Small zinc hardware arrives in mixed batches: flat belt buckles next to deep D-rings with blind holes. One window must clean both. This is how senior staff engineer the load so the hardest geometry sets the cycle, not the average.
<= 60%
Load factor
sets the cycle
Hardest geometry
100% coverage
Blind-hole gate
Geometry classes and their dewaxing demand
| Class | Example | Demand | Cycle change |
|---|---|---|---|
| Open flat | Belt buckle plate | Easy, 3 min | None - use as the reference |
| Open 3D | D-ring | Moderate, 4-5 min | None |
| Blind hole | Clasp cavity | Hard, 6-8 min | Extend cycle; tilt the basket |
| Shielded thread | Screw boss | Hard, needs flow | Rotate parts at 50% cycle |
Basket and flow design rules
| Rule | Do | Do not |
|---|---|---|
| Load factor | Fill to 60% maximum | Brim-full baskets - shadows clean nothing |
| Orientation | Face cavities downward, angled 15-30 deg | Nest blind holes against each other |
| Spacing | One part thickness between layers | Stack flat parts face to face |
| Rotate | Turn the basket at half cycle for mixed geometry | Assume cavitation reaches sealed cavities |
Blind-hole validation (do once per new part)
| Step | Method | Pass |
|---|---|---|
| 1 | Dye the cavity with marker ink | Ink fully removed |
| 2 | Run the production cycle | Cavity free of ink, no residue |
| 3 | Water-break the cavity | Film holds 10 s |
| 4 | Cross-hatch after plating | No lift at the cavity lip |
Watch out
A cycle that cleans a flat buckle is not evidence for a clasp. Validate the hardest geometry in the load and let it set the clock.
Common mistakes and the fix
| ✗ Set cycle time from the easiest part | ✓ Let the blind-hole part set the cycle |
| ✗ Nest parts for volume | ✓ Nesting creates shadowed faces and trapped wax |
| ✗ Skip rotation for mixed loads | ✓ Rotation costs 30 seconds and saves whole batches |
| ✗ Validate on scrap that was already clean | ✓ Use real production parts with real wax |
What each role watches
| Operator | "Keep the basket within the marked load line and rotate at half cycle." |
| Chemist | "If rotation is impossible, raise alkalinity to the top of range and shorten bath life instead." |
| QC | "Own the blind-hole validation record per part number." |
| Manager | "Price mixed loads honestly: the hardest part sets the cost." |
From the shop floor
Case: a shop quoted mixed belt-hardware loads at the flat-part cycle time. Blind-hole clasps failed cross-hatch in one batch out of five. Rotating the basket at half cycle and adding two minutes gave a 100% pass rate at a 4% cost increase - far below the cost of the failures.
FAQ
Can I clean blind holes without rotation?
Sometimes, but only at reduced load and higher temperature. Rotation is cheaper and more reliable.
Does higher alkalinity fix shadowing?
No. Chemistry cannot reach a shadowed cavity; flow and orientation do.
How often should a new geometry be validated?
Once per part number, and again whenever the polishing compound or basket changes.
Where to go next
- Level guide for this trade: Dewaxing & Ultrasonic Cleaning — technician levels
- Other levels: Junior, Intermediate, Management
- Owner view: Process Overview for owners
- Supplier view: Chemistry & Supply · Find content by trade + level
Related reading in this trade
FAQ - Mixed Loads and Blind Holes: Engineering the Dewaxing Window for Small
Straight answers from the shop floor. Ask us on WhatsApp if your case is different.
What problem does this article address?
Small zinc hardware arrives in mixed batches: flat belt buckles next to deep D-rings with blind holes. One window must clean both. This is how senior staff engineer the load so the hardest geometry sets the cycle, not the average.
What are the practical steps or checks?
Operator Load spacing again - parts stacked flat keep wax in the blind holes. Chemist Free alkalinity below 6% and the ultrasonic only pushes wax around the tank. QC Water-break must hold 10 seconds; if the film closes, the load goes back. Manager On
What should the workshop do differently?
Supplier view: Chemistry & Supply · Find content by trade + level