REVIEW 4 major objections 4 minor 27 references
Seismic Isolation of Optical Tables Using Piezo Actuators
T0 review · 4 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read A piezo-actuated optical table suppresses ground motion by 21–36 dB between 0.5 and 1.3 Hz, leaving about 1 nm/√Hz residual horizontal motion at 1 Hz.
desk verdict Useful, honest engineering paper on a new piezo pre-isolator for optical tables, but the headline 21–36 dB suppression rests on a single witness seismometer and lacks independent confirmation. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing mechanism is the five-degree-of-freedom piezo pre-isolator: four vertical piezo stacks inside height-adjustment feet supply vertical, pitch, and roll, while four horizontal piezo stacks mounted on paired flexure stages supply X and Y translation, with the actuator geometry making the table stiff against rotation about the vertical axis. A broadband ground-motion sensor in the feedback loop senses table motion, a second sensor acts as an out-of-loop witness, and a third on the ground records the reference; the controller is a band-pass boost filter shaped as a ratio of two third-order band-pass filters with cutoff pairs (0.2, 3) Hz and (0.4, 1.4) Hz. This mechanism carries the argument by converting ground motion into a measurable open-loop gain and a control-on reduction in the displacement spectral density, with the tilt-coupling model used to explain the low-frequency transfer function shape.
What would settle it
Put an independent displacement sensor on the table, such as an optical interferometer measuring table position against a reference mass, and compare its 0.5–1.3 Hz spectrum with the table's control-on spectrum; if the independent readout does not show the same 21–36 dB suppression relative to ground, the claim is contaminated by sensor effects. A simpler check is a huddle test: place two sensors side by side on the table, difference them with control off, and see whether the fitted self-noise actually bounds the in-loop-versus-witness residual inside the control band.
Extended reading notes
Core claim
On its own terms, the paper claims that a five-degree-of-freedom table, actuated by piezo stacks and controlled by broadband ground-motion sensors, suppresses ground motion by 21 to 36 dB over 0.5 to 1.3 Hz along the optic axis, as measured by a witness sensor on the table, with the in-loop sensor suggesting even deeper suppression that is limited by sensor self-noise. The claim includes a specific performance point: residual horizontal motion of about $1~\mathrm{nm/\sqrt{Hz}}$ at 1 Hz, down from $20~\mathrm{nm/\sqrt{Hz}}$ with control off. The authors also report that this suppression reduces the broadband relative motion between the table and a suspended mirror by a factor of two, while noting that tilt-to-translation coupling and internal table resonances currently prevent extending control down to the 0.03 Hz target.
Load-bearing premise
The claimed suppression numbers come from sensors on the table, and the argument treats the measured sensor self-noise as the floor; if tilting of the table that the sensors register as horizontal motion, or any other unmodeled sensor noise, is present inside 0.5–1.3 Hz, then the 21–36 dB figures would overstate the true isolation.
Editorial extensions
If this is right
- Compact piezo pre-isolation can cut microseism-band table motion by one to two orders of magnitude, bringing auxiliary optical tables in gravitational-wave facilities toward the suppression of much larger active stages.
- Suspended mirrors seated on such tables see less seismic excitation at their suspension resonances; the paper measures a factor-of-two reduction in broadband relative motion between table and mirror in the control band.
- Because the fitted sensor self-noise at 1 Hz is roughly ten times the instrument specification, the demonstrated suppression is sensor-limited rather than actuator-limited, so quieter sensors should push the residual motion lower.
- Correcting tilt-to-translation coupling and damping the roughly 15 Hz table resonance should let the same hardware reach toward the 0.03 Hz target band, as the authors plan.
Reading between the lines
- If the $1~\mathrm{nm/\sqrt{Hz}}$ residual is genuine table motion, the same flexure-plus-piezo architecture could transfer to non-gravitational precision metrology, where the microseism band is often the dominant environmental noise.
- A clean test of the suppression claim would separate sensor from mechanical contributions: an independent optical readout of table position, or a second witness sensor at a different table location, would show whether the 21–36 dB numbers are true isolation or partly sensor correlation.
- Because the tilt-to-translation coupling is estimated from the measured transfer function rather than measured directly, a dedicated tilt channel could make that mechanism quantitative and possibly extend the control band without new actuators.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript describes an active seismic isolation table for auxiliary optics, using piezo actuators in five degrees of freedom (three translations and two tip-tilts) with a seismometer in the feedback loop. The authors report 21-36 dB suppression of ground motion between 0.5 and 1.3 Hz, with a residual horizontal motion of about 1 nm/sqrt(Hz) at 1 Hz, as measured by a witness seismometer on the table. They also discuss sensor self-noise, tilt-to-translation coupling, and table resonances as limitations that prevent control below 0.3 Hz, and they present a suspended-mass OSEM measurement that shows a smaller factor-of-two reduction in relative motion.
Significance. If the suppression claim is correct, this simple piezo-based pre-isolator would be a low-cost, easily manufactured option for reducing microseismic noise on auxiliary optical tables in gravitational wave detectors, with performance in the microseism band comparable to much more complex systems such as AEI-SAS. The paper is generally well organized and includes useful engineering details: measured open-loop transfer functions, a huddle-test self-noise fit, a noise budget (Figure 8), and an explicit discussion of tilt coupling and resonance limitations. However, the central quantitative claim rests on a single sensor type (the witness seismometer) and lacks independent mechanical verification, statistical uncertainty, and a clear statement of whether the residual is sensor-noise-limited.
major comments (4)
- [Section III, Figures 8 and 9] The OSEM measurement of the suspended mirror relative to the table shows only a factor of two (~6 dB) reduction in the control band, whereas Figure 8 claims 21-36 dB suppression of the table itself. The attribution of this discrepancy to acoustic or suspension-related noise is plausible but is not supported by any quantitative analysis. Because the OSEM is the only independent mechanical witness of table motion in the paper, the headline suppression figure is not yet corroborated. Please provide a second independent measurement of table translation or a quantitative noise budget for the OSEM that demonstrates the consistency of a 6 dB relative-motion reduction with a 21-36 dB table-motion suppression.
- [Section III, Figure 8 and surrounding text] The witness seismometer spectrum lies close to the fitted self-noise curve, and the 'Predicted Witness Seismometer' curve is the quadrature sum of self-noise and the seismometer response to predicted table motion. Consequently, the claimed 1 nm/sqrt(Hz) residual at 1 Hz is an interpretation that depends on the self-noise fit, not a direct measurement. The text should state explicitly whether this residual is an upper limit or a best estimate, and should give an uncertainty that includes sensor self-noise and tilt-to-translation coupling. If the witness were self-noise-limited, the suppression ratio might underestimate the true table suppression, so the direction of the bias should be quantified.
- [Section III, experimental methodology] The quantitative claims (21-36 dB and 1 nm/sqrt(Hz)) are presented without error bars or repeated trials. The suppression ratio is the ratio of ground to witness seismometer spectra and therefore depends on the relative calibration of the two instruments, but the calibration procedure is not described. Please add error bars from multiple measurements or from the spectral estimation method, and describe the relative calibration and its uncertainty.
- [Section III, control band statement] The text states 'This informed our target of 40 dB suppression in the control band' and the introduction gives a target band of 0.03-3 Hz, but the demonstrated performance is 21-36 dB over 0.5-1.3 Hz. The paper should clearly acknowledge that the target was not met and separate the achieved band from the target band in the abstract and conclusions to avoid overstatement.
minor comments (4)
- [Abstract] There is a LaTeX typo in the abstract: '1 $\mathrm{\mathrm{nm/\sqrt{Hz}}}$' has a double math-mode command; it should read '1 $\mathrm{nm/\sqrt{Hz}}$'.
- [Section III] The word 'butter-worth' should be capitalized as 'Butterworth' when referring to the filter.
- [Section II, Table I and text] The table lists table mass as 20 kg and frame mass as 10 kg, but the vertical actuator text says 'Each piezo supports a load of 180 kg (for 720 kg total)'. Clarify what the 720 kg includes and how the load path through the feet relates to the table and frame masses.
- [Figure 8] The self-noise curve is labeled as a fit; the fit parameters or the measurement method should be described in the text or caption so the reader can judge the reliability of the noise floor.
Circularity Check
No significant circularity: the 21-36 dB suppression is a direct ground-to-witness seismometer measurement, and the fitted transfer-function and self-noise models are not load-bearing for that claim.
full rationale
The central claim is an experimental result: the suppression ratio is read directly from a ground T120 and an out-of-loop witness T120 (Section III, Figure 8). No fitted parameter enters that ratio. The 'Predicted Witness Seismometer' curve in Figure 8 is an explicit noise budget built from the huddle-test self-noise fit and measured transfer functions, and it is used only to explain why the witness lies above the in-loop sensor; it does not define the suppression. The tilt-coupling model in Figure 6 is an empirical fit to the same transfer function it describes, but the paper openly states that the coupling was estimated from that measured transfer function and does not use the model to generate the headline suppression. There are no load-bearing self-citations, imported uniqueness theorems, or ansatz-by-citation steps. Concern that the witness seismometer may be self-noise limited or tilt-contaminated is an experimental validity issue, not a circularity issue. The derivation chain is therefore self-contained.
Assumptions & free parameters
free parameters (4)
- Table first resonance frequency and Q factor =
~15 Hz, Q ~70
- Tilt-to-translation coupling coefficient =
roughly 0.2 urad tilt per 20 um translation, about 1e-2 rad/m
- Seismometer self-noise fit =
curve shown in Figure 8, about 10x the instrument specification at 1 Hz
- Controller band-pass filter cutoffs =
(0.2, 3) Hz and (0.4, 1.4) Hz
assumptions (5)
- domain assumption Seismometers are calibrated inertial sensors with negligible cross-coupling in the 0.03 to 3 Hz control band.
- standard math Feedback control can be modeled as a linear time-invariant plant within the actuation range.
- domain assumption Ground motion is the dominant disturbance and the ground seismometer is an adequate reference.
- domain assumption The tilt-to-translation coupling is constant and can be described by a single fitted coefficient over the band of interest.
- domain assumption Table internal modes can be represented by fitted resonances that remain stable during the measurement.
Cite this review
Pith. "Pith review of Seismic Isolation of Optical Tables Using Piezo Actuators." pith.science (2026). https://pith.science/paper/5R5PUIM2
@misc{pith2026250513083,
author = {Pith},
title = {Pith review of: Seismic Isolation of Optical Tables Using Piezo Actuators},
year = {2026},
howpublished = {\url{https://pith.science/paper/5R5PUIM2}},
note = {Machine review of arXiv:2505.13083}
}
abstract
Seismic isolation is crucial for gravitational wave detectors as it minimizes ground vibrations, enabling the detection of faint gravitational wave signals. An active seismic isolation platform for precision measurement experiments is described. The table features piezo actuation along five degrees of freedom: three translational actuations and two tip-tilt degrees of freedom along the horizontal axes. It is stiff in rotation about the vertical axes. A seismometer is used to sense table motion. Piezo actuators are used to suppress seismic noise with feedback control bandwidth of 0.3 to 3 Hz. Suppression levels ranging from 21 to 36 dB of seismic noise within the frequency range of 0.5 to 1.3 Hz are demonstrated, as measured by a witness seismometer on the table, with the suppression direction along the axis of the longitudinal translation of the suspended mirror on the table. The suppression results in 1 $\mathrm{\mathrm{nm/\sqrt{Hz}}}$ residual horizontal motion at 1 Hz. Limitations such as tilt-to-translation coupling that prevent actuation over the desired range of 0.03 to 3 Hz are discussed.
Figures
Figures from the paper (5 more)
Reference graph
Works this paper leans on
-
[1]
author author The LIGO Scientific Collaboration ,\ @noop journal journal Classical and Quantum G ravity \ volume 32 ,\ pages 074001 ( year 2015 ) NoStop
work page 2015
-
[2]
author author F. Acernese , author M. Agathos , author K. Agatsuma , author D. Aisa , author N. Allemandou , author A. Allocca , author J. Amarni , author P. Astone , author G. Balestri , author G. Ballardin , et al. ,\ @noop journal journal Classical and Quantum G ravity \ volume 32 ,\ pages 024001 ( year 2015 ) NoStop
work page 2015
-
[3]
author author Y. Aso , author Y. Michimura , author K. Somiya , author M. Ando , author O. Miyakawa , author T. Sekiguchi , author D. Tatsumi , \ and\ author H. Yamamoto ( collaboration The KAGRA Collaboration ),\ 10.1103/PhysRevD.88.043007 journal journal Phys. Rev. D \ volume 88 ,\ pages 043007 ( year 2013 ) NoStop
-
[4]
author author D. V. \ Martynov , author E. D. \ Hall , et al. ,\ @noop journal journal Phys. Rev. D \ volume 93 ,\ pages 112004 ( year 2016 ) NoStop
work page 2016
-
[5]
author author A. Buikema , author C. Cahillane , author G. Mansell , author C. Blair , author R. Abbott , author C. Adams , author R. Adhikari , author A. Ananyeva , author S. Appert , author K. Arai , et al. ,\ @noop journal journal Physical Review D \ volume 102 ,\ pages 062003 ( year 2020 ) NoStop
work page 2020
-
[6]
author author P. Nguyen , author R. Schofield , author A. Effler , author C. Austin , author V. Adya , author M. Ball , author S. Banagiri , author K. Banowetz , author C. Billman , author C. Blair , et al. ,\ @noop journal journal Classical and Quantum Gravity \ volume 38 ,\ pages 145001 ( year 2021 ) NoStop
work page 2021
-
[7]
author author E. Chin , author J. Dumas , author C. Zhao , author L. Ju , \ and\ author D. Blair ,\ 10.1088/1742-6596/32/1/018 journal journal Journal of Physics: Conference Series \ volume 32 ,\ pages 111 ( year 2006 ) NoStop
-
[8]
author author P. Barriga , author J. C Dumas , author A. A Woolley , author C. Zhao , \ and\ author D. G Blair ,\ booktitle booktitle The Review of scientific instruments ,\ @noop journal journal Review of Scientific Instruments \ volume 80 ,\ pages 114501 ( year 2009 ) NoStop
work page 2009
Show all 27 references
-
[9]
Matichard , author B
author author F. Matichard , author B. Lantz , author R. Mittleman , author K. Mason , author J. Kissel , author B. Abbott , author S. Biscans , author J. McIver , author R. Abbott , author S. Abbott , et al. ,\ @noop journal journal Classical and Quantum Gravity \ volume 32 ,...
2015
-
[10]
Blom , author M
author author M. Blom , author M. Beker , author A. Bertolini , author J. van den Brand , author H. Bulten , author E. Hennes , author F. Mul , author D. Rabeling , \ and\ author A. Schimmel ,\ @noop journal journal Nuclear Instruments and Methods in Physics Research Section A...
2013
-
[11]
Wanner , author G
author author A. Wanner , author G. Bergmann , author A. Bertolini , author T. Fricke , author H. L \"u ck , author C. Mow-Lowry , author K. Strain , author S. Gossler , \ and\ author K. Danzmann ,\ @noop journal journal Classical and quantum gravity \ volume 29 ,\ pages 24500...
2012
-
[12]
Kirchhoff ,\ title Implementation of an active seismic isolation system for the AEI 10 m prototype ,\ 10.15488/11634 Ph.D
author author R. Kirchhoff ,\ title Implementation of an active seismic isolation system for the AEI 10 m prototype ,\ 10.15488/11634 Ph.D. thesis ,\ school Leibniz U., Hannover ( year 2022 ) NoStop
2022 doi
-
[13]
Wen , author R
author author S. Wen , author R. Mittleman , author K. Mason , author J. Giaime , author R. Abbott , author J. Kern , author B. OʼReilly , author R. Bork , author M. Hammond , author C. Hardham , et al. ,\ @noop journal journal Classical and Quantum Gravity \ volume 31 ,\ page...
2014
-
[14]
Kissel \ and\ author B
author author J. Kissel \ and\ author B. Lantz ,\ @noop title Enhanced LIGO HAM ISI P rototype P reliminary P erformance R eview , \ type Tech. Rep. \ ( institution T-080251-01 ) NoStop
-
[15]
Aston , author M
author author S. Aston , author M. Barton , author A. Bell , author N. Beveridge , author B. Bland , author A. Brummitt , author G. Cagnoli , author C. Cantley , author L. Carbone , author A. Cumming , et al. ,\ @noop journal journal Classical and Quantum Gravity \ volume 29 ,...
2012
-
[16]
Aasi , author B
author author J. Aasi , author B. Abbott , author R. Abbott , author T. Abbott , author M. Abernathy , author K. Ackley , author C. Adams , author T. Adams , author P. Addesso , author R. Adhikari , et al. ,\ @noop journal journal Classical and quantum gravity \ volume 32 ,\ p...
2015
-
[17]
Abbott , author R
author author R. Abbott , author R. Adhikari , author G. Allen , author D. Baglino , author C. Campbell , author D. Coyne , author E. Daw , author D. DeBra , author J. Faludi , author P. Fritschel , et al. ,\ @noop journal journal Classical and Quantum Gravity \ volume 21 ,\ p...
2004
-
[18]
Blom , author M
author author M. Blom , author M. Beker , author A. Bertolini , author J. Van Den Brand , author H. Bulten , author M. Doets , author E. Hennes , author F. Mul , author D. Rabeling , \ and\ author A. Schimmel ,\ @noop journal journal Physics Procedia \ volume 61 ,\ pages 641 (...
2015
-
[19]
Blom ,\ title Seismic Attenuation for Advanced Virgo : Vibration Isolation for the External Injection Bench ,\ @noop Ph.D
author author M. Blom ,\ title Seismic Attenuation for Advanced Virgo : Vibration Isolation for the External Injection Bench ,\ @noop Ph.D. thesis ,\ school Vrije U., Amsterdam ( year 2015 ) NoStop
2015
-
[20]
author author Thorlabs ,\ https://www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=1095 title Optical T able S upports: A ctive V ibration I solation , \ ( year 2018 ) NoStop
2018
-
[21]
DeSalvo , author J
author author G. DeSalvo , author J. Swanson , \ and\ author S. A. \ Systems ,\ https://books.google.com.sg/books?id=Q3UBSwAACAAJ title ANSYS Engineering Analysis System User's Manual ,\ series ANSYS Engineering Analysis System User's Manual \ No.\ number v. 1 \ ( publisher Sw...
1985
-
[22]
Zhao , author L
author author C. Zhao , author L. Ju , author Q. Fang , author C. Blair , author J. Qin , author D. Blair , author J. Degallaix , \ and\ author H. Yamamoto ,\ @noop journal journal Physical Review D \ volume 91 ,\ pages 092001 ( year 2015 ) NoStop
2015
-
[23]
Jaberian Hamedan , author A
author author V. Jaberian Hamedan , author A. Adam , author C. Blair , author L. Ju , \ and\ author C. Zhao ,\ @noop journal journal Applied Physics Letters \ volume 122 ( year 2023 ) NoStop
2023
-
[24]
Satari , author C
author author H. Satari , author C. Blair , author L. Ju , author D. Blair , author C. Zhao , author E. Saygin , author P. Meyers , \ and\ author D. Lumley ,\ @noop journal journal Classical and Quantum Gravity \ volume 39 ,\ pages 215015 ( year 2022 ) NoStop
2022
-
[25]
Bork , author R
author author R. Bork , author R. Abbott , author D. Barker , \ and\ author J. Heefner ,\ https://arxiv.org/abs/physics/0111077 title An overview of the ligo control and data acquisition system , \ ( year 2001 ),\ http://arxiv.org/abs/physics/0111077 arXiv:physics/0111077 [phy...
2001 arXiv
-
[26]
author author B. P. \ Abbott et al. ( collaboration LIGO Scientific ),\ 10.1088/0034-4885/72/7/076901 journal journal Rept. Prog. Phys. \ volume 72 ,\ pages 076901 ( year 2009 a ) ,\ http://arxiv.org/abs/0711.3041 arXiv:0711.3041 [gr-qc] NoStop
2009 arXiv
-
[27]
Abbott , author M
author author R. Abbott , author M. Barton , author B. Bland , author B. Moore , author C. Osthelder , \ and\ author J. Romie ,\ https://dcc.ligo.org/LIGO-T0900286/public journal journal LIGO Document Control Center \ volume 900286 ,\ pages 2009 ( year 2009 b ) NoStop
2009
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.