To repeat a good mass-finishing result, record the part condition, exact machine setup, media, loading method, compound and water settings, cycle, and inspection requirements that produced it. Then check the next batch against that record. A timer setting alone is not a complete process.
Start with two questions: What did the unfinished part look like, and what did an acceptable finished part look like? If those answers change, the same machine settings may no longer produce the same result.
Record the process, not just the minutes
A process record is a written description of how a particular part is finished and checked. It should be specific enough that another trained operator can follow it without guessing what “the usual load” or “a little compound” means.
Incoming part condition, media condition, loading, compound dosing, water flow and machine settings can all contribute to batch variation. A supplier's example settings are not a recipe for your parts. Use settings established for your actual equipment and application.
Keep two things separate: the approved setup and what actually happened in the batch. Recording an intended setting is not evidence that it was used.
A practical finishing-process record
Use this as a starting checklist, not as a validated production instruction. Your engineering and quality requirements determine the limits, inspection method and release authority.
| Record section | What to capture |
|---|---|
| Part and incoming condition | Part number and drawing revision, material, batch identification, burr location and condition, and relevant oil or surface contamination. |
| Machine and setup | Machine identification, barrel or bowl configuration, dividers or fixtures if used, and the approved operating settings for that equipment. |
| Media and loading | Media specification and size, condition check, media quantity, part quantity, loading method and the units used. If a ratio is specified, say whether it is by mass or volume. |
| Compound and water | Exact compound identity, specified dose or concentration and how it is checked; water quantity or flow as applicable. Do not interchange a pump setting, a concentration and a measured flow. |
| Cycle and downstream handling | Each finishing stage and its settings, actual run time, separation method, rinse and drying sequence, and any interruption. |
| Acceptance and traceability | Required burr, edge, surface and dimensional checks; inspection locations and methods; results, operator, date, process revision, deviations and release decision. |
Do not copy a vibratory bowl's settings into a centrifugal barrel process. Even within one equipment type, a recipe belongs to the configuration and part load that were evaluated.
Give the next operator a usable finish target
“Looks good” leaves too much open. An approved reference part or controlled photograph can help communicate appearance, but it does not replace dimensional, roughness or cleanliness requirements on the drawing or process specification.
Identify which edges must be deburred and which features must retain their geometry. If surface roughness is specified, record the required parameter, units, measurement location and method. Do not treat a brighter surface as proof of a lower roughness reading or an acceptable edge.
For a new application, begin with representative unfinished parts and a clear finish target. Once a process works, preserve the setup and its inspection evidence together.
Check what changed before changing the cycle
If the next batch is different, compare it with the recorded baseline before adding time or changing several settings at once.
- Check the incoming parts. Are the burrs, material and upstream condition comparable?
- Check the load and consumables. Was the same media used, is its condition acceptable, and were quantities recorded in the same units?
- Check the actual machine, compound and water settings against the approved setup.
- Check separation, rinsing, drying and inspection. Did the difference appear during finishing or afterward?
- Follow your quality procedure for holding, investigating and releasing affected parts. Record any approved process change and the evidence supporting it.
Worn media is one possible cause, not the default diagnosis. A batch record helps separate a changed consumable from a changed part, load or handling step.
Common questions
Can we use one process sheet for a family of parts?
Only when the grouping and its limits are supported by your process evaluation. Similar-looking parts can differ in material, burrs, sensitive edges, holes and handling requirements. State which part numbers and revisions the sheet covers rather than assuming that one good batch qualifies everything similar.
Does a saved machine program replace the record?
No. It may preserve the settings it controls, but it does not by itself document incoming burrs, media condition, actual loading, downstream handling or acceptance results. Record those separately where the control system does not capture them.
How many batches prove that a process is repeatable?
There is no universal number for every part and application. Your quality requirements and process risks determine the evidence needed. One successful sample is useful starting evidence, not a guarantee for all future production.
Bring the baseline to the process discussion
If results vary, contact Richwood with the part material, machine and configuration, current media and compound, recorded load and cycle, and a description of what changed. A representative unfinished part and a clear accepted-finish reference make the discussion more useful than the cycle time alone.