Printing at night: what you actually save on electricity

Scheduling the printer to start in the small hours and catch the cheap electricity band is one of the most widely passed-around tips. It sounds like free money: same part, smaller bill.

It is. The problem is how much: ten cents a part, and $18.48 a year if you move your entire output from the expensive band to the cheap one.

Not bad for something you set up once and forget, but far less than people assume. This guide is why, and where the real money is.

What one print actually uses

Let us start with real consumption, which is the part almost nobody has measured. The reference part on this site is 60 grams in 5 h 30, on a printer averaging 120 W.

That is 0.660 kWh. For the entire print.

To put it in perspective, that is roughly what an electric oven uses in twenty minutes. A desktop 3D printer is not one of the appliances that moves your bill: the hotend heater is small, the bed pulls hard only at the start and then just holds, and the motors draw very little.

At $0.15 a kWh, which is the calculator's default, those 0.660 kWh are $0.0990. Ten cents of electricity in a part that costs you $5.11.

That is 1.9% of the cost. That percentage is the reason for everything that follows: you can do whatever you like with the price of a kWh, you are still fighting over 2% of the bill.

Preheating does not change it either

There is a fair objection to the above: for the first few minutes the printer draws far more than its average, because it is heating the bed and the hotend at once. That stretch does not show up in the print time your slicer gives you.

True, and worth putting a number on. This one does not come from the calculator: it is a separate sum, with high power sustained through the minutes of heating.

Ten minutes heating a large bed at 350 W costs less than a cent. It is 8.8% of the whole print's consumption, which was already small.

So the objection is right on physics and wrong on accounting: the spike is real, but it is brief and it does not move the bill. Where preheating does matter is in time, which is far more expensive than electricity, and that is worth watching if you print a lot of small parts one at a time.

The same part on each tariff

Let us put it in a table. Electricity bands change with the country, the contract and the day, so what matters is the range: cheap is around 6 cents a kWh and expensive around 20.

From the cheapest row to the dearest there are 18 cents. And between off-peak and peak, which is the decision you can actually make, there are 10 cents.

You can open both ends and compare: the part off-peak and the same one at peak. The fourth decimal of the electricity line moves, and little else.

What those cents add up to in a year

Ten cents a part looks negligible, and on one part it is. But it is worth looking at a full year before writing it off, because repeated cents sometimes surprise.

At two hundred parts a year, which is 1,100 hours of machine time:

This time it does not surprise. Moving a whole year of output from the expensive band to the cheap one saves $18.48.

Put another way: all the electricity for a year of printing, on the worst band, costs less than two spools of PLA. And the difference between doing it well and doing it badly is less than one spool.

If you want the figure for your own machine, with its power draw and your hours, there is the electricity cost calculator, which already has the measured power of the most common models.

Where the real money is

Here is the important part of this guide. If you are going to spend time cutting the cost of your parts, it helps to know which lever pays most. These are the four, all on the same part.

There it is. Taking ten grams off the part saves three and a half times what moving all your printing to the small hours does, and it does not force you to print at night.

Those 10 grams are the part at 50 grams and 4 h 35, which comes to $4.76. And dropping the infill on a decorative part from 20% to 10% usually does exactly that, without it being noticeable in the hand: it is measured in what infill costs.

And look at the third row, the one most people ignore: a single minute of your time is worth double the entire electricity tariff difference. If taking the part off, cleaning it and packing it takes you one minute less, you have already gained more than scheduling overnight prints all year.

When electricity does start to matter

Everything above applies to a normal desktop printer. There are two cases where the sums change, and it is worth knowing whether you are in one of them.

The first is power draw. 120 W is typical of a small machine printing PLA. A large printer, with a 300 × 300 bed and an enclosure, running ABS at 100 °C bed temperature, is in another league.

At 500 W the electricity is four times higher, but the part only goes up 34 cents: from $5.11 to $5.45. Still not the line that decides anything, though you can see it now.

The important thing here is something else: average power is not what the label says. The label gives the peak, which the machine only touches while heating up. The average over a long print is considerably lower, and that is the one to put in the sums.

The second case: a lot of hours

The other situation where the tariff starts to weigh is the obvious one: when the hours multiply. One machine doing 1,100 hours a year moves nothing. Five machines doing 5,500 is a different conversation.

At 120 W each, those 5,500 hours are 660 kWh a year. Off-peak they cost $39.60; at peak, $132.00.

The difference is $92.40 a year, and there it is worth scheduling the starts. Not because $92 is a lot of money, but because at that scale the measure costs nothing: you queue the jobs for the night once and it is done.

Even so, keep the proportion in mind. On 5,500 hours of output you will be spending around $1,200 on filament, and the full breakdown of where every euro of a machine hour goes is in what an hour of 3D printer time costs. Electricity is still the smallest of the four lines.

How expensive electricity would have to get

There is one more way to look at this, and it is to turn the question around. Instead of asking how much we save, let us ask what would have to happen for electricity to be a serious line.

A kWh would have to reach $1.00 — nearly seven times today's price — for electricity to make up 11.5% of the part. And even then it would weigh less than the machine.

That is the definitive answer to why this article turned out the way it did. It is not that electricity is cheap: it is that a 3D printer uses very little for how long it runs. It spends six hours switched on and uses what an oven does in twenty minutes.

So if what you want is to cut the cost of your parts, the electricity tariff is not your lever. It is not even the second one.

The good reasons to print at night

That the saving is small does not make printing at night a bad idea. It makes money the wrong reason. There are three right ones:

• Noise. A running printer makes a constant hum that bothers people more than you would expect in a flat. At night it bothers nobody, and it leaves your daytime quiet.
• The machine works while you sleep. This is the real one. If the printer sits idle eight hours every night, you are losing a third of the capacity you already paid for. That is the calculation that matters, and it does not show up on the electricity bill: it shows up in depreciation.
• Heat. In summer, an enclosed printer warms the room. Overnight that is welcome.

The reason that does not hold up is saving on the bill. If you are going to print at night, do it for the other three, which are worth a great deal more than ten cents.

And one warning worth stating: leaving a printer running unattended carries risk. If you are going to do it regularly, it is worth having the machine on a surface that will not burn, a smoke alarm nearby and, if you can, a camera to keep an eye on it. That is worth more than the $18 a year.

Working out your own in five minutes

Three steps to settle it with your own numbers.

One. Look at your bill and write down the kWh price of your most expensive band and your cheapest. If you are on a flat rate, you are done: this article is not about you, and you have saved yourself the decision.

Two. Multiply your machine's average power in kW by the hours you print in a year. Those are your kWh. If you do not know the average power, the electricity calculator has it for the most common models.

Three. Multiply those kWh by the difference between the two prices. That is your maximum annual saving from scheduling at night, and it is the maximum because it assumes you currently print everything on the expensive band.

If the result is worth the effort to you, go ahead. If not, you know where to look: the table of four levers above, starting with the grams.

More tools and guides

• How much does 3D printing cost: 20 parts with real numbersThe real cost of 20 typical 3D printed parts: keyring, planter, phone stand, gear and more, with their grams, their hours and a link to recalculate.
• Why 3D prints fail and what it really costs youA failure halfway through costs 2.96 €, and only 32% of that is material. The four causes in order, with the real annual cost of each failure rate.
• How long does a 1 kg filament spool lastHow many metres and how many parts you get from a 1 kg spool, by material and diameter. Tables for PLA, PETG and ABS, and how to gauge what is left.
• 3D printing cost calculatorWork out the real cost and selling price of a 3D printed part: material, power, machine wear, failures and markup. Free, no sign-up, nothing uploaded.
• How much to charge for a 3D printed partPrice your prints properly: real cost, your labour and your markup. Generates a quote you can copy and send straight to the customer.
• How much electricity does a 3D printer useWork out your 3D printer's power use and what it costs per month on your own tariff. With the average wattage of the most common models.

Tools Guides