Why don't the two loss figures add up any more?

Because they never should have. Uptime and yield multiply, so some parts would have been rejected during time the cell was not cycling anyway. That overlap belongs to neither fix and arrives only when both are done. the earlier two rows added up because one of them silently absorbed it, and which one depended purely on the order the factors were applied in.

Should I load more parts per fixture?

Only if the added positions still cure. Gross capacity rises with loading, but a batch cycle cannot end until the worst-lit part has its dose, so extra positions at the edge either lower yield or lengthen the cycle. Run the page both ways with the yield you actually observe at each loading: the fuller fixture wins less often than people expect.

My cure rejects are always the same fixture positions. What does that mean?

That it is a fixturing problem, which is good news because fixturing is cheap to change. Positional failures mean specific locations see less dose. Shadowed by neighbours, further from the source, or at an unfavourable angle. Rotation, respacing, a reflector, or simply not loading those positions all fix it. More lamp power fixes it too, at the cost of overcuring everything else in the fixture.

Where should loading time go: the cycle rate or uptime?

In the cycle rate, if loading is a normal part of every cycle. Reserve uptime for time the cell is not cycling at all: changeovers, breakdowns, waiting for parts. The only real error is counting it in both, which double-charges the same minutes and understates capacity twice.

Which improvement should I do first?

Compare the two single-fix rows and weigh them against effort. On a typical batch cell uptime dominates by a wide margin because handling, not curing, sets the pace, but a cell with positional dose problems can invert that, and the whole reason the rows are separated is that the answer is a property of your numbers rather than a general rule.

Does the cell's cycle rate guarantee the parts cured?

No. Nothing on this page checks dose. A cell cycling fast enough to short the exposure produces at exactly the calculated rate with nothing cured, and the failure is invisible here. The dose and dwell pages are where a cycle time is validated against what the material needs.

UV Curing calculator

UV Batch Cure Capacity Calculator: Honest Loss Attribution

Work out what a batch UV cell really produces and what each improvement would return. Enter the parts per fixture, the cycles achieved per hour, cell uptime and first-pass cure yield. The calculator returns gross and net output, then divides the shortfall into what fixing uptime alone returns, what fixing yield alone returns, and the part that only appears if both are fixed. the earlier version reported two loss figures that summed neatly to the total; they summed because one of them absorbed an overlap that belongs to neither.

What this calculator does

Convert fixture loading and cycle rate into net good output from a batch UV cell, and split the shortfall into what each fix genuinely returns.
Use it for deciding whether to load a fixture more fully, ranking handling improvements against cure improvements, estimating batch cell capacity for scheduling, explaining why two improvement estimates cannot be added, testing what a faster load–unload cycle is worth.
Convert fixture loading and cycle rate into net good output from a batch UV cell, and split the shortfall into what each fix genuinely returns.

Formula used

Gross fixture capacity = parts per fixture × cycles per hour
Net good parts/hr = gross × uptime × first-pass yield
Recovered by fixing uptime alone = gross × (1 − uptime) × yield
Recovered by fixing yield alone = gross × uptime × (1 − yield)
Recovered only by fixing both = gross × (1 − uptime) × (1 − yield)

Inputs explained

Parts loaded per cure fixture: Parts in the fixture at one time. Loading more only helps if every added position still receives the required dose.
Completed cure cycles per hour: Full load–cure–unload cycles achieved in an hour. Cycle time is set by the worst-lit position in the fixture, not by the average.
UV cell uptime: Fraction of scheduled time the cell actually cycles, after loading delays, changeovers and stoppages.
First-pass cure yield: Share of parts passing on cure at the first attempt. In a batch cell this is usually driven by fixture position rather than by the lamp.

How to use the result

Best suited to deciding whether to load a fixture more fully, ranking handling improvements against cure improvements, estimating batch cell capacity for scheduling, explaining why two improvement estimates cannot be added, testing what a faster load–unload cycle is worth.
Cannot see WHICH positions fail. A 3% reject rate spread evenly and a 3% rate concentrated in two fixture positions call for entirely different responses, and only the second is cheap to fix. Assumes uptime and yield are independent; in a manually loaded cell they often share a cause in the operator's pace. Ignores the cycle-time penalty of loading a fixture more fully, which can offset the extra parts per cycle. Says nothing about dose. A cell cycling fast enough to miss the required exposure produces at the calculated rate and cures nothing. Excludes rework, which can return some rejects to good output at a cost carried elsewhere.

Common questions

Why don't the two loss figures add up any more? Because they never should have. Uptime and yield multiply, so some parts would have been rejected during time the cell was not cycling anyway. That overlap belongs to neither fix and arrives only when both are done. the earlier two rows added up because one of them silently absorbed it, and which one depended purely on the order the factors were applied in.
Should I load more parts per fixture? Only if the added positions still cure. Gross capacity rises with loading, but a batch cycle cannot end until the worst-lit part has its dose, so extra positions at the edge either lower yield or lengthen the cycle. Run the page both ways with the yield you actually observe at each loading: the fuller fixture wins less often than people expect.
My cure rejects are always the same fixture positions. What does that mean? That it is a fixturing problem, which is good news because fixturing is cheap to change. Positional failures mean specific locations see less dose. Shadowed by neighbours, further from the source, or at an unfavourable angle. Rotation, respacing, a reflector, or simply not loading those positions all fix it. More lamp power fixes it too, at the cost of overcuring everything else in the fixture.
Where should loading time go: the cycle rate or uptime? In the cycle rate, if loading is a normal part of every cycle. Reserve uptime for time the cell is not cycling at all: changeovers, breakdowns, waiting for parts. The only real error is counting it in both, which double-charges the same minutes and understates capacity twice.
Which improvement should I do first? Compare the two single-fix rows and weigh them against effort. On a typical batch cell uptime dominates by a wide margin because handling, not curing, sets the pace, but a cell with positional dose problems can invert that, and the whole reason the rows are separated is that the answer is a property of your numbers rather than a general rule.

Last reviewed 2026-08-25.