Battery pack prototype acceptance criteria for OEM buyers should be agreed before the sample is built. Define what the sample must demonstrate, its actual configuration, the evaluation conditions and the evidence that will support a decision. “Works in our machine” is too ambiguous to serve as the entire acceptance rule.
A written sample agreement helps purchasing, engineering and the supplier make the same decision from the same results. It also distinguishes permission to continue development from approval to release a production design.
State the purpose of this sample
Identify whether the delivery is for mechanical fit, electrical integration or evaluation of a production-intent design. A fit sample with temporary internals may answer an access question while providing no evidence about final runtime. List every intended use and every known deviation.
Record the sample ID, pack specification revision, relevant component identities, BMS hardware and firmware, connector assembly and charger or host equipment. The required detail should let the teams distinguish the tested build from a later revision.
Use Customization to discuss the sample scope and battery requirements before the order treats an incomplete brief as a finished specification.
Write acceptance criteria as decision rules
Each evaluation item needs a requirement, method, conditions, pass/fail criterion, evidence record and approval owner. NASA’s product verification guidance connects a requirements baseline with acceptance criteria and recorded results. This is a general engineering reference, not a battery certification requirement.
| Evaluation item | Conditions to define | Evidence and decision |
|---|---|---|
| Mechanical fit | Drawing revision and mating fixture or equipment | Measurements and assembly observations against agreed features |
| Operating duty | Load profile, starting condition and installed arrangement | Recorded behavior against the agreed service requirement |
| Charging | Approved charger identity and operating conditions | Charge behavior assessed against the approved specification |
| Communication | Host, protocol and firmware versions | Required information and responses verified |
| Environmental suitability | Application exposures and approved method | Results reviewed against project-specific criteria |
| Documentation | Required files and revision references | Complete, consistent package or identified open items |
The table is a planning template. Qualified teams must fill in actual limits and safe methods. Do not treat it as a procedure for abuse testing or deliberate protection triggering.
Remove ambiguity before the evaluation starts
Consider the phrase “acceptable runtime.” Replace it with a requirement that names the agreed duty profile, starting state, environment, termination condition and minimum service duration. State how instrumentation and measurement uncertainty will be handled where they could change the decision.
Do the same for fit: identify which dimensions and assembly actions determine success. A photograph of a pack inside the equipment may not show connector access or whether the cover can be installed. Define the observation needed for each requirement instead of collecting evidence without a decision rule.
Vanguard’s OEM electrification guide discusses application testing and using prototypes to understand loads. Use that principle to agree a representative host-equipment evaluation rather than relying solely on a brief bench demonstration.
Allocate samples and responsibilities deliberately
Agree how many samples are needed for the planned activities and which units may be reused. Retain the exposure history of each unit. A sample that has been reworked or subjected to a damaging evaluation should not silently become an unchanged reference for another decision.
Assign the test owner, witness where needed, data reviewer and final approval authority. Confirm that the required facilities and equipment are available. Supplier bench evidence and customer integration evidence can complement each other, but their scope should remain explicit.
Close failures and define the release gate
Use a discrepancy record for an unmet criterion: sample identity, test condition, expected result, observed result, evidence, responsible party and next action. Agree which affected evaluations must be repeated after a correction. A resumed demonstration alone does not explain why an earlier result failed.
Conclude with a written decision: accepted for the named stage, accepted with specific conditions, or not accepted. Record unresolved items and who may close them. A prototype acceptance decision does not by itself establish production repeatability or target-market compliance; those need their own agreed evidence.
Review the Process page for battery assembly context. Send the intended sample stage, requirement matrix and evaluation responsibilities when you discuss your requirements via B2B enquiry. Confirm the available evaluation scope for the project before making it part of a supply commitment.