Miners in Satisfactory: Rates, Purity and Overclocking

Everything in your factory is downstream of a miner, and the two numbers that decide what one produces — the mark and the node purity — multiply rather than add. Getting that relationship straight prevents the most common early waste in the game.

The three marks

Miner	Extraction rate	Power	Unlocked at
Miner Mk.1	60/min	5 MW	Tier 0 — HUB Upgrade 5
Miner Mk.2	120/min	15 MW	Tier 4
Miner Mk.3	240/min	45 MW	Tier 8

Those are the rates on a normal purity node. All three occupy the same footprint — 6 by 14 metres, 84 square metres — so upgrading never costs you space.

The Mk.1 is cheap: one Portable Miner, 10 Iron Plate and 10 Concrete. Note the Portable Miner in that list, which is why you unlock the portable version first.

Purity multiplies the rate

Node purity	Multiplier	Mk.1	Mk.2	Mk.3
Impure	0.5x	30/min	60/min	120/min
Normal	1x	60/min	120/min	240/min
Pure	2x	120/min	240/min	480/min

A Mk.1 on a pure node produces exactly what a Mk.2 produces on a normal one. That equivalence is worth holding onto when deciding where to put a scarce upgrade: the mark and the purity are interchangeable, so put your best miners on your best nodes.

Purity is a fixed property of the node, shown when you scan it or point at it. Nothing you build changes it, and rocks or debris sitting on the node have no effect on output once cleared.

The belt mistake

This is the one that quietly costs people half a node.

A Miner Mk.1 on a pure iron node produces 120/min. Put it on a Mk.1 belt rated at 60/min and you have thrown away exactly half the node. Nothing flashes red. The miner simply backs up and idles, and you find out much later when a downstream line will not scale.

The habit worth building: whenever you place an extractor, check the belt mark against its actual output rate before walking away. Belt capacities are 60, 120, 270, 480, 780 and 1200 per minute for Mk.1 through Mk.6.

A Mk.3 miner on a pure node produces 480/min, which needs a Mk.4 belt at minimum. That combination arrives well before most players think to check.

Overclocking a miner

Miners are overclockable and follow the standard production-building rule: power scales by an exponent of 1.321929, output scales linearly.

So a Mk.2 at 250% produces 300/min and draws about 3.36 times its base 15 MW — roughly 50 MW. The output is the same as a Mk.3 at 125%, which would draw considerably less.

The practical guidance:

Overclock when the node is the constraint. A pure node with a Mk.2 on it is producing 240/min; pushing to 250% gets 600/min from the same node. If you have power to spare and no more nodes, this is exactly what shards are for.

Underclock when the consumer is the constraint. If your smelters only want 90/min, running the miner at 75% saves disproportionate power because of the same exponent working in your favour.

The overclock calculator on this site gives the exact figures for any mark and clock speed combination.

The node must be clear first

A miner will not place on an obstructed node. Two things obstruct:

Resource deposit lumps — the spiky ore outcrop on the node. Hand-mine it by holding the interact key until it breaks.

Cracked boulders — grey rocks with visible dark cracks. Hand mining does nothing; these need a Nobelisk Detonator, and nothing else in the game removes them.

If you find a good node blocked by a boulder before you have Nobelisks, a Portable Miner placed at the edge extracts at a much lower rate and keeps a chain moving until you can clear it properly.

Planning around node supply

Node count, not machine count, is the real ceiling on a factory. A few consequences:

Process at the node where it reduces item count. Smelting iron ore into ingots is one to one and does not help, but making plates is three ingots to two plates, which cuts what you transport by a third.

Match extraction to demand rather than maximising it. A miner producing more than the line consumes just backs up, and you paid the power for it.

Alternate recipes extract more from the same node. This is the largest lever available — a recipe using half the ore per unit effectively doubles what your nodes support, without building anything new.

Water Extractors work differently

Worth a note because they are grouped with miners mentally and behave unlike them.

A Water Extractor produces 120 m3/min regardless of anything resembling purity, draws 20 MW, and needs deep enough water rather than a node. Shallow lake edges will not accept one, which is why some apparently large bodies of water refuse placement at the shoreline.

It also carries 10 metres of built-in head lift, which matters when planning a pipe run uphill — your first pump belongs where that 10 metres runs out, not at the extractor itself.

Resource Wells are a third category again: a pressuriser on the well head, with satellite nodes around it that each need their own extractor. They are the fluid equivalent of a node cluster rather than a single node.

Upgrading miners in the right order

When Mk.2 unlocks you will not have the materials to replace every Mk.1 at once, so the order matters.

Pure nodes first. A Mk.2 on a pure node adds 120/min over the Mk.1 it replaces. The same upgrade on an impure node adds 30/min. Same cost, four times the return.

Then whatever is feeding your worst bottleneck, regardless of purity, because throughput where it is needed beats throughput in general.

Impure nodes last, if ever. An impure node with a Mk.1 produces 30/min, and it is often simpler to leave it and put the upgrade somewhere it does more.

The same logic applies to belts: upgrade the line carrying 480/min before the one carrying 60/min, even though both are technically “on the old mark”.

Common mistakes

Belting a pure node with a Mk.1 belt. Half the output disappears silently.

Assuming purity affects quality. It affects rate only. The ore is identical.

Overclocking when you could build another miner. The exponent makes it expensive if you have a spare node.

Trying to hand-mine a cracked boulder. Only a Nobelisk clears those.

Running miners flat out into an undersized line. Underclock to match; the power saving is larger than the rate reduction.