Failed print cost calculator: what failures add to every good part

The formula

A failed print burns four things up to the moment it stops. Call the full print’s filament g grams and its time h hours, and call the share it completed p.

filament     = g × p / 1000 × filament price per kg
electricity  = watts / 1000 × h × p × price per kWh
machine time = machine hourly rate × h × p
labor        = minutes to clear and restart / 60 × labor rate

cost of one failure  C = filament + electricity + machine time + labor

attempts per good part   = 1 / (1 − failure rate)
failure cost per good part = C × rate / (1 − rate)
monthly failure cost     = prints started per month × rate × C
upgrade payback (months) = upgrade cost / (prints per month × (old rate − new rate) × C)

Filament and time come from your slicer’s print summary. Progress comes from the printer screen or the layer where the print stopped. The machine hourly rate is your printer’s purchase price spread over the hours you expect it to run, plus upkeep; the printer hourly rate calculator builds that number step by step.

Where the defaults come from

The defaults describe a mid-size PLA (polylactic acid) print on a single desktop printer. Replace every one with your own figures.

Input	Default	Basis
Filament, full print	120 g	Example part
Print time	6 h	Example part
Progress at failure	50%	Example: failed halfway
Filament price	20 per kg	Example spool price; use your receipt
Power draw	80 W	Prusa’s knowledge base reports an average of 80 W for an MK-series printer on generic PLA at 26 °C room temperature
Electricity	0.18 per kWh	EIA’s US average residential price for June 2026 was 18.34 cents per kWh
Machine rate	0.25 per h	Example: a 750 printer written off over 3,000 print hours
Clear and restart	10 min	Example; time yourself
Labor rate	20 per h	Example; use what your hour is worth
Failure rate	10%	Example round number; measure your own
Prints started per month	60	Example: two a day
Upgrade	100, cutting the rate to 5%	Example to show the payback line

How much does a failed 3D print actually cost?

With the defaults, one failure costs $5.33: $1.20 of filament, $0.04 of electricity, $0.75 of machine time and $3.33 of your time. The exact figure scales with how far the print got. Labor is the biggest line here, because clearing the bed and restarting takes the same ten minutes whatever failed.

Worked example

A 120 g, 6 hour PLA part fails at 50%. That is 60 g of filament and 3 hours of printer time gone.

Line	Working	Cost
Filament	60 g × $20 / 1000	$1.20
Electricity	0.08 kW × 3 h × $0.18	$0.04
Machine time	3 h × $0.25	$0.75
Labor	10 min × $20 / 60	$3.33
One failure		$5.33

The same part printed to completion costs $3.99 ($2.40 filament, $0.09 electricity and $1.50 machine time). A failure halfway through costs more than a good print because of the labor line. A slicer’s cost estimate leaves this out: the second attempt is a full print on top of the failed one.

Filament wasted on failed prints

Filament is the line people notice, yet here it is small: 60 g of PLA at $20 per kg is $1.20. A first-layer failure wastes almost nothing, while one at 95% throws away nearly the whole part. Progress matters more than the spool price.

What percentage of 3D prints fail for hobbyists versus print farms?

No reliable published figure separates hobbyists from print farms. The best peer-reviewed data comes from a university makerspace, where Song and Telenko (Procedia CIRP, 2019) logged a 41.1% failure rate among mostly novice users. Farm operators publish no audited rates. Measure your own: failed prints divided by prints started.

What the published numbers cover

Three figures turn up in the research, and none describes a typical home printer or a production farm:

• 41.1% in an open university studio, with human error behind failures equal to 26.3% of all prints (Song and Telenko, 2019). The authors also found that user experience did not lower the failure rate in their data.
• About 20% for early self-built RepRap printers and about 10% in later estimates, as summarized by Petsiuk and Pearce (Additive Manufacturing, 2020).
• 1% to 20% from an informal poll of an online 3D printing community, also cited by Petsiuk and Pearce. A self-selected poll is not a measurement.

A shared makerspace full of first-time users will not match a tuned printer running the same PLA part every day. Use these figures as a range, not as your number.

What costs does a failed print carry beyond wasted filament?

A failed print also costs the electricity it used, the machine hours it consumed, and your time to clear the bed, check the nozzle and restart. Those hours carry the printer’s depreciation. On a print farm, the lost hours delay other jobs.

Cost	What drives it	In the calculator
Filament	Grams printed before the failure	Yes
Electricity	Printer watts × hours run	Yes
Machine time	Depreciation and upkeep per hour	Yes
Labor	Minutes to clear, inspect and restart	Yes
Lost capacity	Hours the printer could have spent on paid work	No; see below
Repairs	Parts damaged by a blob or crash	No; add it separately
Late delivery	Refunds or rush shipping on a missed date	No

Electricity wasted on failed prints

Electricity is usually the smallest line. At Prusa’s measured 80 W average on PLA, three hours of a failed print use 0.24 kWh. At EIA’s June 2026 US residential average of 18.34 cents per kWh, that is about 4 cents. High-temperature materials draw more: Prusa measured 120 W on ABS (acrylonitrile butadiene styrene). Heated chambers raise that further. The electricity cost calculator works out your printer’s hourly draw.

Why lost machine time matters on a farm

For a hobbyist, lost printer time costs only the depreciation in the machine rate. On a fully booked print farm, those hours had a paying job waiting, so value them at what the printer earns per hour.

How do you calculate the cost of a failed print?

Multiply the full print’s filament and hours by the share completed before it failed. Price the filament per gram, the electricity per kWh and the machine time per hour. Then add your minutes to clear and restart at your labor rate. The sum is the cost of one failure.

1. Open the print summary in your slicer and note the filament weight and print time.
2. Note the progress when the print stopped. A layer count works: layer 150 of 300 is 50%.
3. Filament cost: grams × progress ÷ 1000 × price per kg.
4. Electricity: watts ÷ 1000 × hours × progress × price per kWh.
5. Machine time: hours × progress × machine hourly rate.
6. Labor: minutes to clear and restart ÷ 60 × labor rate.
7. Add the four lines.

How do you track a print failure rate?

Log every print you start in one sheet: date, printer, part, material, and whether it finished. Failure rate is failed prints divided by prints started, over at least a few dozen jobs. Record the cause and progress of each failure too, so you can price your costliest failure modes.

A simple failure log

A spreadsheet with one row per print start is enough for one printer or a small farm:

Column	Example
Date	2026-09-24
Printer	Printer 3
Part	Bracket v2
Material	PLA
Result	Failed
Progress at failure	50%
Cause	First layer lifted
Minutes to recover	10

1. Add a row when a print starts, not when it ends, so abandoned jobs count.
2. Mark the result when you clear the plate.
3. Each month, divide failed rows by total rows for the failure rate.
4. Sort failures by cause and multiply each by its cost from this calculator. The most expensive cause is not always the most frequent one: a rare 90% failure on a long print can cost more than many first-layer failures.

On a print farm, log per printer as well as overall, so one troubled machine stands out.

How should failure rate be built into print pricing?

Add each good part’s share of failures to its cost: failure cost × rate ÷ (1 − rate). With the defaults, a 10% failure rate adds $0.59 to a $3.99 part, a 14.8% increase. Price from that true cost of $4.58, then add margin.

Why the rate is divided by (1 − rate)

A 10% failure rate does not mean 10% extra cost. To finish one good part at a 10% rate you start 1 ÷ 0.9 = 1.11 prints on average. So each good part carries 0.11 failures, not 0.10. The gap widens as the rate climbs. At 20%, each good part carries 0.25 failures. At 50%, it carries one full failure.

Failure rate	Attempts per good part	Failures carried per good part
5%	1.053	0.053
10%	1.111	0.111
20%	1.250	0.250
30%	1.429	0.429
50%	2.000	1.000

A shortcut for quoting

If failures usually happen late and cost about as much as a full print, the true cost is roughly the full print cost ÷ (1 − rate). This calculator is more precise because it prices each failure at the progress it reached, plus labor. The print cost calculator gives the full-print figure to start from.

How failure rate affects print farm profit

The failure cost comes out of margin before anything else. With the defaults, 60 prints started a month at a 10% rate means six failures and $31.96 lost each month. On a farm with ten printers at the same numbers, that becomes about $320 a month. Budget for it as a line item, then track whether it falls.

When is a reliability upgrade worth its price?

An upgrade is worth it when the failures it prevents, over a period you accept, cost more than the upgrade. Payback in months is the upgrade cost divided by monthly savings. With the defaults, a $100 upgrade that cuts failures from 10% to 5% saves $15.98 a month and pays back in 6.3 months.

How the payback is worked out

Monthly saving = prints started per month × (old rate − new rate) × cost of one failure. In the example: 60 × (0.10 − 0.05) × $5.33 = $15.98. Then $100 ÷ $15.98 = 6.3 months.

Three things decide whether the number is real:

1. The new rate is a guess until you measure it. Log failures for a month before the upgrade and a month after. Use the measured drop, not a product claim.
2. The upgrade must target your costliest failure cause. First-layer inspection catches failures that are already cheap, because little progress is lost. Camera-based AI print monitoring pauses a print partway through, which limits waste on failures that would otherwise run for hours. Bambu Lab’s own documentation says its spaghetti detection “is not a foolproof method” and may miss failures, especially with dark filament.
3. Lost capacity is not in the payback. If your printers are fully booked, every avoided failure also frees hours for paid work, and the true payback is shorter.

Do failed prints damage the printer?

Most failed prints do not. A print that detaches and is stopped early costs filament and time only. Two failures can cause damage: a detached print that sticks to the nozzle and grows into a blob, and a nozzle crash into a curled part. Prusa warns that a blob can break the heater and thermistor cables.

Blobs around the hotend

Prusa’s knowledge base describes the blob as forming when the first layer detaches and sticks to the moving hot nozzle while it keeps extruding. The thermistor and heater cartridge cables usually end up inside it. Prusa calls the heater cable “very vulnerable” because its insulation ends before the wire enters the cartridge.

If a MINTEMP or preheat error appears during removal, Prusa says the thermistor or heater cable is already broken. Prusa also notes that blob damage is not covered by warranty. Its main advice is to stop a print as soon as filament starts sticking to the nozzle.

Bambu Lab’s wiki describes the same failure as “hotend clumping”. It gives a cleaning procedure: heat the hotend, remove the front housing and pick the softened plastic away. It warns to avoid pulling or breaking the hotend wires and to avoid overheating the plastic housing.

A blob that has already broken a cable adds replacement parts and downtime to the cost of the failure. Enter that separately, since the calculator does not include repairs.

Crashes into curled parts

Prusa’s crash detection article says layer shifts usually happen when the extruder moves quickly between objects and hits a curled-up piece of the print. Warped overhangs can also rise into the nozzle’s path. Prusa suggests more part cooling or a larger Z-hop (lifting the nozzle on travel moves) in the slicer. A shifted print is a failure in itself, and it counts at whatever progress it reached.