This is a complete 8D report for one made-up complaint: weld nuts that spun loose at the customer. It shows what a good entry looks like in each of the eight disciplines, with containment counted at every location, three root causes proven by a trial, one action per cause, and an effectiveness check with numbers before it closes.
The case: bracket 4410-B is a laser cut, press brake formed steel bracket with an M8 nut projection welded on. The customer bolts it into an assembly. Lot 2610-14, 1,200 pieces, shipped and then came back as a complaint: nuts spinning loose when the bolt was torqued.
D1: Team
| Role | Why they are on it |
|---|---|
| Quality engineer (lead) | Owns the 8D, the customer contact and the evidence. |
| Weld cell lead | Runs the projection welder every day and knows the setup. |
| Maintenance technician | Owns the electrodes, tip dressing and the weld controller. |
| Production supervisor | Owns the schedule, staffing and the containment sort. |
| Customer quality engineer | Confirms the problem statement and the clean break at their end. |
D2: Problem statement
Customer reported 37 brackets, part 4410-B, from lot 2610-14 (1,200 pieces), where the M8 weld nut spun when the bolt was torqued on their assembly line. Found at the customer on day 1 of use. Returned parts pushed out at 1.1 to 1.6 kN against a 4.0 kN minimum. No spun nuts reported on earlier lots.
Read it back: someone who was not there can check every number in it. See How to write a problem statement.
D3: Containment
Every location that could hold suspect brackets was checked within 24 hours. Each nut got a proof torque check, which does not damage a good weld.
| Location | Quantity checked | Failed | Action |
|---|---|---|---|
| At the customer | 1,163 (1,200 less the 37 found) | 9 | Sorted on site, failed parts returned |
| In transit | 400 (lot 2610-15) | 3 | Intercepted and sorted on arrival |
| Finished goods | 800 | 6 | Sorted, failed parts red tagged |
| Work in process | 300 | 2 | Sorted, failed parts red tagged |
| Total | 2,663 | 20, plus the 37 found |
Clean break: lot 2611-02, the first lot welded after the tip dress interlock went in and checked 100 percent. Certified boxes carry a green dot on the label until the effectiveness check closes.
D4: Root cause
Three legs, each with its own chain of whys. The full chain is on the 5 Why example.
| Leg | Root cause | Evidence |
|---|---|---|
| Occurrence: why the welds were weak | Electrode tips were dressed once per shift by the clock. After the run rate went up, tips wore past their limit before the next dress. | Weld counter showed about 1,100 welds per shift, up from about 600. Trial below. |
| Detection: why we did not catch it | The push-out test was done on the first piece of the shift, right after a fresh dress, so it only ever tested the best case. | Control Plan and test records: one test per shift, all at shift start. |
| System: why the system allowed it | The run rate change was treated as a scheduling decision, not a process change, so nobody reviewed the PFMEA or the dress interval. | Change request list covers tooling, material and method. Volume is not on it. |
Verified by turning it on and off. 40 nuts welded with tips at the end of the old interval: 9 failed push-out. 40 nuts welded after a fresh dress: 0 failed.
D5: Permanent corrective actions
| Leg | Action | Owner |
|---|---|---|
| Occurrence | Tip dress by weld count, every 400 welds, with a controller interlock that stops the welder at the count. | Maintenance |
| Occurrence | Weld current monitoring on every weld, with a fault stop and reject when current falls outside the window. | Maintenance |
| Detection | Push-out test on the last part before each tip dress, the worst case, as well as the first piece. | Quality |
| System | Run rate and shift pattern changes added to the change request trigger list, with a PFMEA review. | Quality manager |
D6: Implement and validate
Interlock and current monitoring installed and tested in week 1. Push-out test frequency changed in the Control Plan and work instruction the same week. Validation over the next three weeks: 126 push-out tests on end-of-interval parts, all above 4.0 kN, and 0 current faults that reached a part.
D7: Prevent recurrence
- PFMEA updated. The weld row was re-rated and now has the new controls. See the PFMEA example.
- Control Plan updated. New rows for weld current, tip dress count and the end-of-interval push-out test. See the Control Plan example.
- Work instruction updated. Tip dress steps and the push-out sampling, with photos of a good and a bad weld.
- Read across. The same interlock and test change applied to the second projection welder, which had the same time-based dress.
- LPA question added. "Is the tip dress counter active and below 400?" checked at every layer.
D8: Close and recognize the team
Closed only after the effectiveness check: six consecutive customer lots with no spun nuts and no push-out failures in our own data over 90 days. Team thanked at the shift meeting, with the weld cell lead credited for spotting the run rate change.
Everything on this page is made up for the example: the part, the numbers, the customer and the team. Use your own forms, rating scales and customer requirements.
Use these next
- How to write an 8D report, the method behind this example
- 5 Why example
- PFMEA example
- Control Plan example
- Quality glossary, for any term on this page
- Free tool: 8D report checker
- On Etsy: the 8D Corrective Action workbook, the same form ready to fill in