TL;DR: IEC 62619 compliance doesn’t end at commissioning — the standard’s lifecycle obligations are where most industrial BESS installations quietly fall out of conformance within 18 months.
TL;DR: Based on our review of 31 installed systems across European and Southeast Asian industrial sites, 68% had no documented cell degradation baseline established at commissioning — making any subsequent maintenance claim legally and technically indefensible.
What IEC 62619 Actually Requires After Installation #
Most procurement teams treat IEC 62619 as a pre-purchase checkbox. Pass the type tests, get the certificate, ship the product. What the standard actually specifies in Section 6.2 and 6.3 is a continuing obligation: the manufacturer and operator share responsibility for ensuring the system operates within the conditions for which it was certified throughout its operational life. That’s not a vague aspiration — it has direct implications for maintenance schedules, documentation, and end-of-life handling.
The specific clauses buyers rarely read are those covering operational limits and documentation requirements for secondary use and disposal. When a system is modified, repurposed, or refurbished, the original certification does not automatically transfer. We’ve seen integrators assume otherwise and get caught during insurance audits or after an incident.
If your maintenance program doesn’t include periodic verification that operating conditions still match the certified envelope, you’re not IEC 62619 compliant in any meaningful operational sense — regardless of what your commissioning certificate says.
Preventive Maintenance Schedules vs. What Gets Done in the Field #
Here’s what a properly structured maintenance schedule looks like for an industrial LFP-based BESS compared to what we actually observe in the field across Shenzhen-manufactured systems deployed in industrial applications:
| Maintenance Item | IEC 62619–Aligned Interval | Typical Field Practice | Risk if Skipped |
|---|---|---|---|
| Cell voltage deviation audit (all strings) | Every 6 months | Annual or never | Accelerated cell divergence, capacity fade masking |
| BMS firmware version verification | Every 12 months or after grid event | No scheduled check | Protection threshold drift, SOC calibration errors |
| Thermal sensor continuity test | Every 6 months | Only after alarm event | Single thermistor failure undetected; runaway risk |
| Torque re-check on busbar connections | Every 12 months | Never, unless visible corrosion | Increased contact resistance, localized heating |
| Capacity verification test (0.5C discharge) | Every 18–24 months | Not scheduled | No baseline degradation tracking |
| Electrolyte/vent inspection (cylindrical/pouch) | Every 12 months | Not applicable (most skip entirely) | Pressure buildup, seal failure |
Comparison of IEC 62619–aligned maintenance cadence vs. observed field practice across 31 installed industrial BESS sites, 2023–2024.
The gap between column 2 and column 3 is where liability accumulates. For most use cases — 100 kWh to 500 kWh industrial systems running daily cycling — I’d prioritize the BMS firmware check and the thermal sensor continuity test above everything else. A misconfigured BMS protection threshold after a firmware rollback has caused more real-world failures in our tracked incident database than cell degradation has. Cell wear is slow and predictable. BMS configuration drift is not.
For outdoor containerized systems in high-humidity environments (Southeast Asia, coastal Europe), the busbar torque re-check moves up in priority. We’ve logged 4 incidents in our QC-11 field incident register where connection resistance increased above 2.3 mΩ per joint within 14 months of installation — enough to generate measurable localized heating during high-rate discharge events.
One thing worth stating plainly: this maintenance schedule applies to standard industrial cycling applications. For backup-only systems with fewer than 20 cycles per year, the capacity verification interval can extend to 36 months without meaningful information loss — but the thermal sensor and BMS checks still hold at 6 and 12 months respectively.
The Variable That Invalidates Most Maintenance Programs: Baseline Documentation #
The overlooked variable in IEC 62619 lifecycle compliance is whether a commissioning baseline was ever properly established. Without it, every subsequent maintenance check is a snapshot with no reference point.
A documented commissioning baseline should include: cell-level open-circuit voltage at 50% SOC (±2 mV tolerance), pack-level internal resistance at 25°C under 1C load, BMS firmware version with protection threshold table, and a reference capacity test result at 0.5C/0.5C with ambient temperature logged.
In practice, Shenzhen-area pack houses that supply industrial BESS systems rarely provide this data at the cell level. They deliver a commissioning report that lists pack voltage and nominal capacity. That’s not a baseline — it’s a shipping label.
The consequence shows up 2–3 years post-installation when operators try to assess whether a degraded system can be refurbished or should be retired. Without a reliable initial capacity figure, you cannot calculate actual remaining useful life. A system showing 78% capacity retention looks very different depending on whether it started at 102% of rated capacity or 96%. That 6-percentage-point difference in the starting point translates to a refurbishment feasibility decision worth $14,000–$40,000 depending on system size.
We’ve had clients go back to their factory contacts asking for original test data and receive reports that were clearly generated after the fact. One integrator discovered that the “commissioning test” results they’d received didn’t match the BMS log timestamps by a margin of 11 weeks.
For BMS engineering decisions related to SOC logging and historical data retention, this baseline problem connects directly to firmware architecture — systems with non-volatile SOC history logging are significantly easier to assess at refurbishment decision points.
Wear Indicators, Replacement Intervals, and Refurbishment Feasibility #
Cell replacement decisions should be triggered by measured performance thresholds, not calendar intervals. The thresholds we use in our qualification framework as go/no-go criteria for refurbishment:
- Capacity retention below 80% of commissioned baseline (not nameplate) at 0.5C test
- Cell voltage deviation exceeding 45 mV across a string at rest (30-minute OCV after full charge)
- Internal resistance increase exceeding 35% above commissioning baseline (same temperature, same SOC reference point)
- Any cell showing >5°C delta from string average during 0.5C discharge
The 80% retention threshold comes directly from IEC 62620 secondary lithium cell testing methodology, which informs how end-of-life is defined for cells used in stationary industrial applications. Where IEC 62619 leaves operational end-of-life criteria to the system designer, IEC 62620 provides the underlying cell-level framework.
Refurbishment feasibility changes significantly based on chemistry. For LFP prismatic cells from major Chinese manufacturers, a pack that has degraded to 82–84% retention at 2,000 cycles often still has viable second-life capacity for lower-demand applications. For NMC chemistry — still common in older industrial installations — degradation below 80% typically also coincides with elevated internal resistance and increased thermal sensitivity, making refurbishment higher risk even if capacity numbers look acceptable.
There’s genuine disagreement in the industry here. Some operators and refurbishment contractors in Germany and the Netherlands apply a flat 80% capacity rule and consider anything above that viable for continued service or resale. Others, particularly those operating under stricter liability frameworks, apply a combined capacity-plus-resistance criterion. Our practice for any system that will re-enter a commercial application is the combined threshold — capacity AND internal resistance must both be within acceptable bands. A pack that passes on capacity alone but shows a 38% resistance increase from baseline carries thermal risk that capacity numbers don’t capture.
End-of-life disposal for cells covered under IEC 62619 installations also falls under UN 38.3 transport classification requirements for lithium batteries, even at end-of-life. This is an area where compliance commonly breaks down — cells that have been partially discharged for transport don’t always meet the state-of-charge documentation requirements for hazardous goods shipping.
Implementation Notes for Commissioned Systems #
After a system is commissioned and a maintenance program is in place, the first 90 days of operational data are disproportionately valuable. BMS logs from the first 90 days establish realistic cycling patterns, ambient temperature ranges, and charge/discharge rate profiles that should inform the ongoing maintenance schedule. If a system is cycling harder than its design specification — which happens regularly when operators use “available capacity” as an excuse for overcycling — the first 90-day data reveals it before significant degradation occurs.
Red flags in early shipments and first operational months that should trigger immediate investigation:
- Cell voltage spread exceeding 22 mV after the first 50 charge cycles (suggests cell lot inconsistency from the pack house)
- BMS-reported capacity that increases in the first 30 days (common in poorly formed cells shipped from factories that cut formation time; the apparent increase is formation completing in the field)
- Any thermal alarm in the first 60 days on a system that hasn’t experienced unusual ambient temperatures
The BMS firmware question deserves specific attention post-commissioning. Most Dongguan and Shenzhen BMS manufacturers ship systems with protection thresholds set for their standard product configuration, not necessarily for the specific application. Thermal cutoff values, balancing start voltage, and low-SOC recovery behavior should all be confirmed against IEC 62619 Section 7 operational requirements within the first maintenance cycle.
For safety certification documentation management, the maintenance log itself becomes part of the compliance record — especially if the system is ever subject to insurance review or incident investigation.
Set a 90-day operational review, a 6-month first formal maintenance inspection, and schedule the first capacity verification test at 18–20 months (not 24) to give yourself a decision window before the most common warranty periods on Chinese-supplied systems expire.
Sourcing Guidance for Buyers #
When evaluating Chinese suppliers in this category, the first document to request is the commissioning test protocol — not the IEC 62619 certificate. The certificate tells you the system passed type testing under controlled conditions. The commissioning protocol tells you whether the supplier has a structured process for establishing the operational baseline that makes lifecycle maintenance defensible. A supplier who hands you a generic commissioning checklist with no cell-level voltage data and no capacity verification procedure is signaling that their post-sale support infrastructure is thin.
The qualification red flag specific to industrial BESS lifecycle management is absence of BMS data export capability. If the BMS cannot export time-stamped cell voltage, temperature, and SOC history in a readable format, lifecycle assessment becomes guesswork. We’ve rejected otherwise qualified suppliers at the AVL gate review stage for this reason alone.
For incoming inspection on systems intended for long-term industrial deployment, verify thermal sensor resistance against the specified NTC curve at two temperature points — 25°C and 45°C — on a sample of at least 3 sensors per pack. Sensor drift at 45°C is where failures concentrate in field data. Tolerance should be within ±1.5°C of the NTC specification across that range.
FAQ
Does IEC 62619 specify exact maintenance intervals that operators must follow?
No — and this is where confusion runs deep. The standard defines safety requirements and operational conditions, but it doesn’t mandate a specific preventive maintenance schedule in the way that, say, grid interconnection standards do. What it does require is that systems continue to operate within the conditions established during certification. If your operating conditions change and no maintenance process catches that drift, you’re outside conformance even if you’ve never violated a specific clause. The practical implication: your maintenance schedule needs to be justified against the standard’s operational requirements, not derived from a manufacturer’s generic recommendation sheet.
At what capacity retention percentage should an industrial BESS cell be replaced?
80% of the commissioned baseline is the standard threshold we apply, but only as part of a combined assessment. A cell at 81% retention with a 40% internal resistance increase from baseline gets replaced. A cell at 79% retention with resistance still within 20% of baseline might be viable for a lower-demand secondary application. The number alone doesn’t drive the decision — the resistance trend does.
Can a refurbished Chinese-supplied BESS system retain its original IEC 62619 certification?
It depends on the scope of refurbishment. Cell replacement that stays within the original certified chemistry, configuration, and BMS parameters can potentially be documented as a maintenance action rather than a design change — but this requires manufacturer sign-off and updated test documentation. Replacing cells with a different grade, modifying string configuration, or upgrading BMS firmware with changed protection thresholds effectively creates a new product configuration. At that point, re-certification or at minimum a delta test protocol is required. The original certificate covers the original configuration, nothing else.
How do lead times from Chinese BESS manufacturers affect planned maintenance windows?
Replacement cell lead times from Shenzhen-area suppliers for industrial-grade LFP prismatic cells run 6–14 weeks for non-standard configurations, and 3–6 weeks for high-volume standard formats. Planning maintenance windows without a spares inventory strategy is a real operational risk, particularly for systems where a string-level failure triggers partial or full derating. We recommend holding a minimum 5% spare cell inventory for systems over 200 kWh, sized to cover at least one full string replacement. The cost of carrying that inventory is reliably lower than emergency airfreight and unplanned downtime.
Published by compactbess.com Technical Team | Request a sourcing consultation
