Method
Keeping an honest log of steep times and temperatures
For two years I wrote down eighty degrees in the notebook and brewed at something between seventy-one and eighty-six, depending on the vessel, the room and how long I had been looking for the thermometer. The log only became useful once the numbers in it were true.
The kettle is not where the tea happens
My kettle has a temperature setting and it is reasonably accurate at the moment it switches off. A probe thermometer in the water agrees with it to within about two degrees, which is as good as I need. The problem is everything that happens in the twenty or thirty seconds afterwards, and none of that is in the kettle's control.
Water poured from a kettle at eighty into a cold porcelain pot of about three hundred millilitres arrives at seventy-two. The same water into the same pot, pre-warmed with a rinse of boiling water and emptied, arrives at seventy-eight. That six degree gap is larger than most of the differences I had spent a year arguing with myself about. A glass cup with thinner walls costs less, around three degrees cold and one warmed, and a heavy stoneware mug straight off a shelf in a cold kitchen took nine degrees off the first pour.
So the number worth recording is the water in the vessel with the leaf in it, taken with a probe, and not the number on the kettle. Once I started doing that, entries from different months began to agree with each other for the first time.
How fast it falls in a cold kitchen
This kitchen sits at nineteen degrees for most of the winter and about twenty-three in summer, and the ground floor wall behind the table is cold to the touch from November onwards. I spent one Saturday measuring cooling rather than tasting anything, with a probe in the liquid and a phone timer, which is not a thrilling afternoon but it removed a lot of guesswork.
In an uncovered glass cup holding two hundred millilitres, starting at eighty, the water was at seventy-four after sixty seconds, seventy-one after ninety and sixty-seven after two minutes. Covered with a saucer, the same start was seventy-seven at sixty seconds and seventy-three at two minutes. In a pre-warmed lidded pot holding four hundred millilitres, the fall was much slower, about one and a half degrees a minute over the first three minutes, because there is more water relative to the surface and the lid does most of the work.
This matters more than it sounds. A ninety second steep of a delicate green in an open cup is not a steep at eighty degrees, it is an average of something nearer seventy-two, and the last thirty seconds of it are happening at a temperature where the leaf gives up quite different things than it does at the start. Whether that is good or bad depends entirely on the tea. It is only a problem if the notebook says eighty and you believe it.
Kept on the card
- Measure in the vessel, not in the kettle
- Rinse the pot with boiling water first; a cold one swallows most of a ten degree margin
- An open cup loses roughly six degrees in the first minute at these room temperatures
- Write the room temperature in the entry; in winter it changes the whole curve
Why my second steeps were dull
I had a settled belief that a second steep was always a weaker version of the first and that the good ones I kept reading about belonged to better leaf than I was buying. It turned out to be a temperature error and a drainage error, roughly in equal parts.
The temperature error is that the leaf is now wet, opened and already hot, so pouring the same eighty degree water over it gives a much more aggressive extraction than the first pour did, and a longer time on top of that makes it worse rather than better. When I started going a few degrees cooler and keeping the second steep the same length as the first, or only slightly longer, the second cup stopped being a thin echo and became a different and often better cup.
The drainage error is simpler and slightly embarrassing. I was leaving the leaf sitting in twenty or thirty millilitres of liquid in the bottom of the strainer between steeps, sometimes for ten minutes while I did something else. That residue keeps extracting the whole time at a falling temperature and then joins the next pour. Lifting the strainer clear and letting it drain properly, and not leaving the leaf standing for more than two or three minutes, improved the second steep more than any change of temperature did.
The ratio I keep coming back to
The figure that survives in my notebook is three grams of leaf to two hundred millilitres of water for most things, which is a little heavier than the standard printed on packets, with short steeps to compensate. For rolled or tightly twisted leaf I go to three and a half grams, because the volume in the strainer is deceptive and the same weight of rolled leaf opens to fill twice the space.
Time varies far more than weight. Ninety seconds for a spring green, two minutes for a summer one, two and a half to three for most black leaf, and a first steep of forty seconds for anything rolled and tightly compressed, with the understanding that the second will be better than the first. I do not use a timer app because I forget to stop it; I use a small sand timer I can see from the other side of the room, and when the tea is better than usual I note the time it actually ran rather than the time I meant it to run.
The whole log lives in a paper notebook on the shelf above the table, one line per cup, with the date, the tea, the weight, the water volume, the measured temperature in the vessel, the time, the room temperature, and one short sentence at the end. Three years of that is something like eleven hundred lines, and its usefulness is almost entirely in the boring ones. The exceptional cups are memorable anyway. The ordinary ones are what tell you where the middle of your practice actually sits.