In a secondary-battery cell production line, the winding process — layering and winding the cathode sheet, anode sheet, and separator together to form the electrode assembly — is the key section that governs both production speed and quality at once. The problem is that, during every acceleration and deceleration segment of the winding cycle, tension oscillates slightly because of the rotational inertia of idle rollers (rollers with no drive source that support the material and rotate along with it). Korean patent KR101899991B1 (“Tension control method for secondary battery manufacturing apparatus”) secured, as its claim, a method that factors in even this ‘moment of inertia of a single roller’ to correct tension.
What makes this tense is that although the patent was filed back in 2016, it remains Active, and follow-on patents from the Samsung SDI and LG Energy Solution families cite it as prior art. This piece answers the question of how far the familiar design element of winding tension control is locked up by this patent, and where design freedom remains.
The Basic Mechanism
The idea behind this patent can be compared to ‘timing the push of a swing.’ When you push a swing (the material), the chain it hangs from (the idle roller) absorbs or returns part of the pushing force during acceleration and deceleration, because of its own weight. If you ignore the chain’s inertia and push with a constant force, the swing sways at the start and stop of the motion — and in the same way, the material’s tension oscillates. This patent pre-loads a rotational-moment-of-inertia value calculated from the idle roller’s weight, radius, and thickness, and, in the segments where the winding line speed changes (acceleration/deceleration), preemptively corrects the tension output of the tension-adjustment device (such as a pneumatic regulator) by that value.
Comparison: Summary of Claim Elements
| Element | Role | Combined in Independent Claim (Claim 1)? |
|---|---|---|
| Step of loading idle-roller rotational-moment-of-inertia information | Calculating and storing an inertia value based on roller weight, radius, and thickness | Required |
| Step of loading winding-line-speed transition-segment (acceleration/deceleration) information | Identifying the initial acceleration segment and final deceleration segment of the winding cycle | Required |
| Step of adjusting the tension application amount | Correcting the tension-adjustment device’s output, in the transition segment, by the amount of tension change caused by inertia | Required |
| Measured feedback during the constant-speed segment (dependent claim 3) | Adjusting measured tension to the target tension during the constant-speed segment | Dependent (optional) |
| Pneumatic regulator pivot-angle control (dependent claims 5, 6) | The specific implementation method of the tension-adjustment device | Dependent (optional) |
| Incorporating the moment of inertia of the material’s own mass (dependent claims 8, 9) | Including the material’s own inertia in addition to that of the idle roller | Dependent (optional, second embodiment) |
Source: paraphrased by the editorial team from the claims of KR101899991B1 (Google Patents public record)
What this table shows: the core of independent claim 1 is a three-step combination — ‘① load roller inertia information → ② load acceleration/deceleration segment information → ③ preemptively correct tension in response to that inertia value.’ The specific structure of the tension-adjustment device (pneumatic regulator, dancer-type, and so on) and whether the material’s own inertia is factored in are both pushed down into dependent claims, so the core of the claim scope is designed narrower than it might first appear.
Source: reconstructed by the editorial team from the claims of KR101899991B1
Patent Perspective
Claim gist: The applicant is Pito Co., Ltd. (inventor Cho Seok-jin), and the family was built by claiming priority in China (CN201711225032.1A) and Japan (JP2017229564A) simultaneously with the domestic Korean filing on November 29, 2016 (KR1020160160444A). It was granted and published on September 18, 2018, is an Active patent expected to expire on November 29, 2036, and consists of 9 total claims (1 independent, 8 dependent).
Prosecution history: Among the six prior-cited patents per Google Patents are Samsung SDI’s KR101265196B1 (a secondary battery manufacturing apparatus, explicitly named as this patent’s own background art) and KR101914566B1 (electrode assembly winding apparatus and control method), indicating that this invention positioned itself as an improvement addressing the limitation of Samsung SDI’s existing winding tension-control technology — namely, that it did not account for moment of inertia. Among the twelve patents citing this one (cited-by) are LG Energy Solution’s electrode-assembly manufacturing apparatus patents (KR102690895B1, KR102430866B1) and a follow-on Samsung SDI patent (US12371291B2, granted 2025, “Winding apparatus capable of real-time precision tension measurement”), suggesting it may continue to be cited as one of the foundational technologies in this field.
Design-around points: Assessing literal infringement of this patent requires checking (1) whether the tension correction is in fact structured to pre-load and factor in a value for the ‘idle roller’s rotational moment of inertia,’ and (2) whether that correction is applied in response to the specific ‘acceleration/deceleration segment.’
Design-around scenario (rebuttal): An approach that corrects tension using only real-time PID feedback from a tension sensor (dancer-type or similar), instead of pre-calculating and loading a moment-of-inertia value, may avoid literal infringement because it lacks the ‘loading moment-of-inertia information’ element. However, if it ultimately produces the same tension-stabilization effect, it could become subject to a doctrine-of-equivalents analysis — this is stated only as a possibility requiring case-by-case review.
Limits and Outlook
Optimistically, the accumulation of patents built on such detailed physical models could accelerate the advancement of winding technology among domestic secondary-battery equipment makers. But there is a counter-scenario as well. As more small and mid-sized equipment makers, like Pito, densely secure patents on specific components or process units, large cell makers or new equipment entrants could face growing complexity, having to design around multiple detailed patents simultaneously even for a small change to the winder design.
As a strategic alternative, one option worth considering is to adopt, from the outset, a control architecture based on real-time multi-sensor feedback (simultaneously measuring tension, diameter, and speed) rather than a moment-of-inertia pre-loading approach when designing winding tension control — securing an independent path that is structurally distinct from this patent family.
Action: Equipment-design and IP staff are advised to first determine whether their own winding equipment’s tension-correction logic is ‘moment-of-inertia pre-loading type’ or ‘real-time sensor feedback type,’ and, if it is the former, to run an FTO screening that maps it 1:1 against claim 1 of KR101899991B1 and its related family members (CN201711225032.1A, JP2017229564A).
This patent analysis was prepared based on direct review of the original Google Patents record (patents.google.com/patent/KR101899991B1/ko) on September 10, 2026. Whether the patent registration is maintained, any infringement risk, and the effectiveness of any design-around are all presented only as possibilities, and this analysis does not substitute for definitive legal advice. Before applying any of this to an actual design, please obtain a formal FTO review from a patent attorney or IP law firm.