Rotor Magnet Bonding Process Window for Repeatable Prototype Builds

Most magnetic rotor prototype failures are not caused by magnet grade. They are caused by unstable bonding execution.

When adhesive amount, curing conditions, and fixture references are loosely controlled, pilot lots pass once and fail on repeat.

Related capability page: Turnkey Magnetic Assemblies.

Why "one good sample" is not enough

Prototype success can hide process instability. A rotor that passes once may fail in pilot because bonding variables were not controlled as a system.

For buyers, the objective is not only "part passes drawing." The objective is repeatable process capability across lots, operators, and rework cases.

1. Build the process window from measured pockets, not nominal CAD

Start with real manufacturing data:

• measured pocket size distribution from first-article output;
• pocket surface condition and cleaning readiness;
• fixture contact behavior and positional repeatability.

Then define a process window with lower/target/upper boundaries for:

• adhesive dispense amount;
• placement timing and open-time limits;
• part seating force and alignment controls;
• allowed squeeze-out profile and cleanup boundary.

If this step is skipped, runout drift and balancing rework will rise in pilot phase.

2. Treat adhesive selection as a controlled engineering decision

Do not treat adhesive as a commodity line item. Freeze:

• adhesive grade and approved alternates;
• lot traceability requirements;
• storage and shelf-life controls;
• handling rules before dispense (temperature/humidity/conditioning as required by TDS).

Create an "adhesive change control rule": no lot substitution without documented impact review.

3. Define curing profile as a release parameter

Curing profile must be part of technical release, not tribal knowledge:

All values should follow adhesive supplier TDS and your program validation plan.

4. Insert runout checkpoints into the route card

Runout must be verified in stages to locate variation sources:

1. Post-machining baseline: confirm starting geometry.
2. Post-bonding interim check: detect placement-induced shift.
3. Post-curing cleanup check: capture curing/cleanup effects.
4. Pre-balancing release check: confirm readiness for dynamic test.

For each stage, define:

• fixture/setup reference;
• measurement method;
• acceptance threshold;
• disposition rule if out of limit.

Without staged checks, teams only discover issues at final balance, where correction cost is highest.

5. Add MSA and capability expectations for pilot

Pilot governance should include minimum metrology confidence:

• measurement method repeatability check for key CTQs;
• clear operator instructions for setup and zeroing;
• evidence that the method can discriminate near-threshold parts.

For capability management, classify CTQs by risk:

• high-risk CTQs: stricter sampling and containment;
• medium-risk CTQs: standard pilot controls;
• low-risk CTQs: reduced frequency after stability confirmation.

This keeps quality effort proportional to technical risk.

6. Define rework boundaries before production starts

Rework rules should be explicit and pre-approved:

If this table does not exist, teams make inconsistent decisions under delivery pressure.

7. Link balancing acceptance to bonding process control

Balancing should not be treated as a separate department issue. It is an output of upstream bonding stability.

Require in pilot reports:

• balancing result records tied to lot/serial IDs;
• linkage to runout stage data;
• trend view across pilot batches (not single-point pass/fail only).

This helps buyers decide if the process is stabilizing or just statistically lucky.

8. Pilot-to-repeat release checklist

Before moving to repeat lots, confirm:

1. Process window boundaries are documented and signed.
2. Curing profile and hold/transfer rules are frozen.
3. Stage-gate runout data shows stable behavior.
4. Rework matrix is approved by engineering and quality.
5. Pilot report pack includes traceable balance/runout evidence.

If any item is missing, keep the program in pilot mode.

RFQ block for bonding-centric projects

Program: [name]
Assembly type: [rotor variant]
Drawing revision: [rev]
Pilot qty / repeat qty: [x / y]

Bonding controls requested:
- Adhesive class and approved alternate rule
- Cure profile and hold-time definition
- Runout checkpoints by process stage
- Rework boundary matrix

Evidence required:
- Stage-wise runout records
- Balancing report linked to lot IDs
- Nonconformance and corrective-action summary

This level of RFQ precision improves supplier comparability and reduces late-stage surprises.