some obsolete or specialized
> video equipment (with a lower frame rate). I'd assume the source in your
> case is medical equipment (although I can't rule out, for example, a
> security alarm).
> >>
> >> The best course of action is to find the source of the interference and
> turn it off. However, this isn't always possible, especially with medical
> equipment. A similar situation occurred with an infusion pump in the
> intensive care unit: the intermittently activated drive solenoid created
> regular pulsed interference. A possible solution, also used to combat a
> physiological artifact—an electrocardiographic signal interfering with the
> EEG—is to evaluate the shape of the interference (the presence of harmonics
> indicates that the shape is clearly not sinusoidal) and subtract it from
> the signal. To do this, narrowband filtering isolates the fundamental 20 Hz
> harmonic, its peaks (or zero crossings) are used as a trigger for
> averaging, the averaged response is calculated for each channel, and then
> subtracted from the signals for the individual channels. Naturally, this is
> a post-processing method and is not suitable for online processing.
> >>
> >> Your truly
> >>
> >> Eugen Masherov
> >>
> >>> Dear EEGLAB List,
> >>
> >>>
> >>
> >>> I am currently dealing with persistent line-noise contamination in my
> EEG
> >>
> >>> recordings, specifically strong peaks around 20 Hz and its harmonics
> (40
> >>
> >>> Hz, 60 Hz, etc.). The contamination remains substantial even after
> multiple
> >>
> >>> preprocessing attempts.
> >>
> >>>
> >>
> >>> Raw Data PSD for reference:
> https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=https*3A*2F*2Furldefense.com*2Fv3*2F__https*3A*2F*2Fibb.co*2F5WbNqYCW__*3B!!Mih3wA!EfdhaFV_tpprT4Wc8_HcfK7KDVTXA0O0NPLW3Ws732nxP-TmPxHUXI_qXevcPIO8hfndJk8euI6BWbB1_dCuBxHG_A*24&data=05*7C02*7Crawlse*40uncw.edu*7C5687d65a15aa4e64d03908deb033b1dc*7C2213678197534c75af2868a078871ebf*7C0*7C0*7C639141933253242409*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=eIaQQjDBsZ5HLbBJlpYPu2JekmgTwMZu5OqKhVX6CsA*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUlJQ!!Mih3wA!AIhiGfU3uPoKEyYZGULe3B4YLszx9mBhQGd07e2ZK9SnCFUO7t1H_VXRrWWbSCO6Sxru1LMboIkoQD7L52136w$ <https://urldefense.com/v3/__https://ibb.co/5WbNqYCW__;!!Mih3wA!EfdhaFV_tpprT4Wc8_HcfK7KDVTXA0O0NPLW3Ws732nxP-TmPxHUXI_qXevcPIO8hfndJk8euI6BWbB1_dCuBxHG_A$>
> >>
> >>>
> >>
> >>> I have already tried:
> >>
> >>>
> >>
> >>> -
> >>
> >>>
> >>
> >>> Average re-referencing before notch filtering
> >>
> >>> -
> >>
> >>>
> >>
> >>> Average re-referencing after notch filtering
> >>
> >>> -
> >>
> >>>
> >>
> >>> Multiple full-rank average reference implementations, including:
> >>
> >>> -
> >>
> >>>
> >>
> >>> Makoto’s fullRankAveRef()
> >>
> >>> -
> >>
> >>>
> >>
> >>> Manual full-rank average reference using (nchan + 1) denominator
> >>
> >>>
> >>
> >>> Despite this, a large amount of the harmonic noise is still present in
> the
> >>
> >>> spectra.
> >>
> >>>
> >>
> >>> For reference, the issue looks somewhat similar to this discussion:
> >>
> >>>
> https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=https*3A*2F*2Furldefense.com*2Fv3*2F__https*3A*2F*2Fmne.discourse.group*2Ft*2Fline-noise-eeg*2F11817__*3B!!Mih3wA!EfdhaFV_tpprT4Wc8_HcfK7KDVTXA0O0NPLW3Ws732nxP-TmPxHUXI_qXevcPIO8hfndJk8euI6BWbB1_dB2ZPVDPg*24&data=05*7C02*7Crawlse*40uncw.edu*7C5687d65a15aa4e64d03908deb033b1dc*7C2213678197534c75af2868a078871ebf*7C0*7C0*7C639141933253274372*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=FJlsgfiAAdK1fqkeYAwd1vB8OyNv5OSzHhPdOJQVKAI*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUl!!Mih3wA!AIhiGfU3uPoKEyYZGULe3B4YLszx9mBhQGd07e2ZK9SnCFUO7t1H_VXRrWWbSCO6Sxru1LMboIkoQD4f1vGU-Q$ <https://urldefense.com/v3/__https://mne.discourse.group/t/line-noise-eeg/11817__;!!Mih3wA!EfdhaFV_tpprT4Wc8_HcfK7KDVTXA0O0NPLW3Ws732nxP-TmPxHUXI_qXevcPIO8hfndJk8euI6BWbB1_dB2ZPVDPg$>
> >>
> >>>
> >>
> >>> I would greatly appreciate advice on how people in the EEGLAB community
> >>
> >>> typically handle this kind of persistent harmonic contamination.
> >>
> >>>
> >>
> >>> Would you recommend CleanLine, Zapline-plus, spectrum interpolation,
> ASR,
> >>
> >>> ICA-based approaches, or something else?
> >>
> >>>
> >>
> >>> Thank you very much for your help.
> >>
> >>>
> >>
> >>> Best regards,
> >>
> >>> Maeghal Jain
> >>
> >>> BISE Lab
> >>
> >>> _______________________________________________
> >>
> >>> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu or visit
> https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=https*3A*2F*2Fsccn.ucsd.edu*2Fmailman*2Flistinfo*2Feeglablist&data=05*7C02*7Crawlse*40uncw.edu*7C5687d65a15aa4e64d03908deb033b1dc*7C2213678197534c75af2868a078871ebf*7C0*7C0*7C639141933253298283*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=lisqxIA*2B4OVEAiMrGK6cWHUvIUSghekFoxf6DSQHCm8*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUl!!Mih3wA!AIhiGfU3uPoKEyYZGULe3B4YLszx9mBhQGd07e2ZK9SnCFUO7t1H_VXRrWWbSCO6Sxru1LMboIkoQD6inTlV8w$ <https://sccn.ucsd.edu/mailman/listinfo/eeglablist  >  .
> >>
> >> _______________________________________________
> >>
> >> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu or visit
> https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=https*3A*2F*2Fsccn.ucsd.edu*2Fmailman*2Flistinfo*2Feeglablist&data=05*7C02*7Crawlse*40uncw.edu*7C5687d65a15aa4e64d03908deb033b1dc*7C2213678197534c75af2868a078871ebf*7C0*7C0*7C639141933253319451*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=0ZWknYHCiIXcVrW2*2FTmrgY2hufRp4KeeR81*2BCOtUB0w*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUlJQ!!Mih3wA!AIhiGfU3uPoKEyYZGULe3B4YLszx9mBhQGd07e2ZK9SnCFUO7t1H_VXRrWWbSCO6Sxru1LMboIkoQD7Ur3gyKA$ <https://sccn.ucsd.edu/mailman/listinfo/eeglablist  > .
> >
> > --
> > Scott Makeig, Research Scientist and Director, Swartz Center for
> Computational Neuroscience, Institute for Neural Computation, University of
> California San Diego, La Jolla CA 92093-0559, https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=http*3A*2F*2Fsccn.ucsd.edu*2F*scott&data=05*7C02*7Crawlse*40uncw.edu*7C5687d65a15aa4e64d03908deb033b1dc*7C2213678197534c75af2868a078871ebf*7C0*7C0*7C639141933253340446*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=0ireFT9mF795bN4bgEOg9SzzOiBzfbLhvBEBSl*2FXZKg*3D&reserved=0__;JSUlJX4lJSUlJSUlJSUlJSUlJSUlJSU!!Mih3wA!AIhiGfU3uPoKEyYZGULe3B4YLszx9mBhQGd07e2ZK9SnCFUO7t1H_VXRrWWbSCO6Sxru1LMboIkoQD7giQR85g$ <http://sccn.ucsd.edu/~scott  >
>

--
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0559, https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=http*3A*2F*2Fsccn.ucsd.edu*2F*scott&data=05*7C02*7Crawlse*40uncw.edu*7C5687d65a15aa4e64d03908deb033b1dc*7C2213678197534c75af2868a078871ebf*7C0*7C0*7C639141933253360104*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=AexTc5v9lgdNZw1jdpPGQgxXlcV6MHNOsEemQodk5bs*3D&reserved=0__;JSUlJX4lJSUlJSUlJSUlJSUlJSUlJQ!!Mih3wA!AIhiGfU3uPoKEyYZGULe3B4YLszx9mBhQGd07e2ZK9SnCFUO7t1H_VXRrWWbSCO6Sxru1LMboIkoQD7fiWqVIw$ <http://sccn.ucsd.edu/~scott  >
_______________________________________________
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu or visit https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=https*3A*2F*2Fsccn.ucsd.edu*2Fmailman*2Flistinfo*2Feeglablist&data=05*7C02*7Crawlse*40uncw.edu*7C5687d65a15aa4e64d03908deb033b1dc*7C2213678197534c75af2868a078871ebf*7C0*7C0*7C639141933253379386*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=rylAi2wjbaU*2FPmfgMeLbNrbHSVWdvNxNV62U*2FMZmkrQ*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUlJQ!!Mih3wA!AIhiGfU3uPoKEyYZGULe3B4YLszx9mBhQGd07e2ZK9SnCFUO7t1H_VXRrWWbSCO6Sxru1LMboIkoQD796RxNIg$ <https://sccn.ucsd.edu/mailman/listinfo/eeglablist  >.
_______________________________________________
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu or visit https://sccn.ucsd.edu/mailman/listinfo/eeglablist .