Pith. sign in

REVIEW 3 major objections 6 minor 25 references

Screening Current-Induced Field and Field Drift Study in HTS coils using T-A homogenous model

T0 review · 3 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A homogenized T-A finite-element model makes real-time simulation of slow-ramping large-scale HTS coils feasible and predicts that a 20% transport-current overshoot eliminates central-field drift in an 800-turn coil.

desk verdict A useful case study of the homogeneous T-A model on an 800-turn coil, with credible speedups and a useful design hint about overshoot; the main gap is the missing benchmark, so the quantitative predictions should be treated with caution. read the letter →

arxiv 1908.06330 v1 pith:CBC6DKLZ submitted 2019-08-17 physics.app-ph

Edgar Berrospe-Juarez , Frederic Trillaud , Víctor M R Zermeño , Francesco Grilli This is my paper · ORCID
classification physics.app-ph
keywords T-Aformulationhomogeneousmodelscreeningcurrent-inducedfielddriftcurrentsweepreversalstriationHTScoilreal-timesimulation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that a homogenized T-A finite-element model can simulate screening-current-induced field (SCIF) and field drift in an 800-turn HTS solenoidal coil in about one hour per run on a desktop computer, making real-time simulation of slow ramping cycles feasible. It uses that speed to vary the critical current density $J_c$ and the $n$ power-law index, and shows that both SCIF and drift grow with $J_c$ and $n$, while lower values reduce them at the cost of tape performance. It also tests two remedies: striating tapes into narrow strips reduces SCIF, and a 20% transport-current overshoot eliminates central-field drift for this coil, where 5% and 10% overshoots only postpone it. The practical consequence is that drift mitigation must be evaluated coil by coil rather than assumed from literature values.

What carries the argument

The machinery is the T-A homogeneous formulation. The current vector potential $\mathbf{T}$ is defined only along the 1D HTS layers, or inside the homogeneous bulks obtained by replacing tape stacks, while the magnetic vector potential $\mathbf{A}$ is defined over the whole domain. Transport current enters through a Dirichlet boundary condition $I = (T_1 - T_2)\delta$, imposing a sheet current. Conduction follows the E-J power law $\rho = E_c/J_c(\mathbf{B})\,|\mathbf{J}/J_c(\mathbf{B})|^{n-1}$, with an anisotropic field-dependent $J_c$, so the tape-parallel field component that is neglected in the $\mathbf{T}$ equation still affects the local critical current density. Homogenization is the limit where inter-layer gaps vanish and the surrounding medium is insulating, collapsing the number of degrees of freedom so that 6-h and 30-h cycles simulate in about one hour.

What would settle it

Place Hall sensors at the center of the same 800-turn coil and record the central field through the triangular 26.5 A cycle and a 30 h plateau; if measured SCIF at peak departs from the predicted near-4.1% beyond sensor uncertainty, or if measurable drift reappears after the 20% overshoot, the homogeneous-model prediction fails for this coil.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that replacing each stack of HTS tapes in a solenoid by one homogenized bulk, with the current vector potential $\mathbf{T}$ defined only in the bulk and the magnetic vector potential $\mathbf{A}$ everywhere, brings computational cost low enough for real-time simulation of slow ramps while retaining the essential screening-current physics. For the 800-turn test coil (10 pancakes, 80 turns each) at 26.5 A, the model gives a central field of $B_{\mathrm{sim}} = 0.345$ T against a nominal uniform-current value of $B_n = 0.36$ T, a SCIF of about 4.1%. Systematically varying $J_{c0}$ and $n$ shows that lower $J_c$ and lower $n$ shrink SCIF and drift, that smaller $n$ rounds the hysteresis loop, and that the largest drift (2.2% over 30 h) appears at $n=10$. Two remedies are tested: striation of tapes into up to eight uncoupled strips reduces SCIF, and a 20% current overshoot eliminates drift for this coil while 5% and 10% overshoots only delay its reappearance.

Load-bearing premise

The load-bearing premise is that replacing the many thin tapes in each pancake by one solid block, with no current transfer between turns, still represents how screening currents penetrate; if homogenization distorts the current-front pattern, the predicted SCIF and drift timing would shift.

Editorial extensions

If this is right

  • For the studied 800-turn coil, a 20% transport-current overshoot removes central-field drift over the simulated 30 h plateau, while 5% and 10% overshoots only delay it.
  • The overshoot level needed to stabilize the field must be chosen per coil and per operating cycle; the paper explicitly notes that a 1% overshoot reported earlier is insufficient for this coil.
  • Striating tapes into 2, 4, or 8 uncoupled strips reduces SCIF because each strip develops its own current fronts.
  • Lowering $J_c$ and $n$ reduces SCIF and drift, but this trades away the high critical current and sharp transition that make 2G HTS tapes attractive for high-field magnets.
  • Because a 30-hour drift run computes in about one hour, the same approach can be used to screen operating cycles before energizing real magnets.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • An extension left implicit is that the speed of the homogeneous model could support a real-time control loop that adjusts the overshoot level during magnet operation, not just a pre-computed ramp.
  • The 20% overshoot threshold is likely sensitive to coil geometry, ramp rate, $J_c(\mathbf{B})$ anisotropy, and operating current; a useful extension would map the threshold across these parameters.
  • A direct experimental check would compare the predicted near-4.1% SCIF and log-linear drift against Hall-probe measurements on the same coil; the paper does not report such a comparison.
  • Because the simulated drift grows linearly in log time, longer plateaus could be extrapolated from the 30 h data, but that extrapolation is not made in the paper.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The manuscript presents a homogeneous T-A formulation for modeling HTS solenoid coils, in which stacks of HTS tapes are replaced by continuous bulk regions. The model is applied to a 10-pancake coil with 80 turns per pancake to study the screening current-induced field (SCIF) and central field drift as functions of the critical current density and the power-law index n. The paper also numerically tests two mitigation strategies: tape striation and current sweep reversal with overshoot. The main reported results include a 4.1% SCIF at peak current for the reference parameters, a maximum drift of 2.2%, and the claim that a 20% current overshoot eliminates the field drift. The authors further report that simulations take about one hour on a desktop computer and claim that this enables real-time simulation of slow ramping cycles of large-scale systems.

Significance. If the homogenized T-A model is quantitatively accurate, its computational speed advantage would be significant and would make systematic studies of large-scale HTS magnets feasible. The parametric investigation of Jc and n on SCIF and drift, and the numerical exploration of striation and overshoot, address questions of practical interest to magnet designers. However, the central quantitative claims rest on an unvalidated homogenization approximation and on simulation-only predictions; no comparison against a resolved-tape model, an H-formulation benchmark, or experimental data is provided for this coil. The significance is therefore conditional on an additional validation step.

major comments (3)
  1. [Section 2] The homogeneous-bulk approximation is presented without a benchmark against a resolved-turn model or against experimental data for this 800-turn coil. Replacing each 80-turn pancake by a continuous bulk removes the constraint that every insulated turn must carry the same transport current; only the total pancake current is imposed through Eq. (7). This can alter the radial distribution of transport current and the penetration of current fronts, which directly affects all quantitative outputs: the 4.1% SCIF in Section 4, the drift rates in Section 5, and the 20% overshoot recommendation in Section 6. The paper should add at least one validation case, for example comparing a single pancake against a fully resolved T-A or H-formulation model, or comparing the predicted SCIF against measured data for a similar coil.
  2. [Section 2, Eq. (6)] The T equation in Eq. (6) uses only the radial field component, ∂B_r/∂t, while the critical current density in Eq. (9) depends on both B_r and B_z. The manuscript acknowledges that the parallel-field contribution is neglected, but it does not quantify the resulting error for the analyzed geometry. In a solenoid, the field angle varies significantly with position, especially near the inner and outer radii, and the omitted term can change the local resistivity and, in turn, the dynamics of current front penetration. Please provide a quantitative estimate of the error (for example, by including the B_z term in a test case) or a physical justification for why this term is negligible here.
  3. [Sections 5 and 6] The field-drift and overshoot results are simulation-only; no comparison to experimental measurements or to an alternative numerical formulation is provided. The conclusion that a 20% overshoot 'eliminates' the field drift is based on the homogenized model within a 30-hour window, and the reappearance of drift after 5.5 hours for the 10% overshoot case indicates that the outcome is sensitive to the time horizon and to the model assumptions. The statement in Section 7 that H-formulation simulations 'would have required computation times in the order of months' is also unsupported by any runtime data for this problem. The authors should either provide a benchmark comparison or substantially weaken the claims, for example by clearly labeling the results as model predictions for the specific simulated conditions.
minor comments (6)
  1. [Title and throughout] The spelling 'homogenous' is used in the title while the text uses 'homogeneous'; please unify to a single spelling.
  2. [Section 6] The sentence 'These ramps are shown in figure' lacks a figure number; it should refer to Figure 10.
  3. [Section 4] The notation Jc0n is used without an explicit definition; please define it as the modified reference critical current density used in place of Jc0 in Eq. (9).
  4. [Tables 1 and 2] The relationship between the physical tape thickness and the thickness of the homogeneous bulk is not stated; please explain how the bulk dimensions and the HTS layer thickness δ (Table 2, Eq. (7)) are related.
  5. [Section 7] The 'real-time' claim in the abstract and conclusions would benefit from clarification: a 1-hour computation time for a 6-hour simulated cycle is faster than the cycle duration, but no extrapolation to 'large-scale systems' is demonstrated; please state the precise meaning of real-time used here.
  6. [Section 5] The sentence describing the case {Jc0, n=10} says it exhibits 'the larger drift' (2.2%) and then mentions two cases with 'the same 1.5% drift'; the wording is confusing and should be revised for clarity.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: SCIF and drift are computed outputs, not fitted inputs.

full rationale

No circular step is present. The T-A homogeneous model is stated through the governing equations (3)-(9), with the transport current imposed by Eq. (7) and the anisotropic critical-current law by Eq. (9). The parameters in Tables 1 and 2 are taken from tape data, not fitted to the target results. The screening-current-induced field B_SC is defined in Eq. (10) as the difference between the simulated field and the field from a uniform current distribution; it is a computed output, not an input. The field-drift curves and the effect of current overshoot are likewise obtained by time-dependent simulation and are not used to define any model parameter. The homogenization assumption is inherited from prior work by the authors (Refs. [15] and [18]), but the present paper restates the model equations explicitly and does not invoke that prior work to forbid alternatives or to force a particular numerical result. The lack of a resolved-tape benchmark for this specific 800-turn coil is a validity or accuracy concern about the homogenization approximation, not a circularity. The derivation chain is therefore self-contained with respect to circularity, and the score is 0.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central simulation outputs depend on the E-J power law and anisotropic Jc(B) parameters taken from tape data in Table 2, on the homogenization assumption in Section 2, and on the axisymmetric coil representation. No new physical entity is introduced. The listed free parameters are material and operating inputs, not parameters fitted to the predicted SCIF or drift targets.

free parameters (5)
  • Jc0 (reference critical current density) = 4.5e10 A/m^2
    Material input in Eq. (9) from Table 2; varied by factors 0.25 to 1.5 in Section 4. Not fitted to the paper's SCIF or drift outputs.
  • n (power-law index) = 25 (reference)
    E-J power-law exponent in Eq. (8); varied to 100, 25, 10, and 5. It controls the modeled flux-creep-like relaxation.
  • B0 = 0.03 T
    Field-scale parameter in the anisotropic Jc(B) relation in Eq. (9), taken from Table 2.
  • k = 0.2
    Anisotropy ratio in Eq. (9), taken from Table 2.
  • alpha = 0.6
    Exponent in Eq. (9), taken from Table 2.
assumptions (5)
  • domain assumption HTS resistivity follows the E-J power law with anisotropic field-dependent Jc, Eqs. (8)-(9).
    This is the constitutive model for all screening-current and drift dynamics. It is standard in HTS modeling but approximate.
  • domain assumption A stack of tapes can be represented by a homogeneous bulk, and the magnetic field component parallel to the tape surface is neglected in the T equation.
    Stated in Section 2. It removes inter-turn gaps and part of the field-angle dependence, with no validation against a non-homogenized model in this paper.
  • domain assumption The coil is adequately modeled as 2D axisymmetric with 1D HTS layers in the T-A formulation.
    Assumes azimuthal symmetry and neglects turn-transition and lead effects. Used throughout Sections 4 to 6.
  • domain assumption The transport current is imposed via Dirichlet boundary condition on T, Eq. (7), with zero-conductivity surroundings for the bulk.
    Defines how the coil current is injected and how strips are uncoupled in the striation model.
  • domain assumption The magnetic permeability is mu0 everywhere.
    Stated in Section 3; valid because no magnetic materials are present.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Screening Current-Induced Field and Field Drift Study in HTS coils using T-A homogenous model." pith.science (2026). https://pith.science/paper/CBC6DKLZ

@misc{pith2026190806330,
  author       = {Pith},
  title        = {Pith review of: Screening Current-Induced Field and Field Drift Study in HTS coils using T-A homogenous model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CBC6DKLZ}},
  note         = {Machine review of arXiv:1908.06330}
}
read the original abstract

The emergence of second generation (2G) high-temperature superconductor (HTS) tapes has favored the development of HTS magnets for their applications in areas such as NMR, MRI and high field magnets. The screening current-induced field and the field drift are two major problems hindering the use of HTS tapes in the mentioned areas. Both problems are caused by the screening current, then it is necessary to have a modeling strategy capable to estimate such phenomena. Thus far, the H formulation has been the most successful and used approach to model medium-size systems (hundreds of tapes). However, its application to large-scale systems is still impaired by excessive computation times and memory requirements. Homogenization and multi-scaling strategies have been successfully implemented to increase the computational efficiency. In this contribution, we show that using the homogenization technique with the recently developed T-A formulation allows reducing the computation time and the amount of memory up to the point that real-time simulations of slow ramping cycles of large-scale systems are possible. The T-A homogeneous model also allows systematically investigating the screening current using numerical simulations.

Figures

Figures reproduced from arXiv: 1908.06330 by the authors.

Figure 1
Figure 1. 2D axisymmetric representation of a solenoidal coil where the HTS layers are considered to be 1D lines. The T-A formulation combines the T and A formulations, T is defined just along the HTS lines. The homogenization process transforms the stacks of lines into homogeneous bulks. The homogeneous model assumes that a stack of HTS tapes can be represented by a homogeneous bulk, such process is depicted in figure 1. The… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 25 canonical work pages

  1. [1]

    Melhem Z 2012 High temperature superconductors (HTS) for energy applications (Cambridge, U.K.: Woodhead Publishing)

    work page 2012

  2. [2]

    and Yanagisawa, Y

    Maeda, H. and Yanagisawa, Y. (2014). Recent Developments in High -Temperature Superconducting Magnet Technology (Review). IEEE Transactions on Applied Superconductivity, 24(3), pp.1-12

    work page 2014

  3. [3]

    and Akachi, K

    Amemiya, N. and Akachi, K. (2008). Magnetic field generated by shielding current in high Tc superconducting coils for NMR magnets. Superconductor Science and Technology , 21(9), p.095001

    work page 2008

  4. [4]

    and Maeda, H

    Yanagisawa, Y., Kominato, Y., Nakagome, H., Hu, R., Takematsu, T., Takao, T., Uglietti, D., Kiyoshi, T., Takahashi, M. and Maeda, H. (2011). Magnitude of the Screening Field for YBCO Coils. IEEE Transactions on Applied Superconductivity, 21(3), pp.1640-1643

    work page 2011

  5. [5]

    and Noguchi, S

    Ueda, H., Fukuda, M., Hatanaka, K., Tao Wang, Ishiyama, A. and Noguchi, S. (2013). Spatial and Temporal Behavior of Magnetic Field Distribution Due to Shielding Current in HTS Coil for Cyclotron Application. IEEE Transactions on Applied Superconductivity , 23(3), pp.4100805-4100805

    work page 2013

  6. [6]

    and Larbalestier, D

    Weijers, H., Trociewitz, U., Markiewicz, W., Jiang, J., Myers, D., Hellstrom, E., Xu, A., Jaroszynski, J., Noyes, P., Viouchkov, Y. and Larbalestier, D. (2010). High Field Magnets With HTS Conductors. IEEE Transactions on Applied Superconductivity, 20(3), pp.576-582

    work page 2010

  7. [7]

    and Tixador, P

    Benkel, T., Miyoshi, Y., Escamez, G., Gonzales, D., Chaud, X., Badel, A. and Tixador, P. (2016). REBCO Performance at High Field With Low Incident Angle and Preliminary Tests for a 10-T Insert. IEEE Transactions on Applied Superconductivity, 26(3), pp.1-5

    work page 2016

  8. [8]

    and Tixador, P

    Benkel, T., Richel, N., Badel, A., Chaud, X., Lecrevisse, T., Borgnolutti, F., Fazilleau, P., Takahashi, K., Awaji, S. and Tixador, P. (2017). Preliminary Tests and Margin Estimate for a REBCO Insulated 10 T Insert Under High Magnetic Field. IEEE Transactions on Applied Superconductivity, 27(4), pp.1-5

    work page 2017

Show all 25 references
  1. [9]

    and Murphy, T

    Weijers, H., Markiewicz, W., Gavrilin, A., Voran, A., Viouchkov, Y., Gundlach, S., Noyes, P., Abraimov, D., Bai, H., Hannahs, S. and Murphy, T. (2016). Progress in the Development and Construction of a 32 -T Superconducting Magnet. IEEE Transactions on Applied Superconductivit...

    2016

  2. [10]

    and Weijer s, H

    Berrospe-Juarez, E., Zermeno, V., Trillaud, F., Gavrilin, A., Grilli, F., Abraimov, D., Hilton, D. and Weijer s, H. (2018). Estimation of Losses in the (RE)BCO Two -Coil Insert of the NHMFL 32 T All-Superconducting Magnet. IEEE Transactions on Applied Superconductivity, 28(3), pp.1-5

    2018

  3. [11]

    and Maeda, H

    Yanagisawa, Y., Xu, Y., Jin, X., Nakagome, H. and Maeda, H. (2015). Reduction of Screening Current-Induced Magnetic Field of REBCO Coils by the Use of Multi -Filamentary Tapes. IEEE Transactions on Applied Superconductivity, 25(3), pp.1-5

    2015

  4. [12]

    and Martini, L

    Brambilla, R., Grilli, F. and Martini, L. (2006). Development of an edge -element model f or AC loss computation of high -temperature superconductors. Superconductor Science and Technology, 20(1), pp.16-24

    2006

  5. [13]

    and Sørensen, M

    Zermeno, V., Abrahamsen, A., Mijatovic, N., Jensen, B. and Sørensen, M. (2013). Calculation of alternating current losses in stacks and coils made of second generation high temperature superconducting tapes for large scale applications. Journal of Applied Physics , 114(17), p.173901

    2013

  6. [14]

    and Grilli, F

    Quéval, L., Zermeño, V. and Grilli, F. (2016). Numerical models for ac loss calculation in large-scale applications of HTS coated conductors. Superconductor Science and Technology , 29(2), p.024007

    2016

  7. [15]

    and Grilli, F

    Berrospe-Juarez, E., Zermeño, V., Trillaud, F. and Grilli, F. (2018). Iterative multi -scale method for estimation of hysteresis losses and current density in large -scale HTS systems. Superconductor Science and Technology, 31(9), p.095002

    2018

  8. [16]

    and Yuan, W

    Zhang, H., Zhang, M. and Yuan, W. (2016). An efficient 3D finite element method model based on the T–A formulation for superconducting coated conductors. Superconductor Science and Technology, 30(2), p.024005

    2016

  9. [17]

    and Yuan, W

    Liang, F., Venuturumilli, S., Zhang, H., Zhang, M., Kvitkovic, J., Pamidi, S., Wang, Y. and Yuan, W. (2017). A finite element model for simulating second generation high temperature superconducting coils/stacks with large number of turns . Journal of Applied Physics , 122(4), p.043903

    2017

  10. [18]

    and Grilli, F

    Berrospe-Juarez, E., Zermeño, V., Trillaud, F. and Grilli, F. (2019). Real -time simulation of large-scale HTS systems: multi -scale and homogeneous models using the T –A formulation. Superconductor Science and Technology, 32(6), p.065003

    2019

  11. [19]

    and Jin, Z

    Wang, M., Li, Z., Sheng, J., Dong, F., Zhong, L., Duan, X., Song, M., Hong, Z. and Jin, Z. (2019). Study on the Screening Current Induced Field of a Novel Narrow HTS Tape with 1 mm Width. IEEE Transactions on Applied Superconductivity, 29(5), pp.1-5

    2019

  12. [20]

    and Oh, S

    Shin, H., Kim, K., Dizon, J., Kim, T., Ko, R. and Oh, S. (2005). The strain effect on critical current in YBCO coated conductors with different stabilizing layers. Superconductor Science and Technology, 18(12), pp.S364-S368

    2005

  13. [21]

    and Sirois, F

    Sass, F., Sotelo, G., Junior, R. and Sirois, F. (2015). H -formulation for simulating levitation forces acting on HTS bulks and stacks of 2G coated conductors. Superconductor Science and Technology, 28(12), p.125012

    2015

  14. [22]

    and Kiyoshi, T

    Uglietti, D., Yanagisawa, Y., Maeda, H. and Kiyoshi, T. (2010). Measurements of magnetic field induced by screening currents in YBCO solenoid coils. Superconductor Science and Technology, 23(11), p.115002

    2010

  15. [23]

    and Maeda, H

    Yanagisawa, Y., Nakagome, H., Ugliett i, D., Kiyoshi, T., Ruixin Hu, Takematsu, T., Takao, T., Takahashi, M. and Maeda, H. (2010). Effect of YBCO -Coil Shape on the Screening Current-Induced Magnetic Field Intensity. IEEE Transactions on Applied Superconductivity , 20(3), pp.744-747

    2010

  16. [24]

    and Lee, H

    Kim, K., Song, J., Yang, D., Kim, Y., Kim, T., Kim, S., Park, M. and Lee, H. (2016). Study on elimination of screening -current-induced field in pancake -type non -insulated HTS coil. Superconductor Science and Technology, 29(3), p.035009

    2016

  17. [25]

    and Trociewitz, U

    Zhang, M., Yuan, W., Hilton, D., Canassy, M. and Trociewitz, U. (2014). Study of second - generation high -temperature superconducting magnets: the self -field screening effect. Superconductor Science and Technology, 27(9), p.095010

    2014

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.