Revisiting: Achieving High Extraction with Low Agitation

This article introduces a Pulsar recipe aimed at achieving high extraction with low agitation

Disclaimer: All English content in this blog is AI-translated. Apologies for any quirks – blame the bots!

This blog’s IG account^[A] is live! Going forward, shorter reviews and thoughts will be posted on IG. Please follow!

The Most Anticipated Dripper of 2023: NextLevel Pulsar

In The Physics of Filter Coffee, Jonathan Gagné outlines his ideal dripper: cylindrical to minimize bypass, even water distribution at the bottom to maximize filter paper usage, and a valve to control flow rate. NextLevel, the creators of the LVL-10 dripper, took note. They collaborated with Jonathan to create the Pulsar, a name that not only nods to his profession as an astronomer but also hints at its key feature with the homophone “Pause”: the valve allows for precise control over water flow, even pausing it entirely.

After the announcement of this collaboration, coffee enthusiasts around the globe eagerly awaited its release. Two years flew by, and in August 2023, Pulsar announced pre-orders. I quickly ordered two, and after a month of anticipation, they finally arrived.

My initial experiments with the Pulsar were both challenging and somewhat disappointing.

The Problems I Encountered with Pulsar

The consensus seems to be that Pulsar excels at reducing astringency. However, I found this to be only partially true. The recommended Pulsar recipes, including those from Jonathan Gagné and others^[B], advocate for multiple pours to maintain a low water level and reduce pressure, which is believed to help minimize astringency. Despite numerous attempts and adjustments, I found the astringency to be higher than anticipated (or to clarify, about the same as my other brews, but I had higher expectations!).

I suspect the issue lies in excessive agitation. While multiple pours keep the water level low, they also increase agitation. Observing the coffee bed during brewing with similar recipes, it’s common to find it uneven, potentially leading to inconsistent water flow. The primary issue is Pulsar’s height. To ensure adequate capacity, there’s a significant distance between Pulsar’s disperser and the coffee bed, causing more agitation than expected, even with the disperser.

Before receiving the Pulsar, I had high hopes it would outperform the Tricolate, which required considerable effort to dial in. However, after nearly 20 brews, the results, in terms of both flavor and data, were not noticeably better, which was quite disheartening.

Discussion with SEY

I learned that SEY was also experimenting with the Pulsar, so I reached out to them for their insights.

They shared my concerns, and believed the greatest challenge was “how to achieve high extraction with minimal agitation,” echoing the focus of my previous article. After numerous discussions, revisions, and over 50 of my own experiments, I settled on this recipe.

Recipe Introduction

Here are the detailed parameters and brewing steps, largely mirroring the previous recipe:

• Coffee to water ratio: 18g:360g (1:20)
  • Avoid using more than 20g of coffee, as the Pulsar’s capacity may be exceeded, risking overflow.
• Water temperature: 100°C
• Brewing water: distilled water, re-mineralized post-brew (minerals are added directly to the brewed coffee^[C]).
  • Aim for water with 30–40GH, 15–20KH^[D], with equal amounts of Ca²⁺ and Mg²⁺.
  • If unsure about your water, normal tap water with a low TDS is acceptable.
• Grind size: Burr gap set between 150 to 200 µm from lock.
  • Comparable to traditional 9 bar espresso settings.
  • I used the Lagom P100 with the 98mm SSP Brew burrs, with grind settings between 2.0 and 3.0.
  • If your grinder produces more fines, consider a slightly coarser grind.
• Target concentration (TDS): 1.35%+
  • African coffee can often reach 1.45%+.
• Target extraction yield: 23.5%+
  • African coffee can often reach 25%+.

Required Equipment

• Very lightly roasted coffee
  • Given our aim for very high extraction yields, it’s crucial that the coffee tastes good when highly extracted. Lighter roasted coffees typically fare better at high extractions.
  • Roasteries I’ve had success with include SEY, The Picky Chemist.
• Pulsar and its dedicated filter papers
• Grinder, carafe, digital scale, and other basic pour-over gear (a pouring kettle might not even be necessary!).
• A WDT tool
  • Alternatively, a spoon, chopsticks, or glass stir stick can be used to stir the coffee bed during the bloom.

Detailed Brewing Steps

1. Separate the Pulsar’s wall and base.
2. Close the valve, pour a small amount of hot water into the base, just enough to submerge the ribs at the bottom.
3. Place the filter paper inside; it will float on the water.
4. Open the valve; the filter paper will adhere neatly to the bottom.
   • This ensures no air bubbles are trapped between the valve and filter paper, allowing for smoother water flow.
5. Install the wall onto the base and close the valve.
6. Measure 18g of coffee, grind, transfer to the filter, and level the grounds (a WDT tool is recommended).
7. (With the valve closed) Place the disperser on top, pour in 100g of water quickly, and stir with a WDT tool or stir stick to ensure all grounds are wet.
   • If the coffee is very fresh, stir more carefully and evenly to prevent the coffee bed from bubbling and forming many small holes.
8. Perform a gentle spin (i.e., Rao Spin) to ensure the coffee bed is flat.
9. (With the valve closed) At 1:00, begin pouring slowly in a circular pattern (about 2.5g/s) until reaching 360g. The pour should not finish before 2:30.
10. Half-open the valve, about 1 to 2 o’clock (or 10 to 11 o’clock). The smaller the valve opening, the better, but ensure a smooth water flow (not dripping).
11. Wait for the water to fully drain. The total brew time should be around 6–8 minutes.
12. Re-mineralize using Lotus Coffee Brew Water, adding one drop each of CaCl₂ and MgCl₂ solutions, and two drops of NaHCO₃, achieving roughly 15 ppm as CaCO₃ for Ca²⁺, Mg²⁺, and HCO₃^-.

Recipe Design Thoughts

Below, I’ll share some unique aspects of this recipe and the rationale behind them.

Immersion-Style Bloom with a Large Amount of Water

Typically, the water amount for blooming is 2.5 to 3.5 times the coffee dose. A bloom with a large amount of water might seem wasteful (since it would extract more efficiently if added later) and is usually found in recipes not aiming for high extraction, such as Patrik’s April recipe.

However, the Pulsar differs from other drippers due to its “valve.” We bloom with the valve closed, ensuring all the water has ample contact time with the coffee.

This extra water plays a crucial role: it ensures that subsequent extraction is “low agitation.” During the brewing process, stirring during blooming is emphasized to ensure all grounds are wet. If you’ve cupped or brewed with a Syphon, you’ll notice that after thorough stirring, most of the suspended grounds begin to sink. Ideally, by the end of the bloom, all the grounds should be evenly settled at the bottom, with the water above acting as a buffer layer for the subsequent pour, further reducing agitation.

Therefore, it’s important to ensure the coffee bed is sunken and flat at the end of the bloom. If the coffee is too fresh and stirring is not thorough, the coffee bed will continue to absorb water and bubble during brewing, causing an uneven bed and risking uneven water flow or channeling when opening the valve later. In such cases, more thorough stirring is advised.

Single Continuous Pour

A single continuous pour is a technique often seen in low extraction recipes. It not only reduces the agitation of multiple pours but also tends to shorten the total extraction time due to the high flow rate caused by the high water level. However, the Pulsar’s “valve” allows us to control the flow rate independently of the water level height, enabling us to ensure sufficient brewing time by adjusting the valve opening.

The choice of a single continuous pour stems from our original goal of “low agitation.” It ensures a sufficient buffer layer between the coffee bed and the water surface, and we pour very slowly throughout the entire process, further reducing agitation.

Brewing Log and Discussion

A successful brew log using this recipe
The coffee used was SEY's Colombia Washed Pink Bourbon
The TDS and extraction yield were 1.39% and 24.43% respectively

Why Not Use Multiple Pours?

I acknowledge I was somewhat evasive earlier.

Although the overall process is quite different, as highlighted in the title, this recipe shares the same core idea and goal as my previous article Achieving High Extraction with Low Agitation. To recap, the previous recipe used the OREA V3 dripper, with the entire brew including five pours, bloom included. So, why does this recipe opt for a unique single continuous pour after the bloom?

Generally, increasing the number of pours has three effects:

1. Increases agitation, enhancing extraction efficiency.
   • However, we aim to avoid this in this recipe, making it a “side effect” of multiple pours.
2. Increases the concentration gradient, enhancing extraction efficiency.
   • When we pour, we add completely clean water to the dripper, creating a larger concentration gradient between the water and coffee grounds.
3. Lowers the average water level (because we pour less each time), reduces flow rate, increases total extraction time, and thus enhances extraction yield.

In the original OREA recipe, the main purposes of multiple pours were 2. and 3., but it also produced the “side effect” of 1. With the Pulsar, I can achieve the goal of increasing total extraction time in 3. by controlling the valve opening size, so the only advantage of multiple pours is 2. In my tests, I felt the “advantage” of 2. was not enough to offset the “disadvantage” of 1., which is why the single continuous pour method is adopted.

But the lack of a concentration gradient also required me to use a finer grind to achieve a similar extraction yield. In the previous article’s recipe, I used a grind size of 200 to 300 µm, but in this Pulsar recipe, I used 150 to 200 µm, which may increase the bitterness and astringency of some beans. When this occurs, I recommend trying to split the pour and use a slightly coarser grind.

Isn’t This a Full Immersion Recipe?

When I first received the prototype of this recipe from Lance, my initial question was, “Isn’t this essentially a full immersion recipe? Then why is blooming necessary?” Yes, typically, full immersion brews (or “near” full immersion, as most brews still undergo filtration to remove fines) like the inverted AeroPress or Syphon don’t require blooming. However, upon closer examination, you’ll realize this is not a full immersion recipe but rather a very slow percolation extraction (or a hybrid of immersion and percolation, with percolation likely playing a larger role). Why? Because there is no “immersion” process.

Immersion extraction is generally considered a low-efficiency method because, midway through the extraction, it essentially becomes “extracting coffee with coffee.” Compared to percolation extraction’s “extracting coffee with clean water,” the lower concentration gradient results in lower extraction efficiency.

However, in this recipe, because we don’t stir after the bloom and pour very slowly, it can be considered that we are gently “layering” the water on top. This keeps the upper layer of water’s concentration low, and most of the extraction occurs as the water passes through the coffee bed, akin to percolation extraction. How can we verify this? By observing the coffee flowing out. You’ll notice that in the early stages of extraction, the coffee liquid is very dark, while the liquid flowing out at the end becomes extremely light. If it were full immersion extraction, the color of the coffee liquid would be consistent throughout.

Since most of the extraction occurs during the percolation stage, blooming is necessary, and our immersion-style bloom with a large amount of water further reduces agitation, as mentioned earlier.

Conclusion

This recipe, while significantly different from the one introduced in my previous article Achieving High Extraction with Low Agitation, is based on similar concepts and thus produces a similar flavor profile, with “sweet, complex, and transparent” as the overarching extraction goal.

In my attempts, this recipe proved more convenient (as it requires only one pour) but had a lower success rate. Some beans produced excessive bitterness no matter how I adjusted this recipe. I’m still puzzling over this, so I invite everyone to try it and share their thoughts with me^[E]! Together, perhaps we can further refine this method.