A DC isolation measurement and an alternating-excitation observation do not test the same physical question. The first examines an electrical path between specified terminals under its selected conditions. The second can create induced voltage around closed paths linked by changing magnetic flux.
Before comparing the results, draw both test boundaries. Identify terminals, included layers, possible returns and the mechanical state. The phrase packet insulation test is too broad to show which conductive region the instrument actually evaluates.
Read the DC result within its path
A low measured resistance can support the existence of a conductive path between the selected terminals. A high result may arise because that path is absent, outside the terminal boundary, open under the current pressure, or poorly observable under the chosen electrical condition.
Contact behavior may be nonlinear or history-dependent. A DC observation at one operating point should not be converted into an unrestricted contact resistance for all excitation conditions. The method must also avoid altering the contact in a way that invalidates the investigation.
| Evidence type | Quantity principally examined | Boundary it can miss |
|---|---|---|
| Defined DC isolation check | Terminal-path conduction under the tested state | Closed paths outside those terminals |
| Qualified alternating observation | Response to changing linked flux | Unexcited, concealed or weakly linked paths |
| Local thermal measurement | Accessible thermal consequence | Buried heat and radiometric blind regions |
| Controlled contact-state comparison | State dependence | Other untested load or movement histories |
Use the methods as complementary observations, with their coverage described, rather than competing labels of good or bad.
Read the AC response within its excitation
An induced-current loop requires a complete return and relevant linked flux. A conductive contact that has little linked changing flux in a particular test may show little heating. That observation does not prove the contact remains insignificant in every magnetic configuration.
Conversely, a measured alternating signal can include background pickup, capacitance or other field effects. Distinguish the candidate current path using calibrated phase, position and reference observations rather than interpreting every signal as conductive heating.
Preserve mechanical state across comparisons
If DC testing is performed after unclamping while the alternating observation came from the assembled packet, the two results describe different contact conditions. Record that difference explicitly. Pressure, debris and alignment can change whether a path exists.
Where practical within qualified procedures, compare measurements in reproducible states and investigate disagreements. A DC anomaly without a local thermal signal may require a different excitation or accessible region. A thermal anomaly with a normal DC result may require a different terminal boundary or another source explanation.
The result should be a coverage statement: which terminal paths, flux-linked loops and observation regions have been evaluated. That statement is more useful for repair acceptance than saying that one DC pass covers all alternating local heating behavior.
Compare a terminal path with a flux-linked loop
Catalogue thickness 0.23 mm [1] gives a 4.60 mm nominal separation across 20 sheet thicknesses. Assume a rectangular conductive loop spans that separation and 100 mm of sheet length, with a uniform normal sinusoidal field of 1.0 T peak at 50 Hz, total resistance 0.10 Ω and negligible inductance. Area is 460 mm²; Faraday induction gives Erms = 2πfBA/√2 = 0.1022 V and I²R heating of 0.1044 W. Halving the linked separation halves voltage and quarters heating. A DC check outside this loop can miss it. Draw the terminal boundary and verify flux orientation before assigning significance; the assumed circuit parameters require their own uncertainty [2]. Faraday induction supplies the voltage relation [3]. The assumed field is normal to this loop; main rolling-direction flux is not automatically linked.
Manufacturing controls. Burr bridges, coating damage, sheet spacing, clamp pressure and external returns establish the circuit. A conductive contact alone gives no heating estimate until the loop and linked changing flux are identified.
References
- Nippon Steel Corporation, Grain-Oriented Electrical Steel Sheets, catalogue D004je_03_202004, 2020, Specifications; Size and Form Tolerances.
- JCGM, JCGM 100:2008, Evaluation of measurement data — Guide to the expression of uncertainty in measurement, 2008.
- Zhao, X. et al., Magnetostrictive Properties of the Grain-Oriented Silicon Steel Sheet under DC-Biased and Multisinusoidal Magnetizations, Materials 12, 2156, 2019, sections 2.1 and 2.3; DOI: 10.3390/ma12132156.