The Most Expensive Mistake I Ever Made on a Deep Hole Drill

The Setup That Felt Routine

I was drilling a batch of titanium shafts for an aerospace customer. The material cost was $800 per shaft. I had 20 shafts to drill, each one 600mm long with a 16mm bore running through the center. The job specification called for a tolerance of H7, which meant the bore had to stay within 0.018mm of the nominal diameter. I’d run similar jobs before without any trouble, so I treated this one like any other production run.

The machine was a TBT ML-800 gun drilling machine that I had been using for about three years at that point. I knew its quirks. The spindle would warm up after about 45 minutes of running, and I’d see a 0.005mm shift in tool position if I didn’t account for thermal growth. I had compensated for that on every hole from the first shaft through the seventeenth. Those seventeen shafts passed inspection without a single deviation.

Then I got to shaft number eighteen.

The job was going well until I broke a drill in the 18th shaft. The break happened at about 450mm depth. I heard the pitch of the spindle change for about half a second before the torque limiter kicked in. I retracted the broken tool and installed a new one. But I forgot to check the depth offset after changing the tool. The new drill was 5mm longer than the old one.

The drill advanced 5mm too far and punched through the end of the shaft into the tailstock center. The shaft was scrap. The tailstock center was damaged beyond repair.

The Real Cost Breakdown

When I tallied up the total cost, the number was worse than I expected. Here is how it broke down:

Cost of the mistake: $800 for the shaft, $600 for the tailstock repair, and three days of downtime. All because I didn’t check the tool offset after a tool change.

The downtime was the worst part. I had to wait for the replacement tailstock center to arrive. The customer delivery date slipped by a week. I had to explain to the production manager why the remaining two shafts couldn’t be finished on schedule. That conversation was not pleasant.

The Root Cause I Nearly Missed

At first I blamed myself for forgetting the offset check. That was true as far as it went, but I wanted to understand why I forgot. I traced the sequence backward. The broken tool had caused a moment of frustration. I had spent thirty minutes extracting the broken carbide remnants from the bore. By the time I had the hole cleared and a new tool in the spindle, I was focused on getting back to cutting metal. The offset check was a step I had performed hundreds of times before, but it was not part of a written setup procedure. It lived in my head.

I realized that my memory-based checklist was the real problem. I had no written standard work for tool change after a breakage. The steps were all in my head, and under stress, I skipped one. Since that day, I have written a one-page tool-change checklist and laminated it next to the machine control panel. It lists six steps in order: retract, remove broken tool, inspect bore, measure new tool, update offset, restart coolant. Every operator on that shift uses the same checklist now.

The Procedure I Follow Today

I triple-check tool offsets after every tool change now. Here is the exact sequence I use:

First, I measure the new tool with a micrometer before mounting it. I record the length in a logbook. Second, I enter the measured length into the CNC control and verify it against the previous tool length. Third, I run the tool to a reference surface at low feed and confirm the actual position matches the commanded position within 0.01mm. I do not skip the reference surface check even when the production schedule is tight. That third step would have caught the 5mm discrepancy on shaft eighteen.

I have added a fourth step for critical jobs: I program a dwell at 5mm before final depth and verify the tool tip position with a dial indicator mounted on the tailstock. This takes an extra thirty seconds per hole. It has saved me four times since I started doing it.

What I Tell New Operators

When I train new deep hole drill operators, I tell them this story on their first day. I show them the cost spreadsheet. I walk them through the checklist. I tell them that any operator who breaks a tool and does not re-check the offset will be sent home for the day. That sounds harsh, but I have seen the numbers. A $3,920 mistake is not a learning opportunity. It is a failure of the system.

I also tell them that the most dangerous time is right after a tool breakage. The frustration creates tunnel vision. The pressure to recover lost time pushes you to skip steps. That is exactly when mistakes happen. I have seen operators skip coolant checks, ignore spindle load monitoring, and forget to reset peck cycles in the same situation. Every one of those shortcuts cost money.

The Machine-Specific Factors I Had Overlooked

Looking back, there were two machine-specific details that made this mistake more likely than it should have been. First, the TBT ML-800 uses a manual tool length entry system. There is no automatic tool measurement. The operator measures the tool with a micrometer, types the value into the control, and hopes the entry is correct. There is no verification loop built into the software. A modern CNC might refuse to cut if the tool length change exceeds a threshold percentage. This machine did not have that safeguard.

Second, the control panel displays the current tool length offset on a secondary screen. I had to press a button to cycle through display modes to see it. During a normal tool change, I would check that screen. After the breakage, I was in a hurry and did not cycle through the display modes. The information was there, but the interface hid it behind an extra button press. I have since re-programmed the display to show tool offset on the main screen at all times. I also added an audible alarm that sounds if the tool length changes by more than 3mm between cycles. That would have caught the 5mm difference immediately.

The Statistical Argument for Checklists

After the incident, I ran the numbers on how often tool breakages happen in our shop. I pulled the data from the maintenance logbook for the previous two years. The results surprised me:

Nine out of twenty-two tool changes after breakage involved a skipped step. That is a 41% failure rate. The secondary damage incidents cost an average of $2,100 each. My $3,920 mistake was on the high end, but it was not an outlier. The data showed a systemic problem, not a personal failure.

I presented these numbers at the next production meeting. I proposed a mandatory written checklist for every tool change after breakage. The production manager agreed. We implemented the checklist across all three shifts. In the twelve months since then, we have had zero secondary damage incidents from skipped steps. The checklist cost nothing to produce. It saved an estimated $8,400 in the first year alone.

Key Takeaways

• A single forgotten step after a tool change cost $3,920 in material, repair, and downtime
• The root cause was not carelessness but the lack of a written procedure for tool change after breakage
• Memory-based checklists fail under stress. Write them down and laminate them at the machine
• The reference surface verification step takes thirty seconds and catches offset errors before they damage the workpiece
• The most expensive mistakes happen right after a tool breakage, when frustration and schedule pressure combine
• A written offset verification procedure, followed by every operator, prevents a repeat of this failure
• Historical data from our shop showed that 41% of tool changes after breakage involved a skipped step. A written checklist eliminated secondary damage entirely