Chapter 25 · Phase 5 · Brewing Methods
Immersion vs Percolation
The question this chapter answers：Why is this the most useful brewing distinction?
Phenomenon
The wall of a coffee shop holds a dozen brewers: glass drippers, ceramic drippers, metal mesh cones, a glass pot with a plunger, a plastic tube, a two-chambered glass globe, an octagonal aluminum pot... Each has its own name, its own recipes, its own fans.
But after a few months of brewing, you notice something. A few of them taste about as good no matter how you use them. Others change a lot when you change how you pour.
There are dozens of brewers. Are there dozens of different principles of brewing?
Variables
Set aside shape, material and brand for now, and look at one thing only: how water and coffee meet.
| Question | Immersion | Percolation |
|---|---|---|
| How water and coffee come together | All the water steeps with all the coffee | Water flows through a bed of coffee |
| The water around the coffee | The same batch from start to finish, getting more concentrated | Continually replaced by fresh water flowing down |
| Who decides the time | You do | Grind, bed and pouring together |
| How it ends | You separate the grounds from the liquid | When the water has run through, it is over |
| Typical brewers | French press, cupping bowl | Pour-over dripper, drip coffee machine |
The variables of the coming chapters — grind, ratio, temperature, agitation — exist on both sides, but they act differently on each side.
Experiment
A thought experiment: same 15 g of coffee, same 250 g of water, same grind.
- A: Put all the coffee and water in one cup, steep for four minutes, then filter.
- B: Put the coffee in a dripper and slowly pour the 250 g of water through it, also taking about four minutes.
Write down your prediction before reading on
Which cup extracts more? Why?
Stored only in this browser on this device. Never uploaded.
A second thought experiment: grind two steps finer for both methods, changing nothing else.
Write down your prediction before reading on
In which method does the time change too? In which does the flavor change more?
Stored only in this browser on this device. Never uploaded.
One last one: stretch both methods to eight minutes of steeping (or flowing).
Write down your prediction before reading on
In which method does extraction rise more and more slowly until it nearly stops?
Stored only in this browser on this device. Never uploaded.
Observation
In the first experiment, with everything else equal, B usually extracts more Confidence: Established Mechanism. The water in A keeps getting more concentrated, so dissolving slows in the second half; in B the coffee is always surrounded by relatively dilute fresh water, so dissolving keeps going.
In the second: A's time does not change, because you set it; B's drawdown gets longer, because finer coffee slows the flow (Chapter 20). So the same two steps finer usually changes B's flavor more Confidence: Rule of Thumb: it changed two things at once.
In the third: immersion extraction rises fast, then slowly, and finally levels off at a plateau Confidence: Strong Evidence. Past the plateau, longer steeping adds almost nothing. Percolation has no such plateau: as long as fresh water flows through, what is left in the coffee keeps being carried away — right down to the point where only bitterness and dryness remain.
Model
Everything comes from the concentration difference
Dissolving runs on a concentration difference: there is a lot of soluble material in the particles and little in the surrounding water, so it moves into the water. The smaller the difference, the slower the dissolving Confidence: Established Mechanism.
- Immersion: all the water is one batch. It gets more concentrated, the difference shrinks, and dissolving slows itself down.
- Percolation: fresh water keeps flowing down and carries away what has dissolved, so the concentration difference around the particles stays relatively large and dissolving can continue Confidence: Established Mechanism.
Almost every trait of the two families follows from that one difference.
One difference, one table
| Aspect | Immersion | Percolation |
|---|---|---|
| Ceiling on extraction | A plateau; steeping longer barely pushes it higher Confidence: Strong Evidence | No such cap; it can extract more, and overshoot more easily |
| Time | A dial you set directly | A result of grind and flow — a reading |
| Grind | Changes only how fast things dissolve | Changes how fast things dissolve and the flow (and so the time) |
| Ratio | Once at the plateau, changes almost only strength Confidence: Strong Evidence | Mainly changes strength, but also bed depth and flow |
| Agitation and pouring | Relatively small effect | Large effect: decides whether water passes evenly through the bed |
| Risk of unevenness | Lower: all the coffee steeps in the same water | Higher: water takes the path of least resistance |
The last row deserves a word. In immersion, the water around every particle is much the same. In percolation, wherever the water goes, the coffee there gets washed more. Espresso research found that when flow becomes uneven, overall extraction actually drops Confidence: Strong Evidence. That is the root of why pouring technique matters in pour-over (Chapter 27).
Real brewers are usually mixtures
These are two ideal endpoints. Real brewers often sit in between:
- A pour-over's bloom is a short immersion before the percolation starts;
- When water pools in a pour-over or drip-machine basket, the top of the bed is in fact steeping;
- Some brewers steep the coffee for a few minutes, then open a valve so the liquid drains out through the bed.
That does not break the classification. Quite the opposite: once you know which stage of a brewer is immersion and which is percolation, you can predict which dials it is sensitive to. Chapters 29 and 30 use this ruler on many more brewers.
Transfer
When you see a new brewer, ask first: is all of the water steeping together with all of the coffee?
| Answer | Then |
|---|---|
| Yes: steeped together, separated at the end | Think immersion: time is a dial, the ratio mostly sets strength, extraction slows itself down, forgiving of technique (Chapter 28) |
| No: water flows through a bed | Think percolation: time is a reading, grind changes two things at once, even pouring and an even bed matter (Chapter 27) |
| It steeps first, then drains through | Think in two stages: time for the steep, grind and bed for the drain |
And a troubleshooting order to go with it:
| What you taste | Immersion: change first | Percolation: change first |
|---|---|---|
| Sour, thin | Steep a little longer | Grind a little finer |
| Bitter, drying | Steep shorter, or grind coarser | Grind coarser; check whether the pour is too aggressive |
| Sour and drying at once | Rare; check whether the grind is very uneven | Common; check whether pouring and bed are even |
| Balanced, just too weak or strong | Change the ratio | Change the ratio |
Misconceptions
"More brewers means more principles to learn." Most brewers fit first into immersion, percolation or a combination of the two. Learn the family first, then the details.
"Steep long enough and any method extracts just as much." Immersion extraction is capped by the shrinking concentration difference and barely rises with longer steeping; percolation keeps bringing fresh water, which is how it can extract more.
"Immersion under-extracts, so it tastes weaker." Weak or strong is a strength question, mostly about the ratio (Chapter 22). Immersion extraction does have a ceiling, but that is a separate question from how strong a cup is.
Real Lab
Real lab
Immersion vs Pour-over
Change only：Extraction geometry
Keep constant：Same bean
This chapter is this experiment's home. Here is how:
- Same bean, same dose and water (for example 15 g of coffee and 250 g of water).
- Steep one cup in a French press (or any cup, then a paper filter) for four minutes and filter it; brew the other through a pour-over dripper.
- Grind each the way you normally would — immersion can be a little coarser. Record the pour-over's drawdown time.
- Before tasting, compare the color: if one cup is clearly darker, add a little hot water until the strengths are close, then taste. That step controls strength, so you compare only "immersion versus percolation".
Then do a second round: grind both two steps finer, changing nothing else. Which cup's flavor moved more? Which cup's time changed?
Check Your Model
Predict. You have a dripper with a valve at the bottom that can be closed. You close it, add all the coffee and water, steep for three minutes, then open the valve so the coffee drains through the bed. Before looking at any result, write down: is this immersion or percolation? Which dial matters in the first stage, and which in the second?
One way to think about it · Predict
It is a combination of the two. The three minutes with the valve closed are immersion: you set the time, and the ratio mostly sets strength. Once the valve opens, the liquid has to flow through the bed that has settled, and that stage behaves like percolation: the finer the grind and the deeper the bed, the slower it drains, and the drain time becomes a reading again. If the cup tastes bitter, check first whether the second stage took too long.
Diagnose. Your pour-overs have always been good. One lazy day, you put the same coffee and water in a cup to steep for about the same time as a pour-over, then filter it. The cup is weaker and flatter than your pour-over. What happened?
One way to think about it · Diagnose
Same coffee, same water, similar time — but in immersion the water gets more concentrated and dissolving slows in the second half, so extraction is usually lower than in percolation, and the cup tastes weak and flat. To make up for it, grind a little finer or steep a little longer; if it is only too weak, changing the ratio is more direct.
Explain. In your own words: why does changing the grind change the time in a pour-over but not in a French press?
One way to think about it · Explain
Pour-over is percolation: its time is the time it takes water to flow through the bed. Finer coffee means smaller gaps and slower flow, so the time gets longer. The French press is immersion: the time is whenever you decide to stop the timer, and the water does not have to flow through a bed, so grind only changes how fast things dissolve, not the time.