Coffee Talk (and Tea)

Yep, matcha chocolate. Very easy to find here.

And the peach is wagashi, traditional Japanese sweets. That stuff takes years of training to make properly.

Iā€™m super into this.

1 Like

Iā€™m looking forward to the new wagashi thread with some of your creations.

1 Like

It is tempting but the amount of sugar is off-putting. Iā€™m not into sugar at all and avoid it as much as possible, limiting myself to a fistful of fruit for breakfast.

Have you ever monitored the slurry temperature throughout the process of making a pour over?
  • Yes
  • No

0 voters

Iā€™ve been having success getting the kind of cups I enjoy by doing this and am starting to think this might be a significant piece of the puzzle in making pour overs. An interesting observation Iā€™ve had is that temperature stability on V60s doesnā€™t seem to be consistent. What I mean by that is the pouring technique I used yesterday to keep the slurry above 190 F didnā€™t work today, and Iā€™m not entirely sure why. Yesterday I was able to do pulse pours of about 50 mL every 5-10 seconds, but today I had to maintain one slow, constant pour to keep the temp up and failed to hit 190 F for the majority of the brew. Not only could I not seem to maintain 190, I also seemed to lose heat much faster and I dropped below 180 F multiple times, and the result was unbalanced, underextracted coffee that wasnā€™t enjoyable.

The first and obvious answer is that the kettle temperature must have been lower. I checked it right before starting the pour and it was the same as yesterday (202 F), but I donā€™t think I had quite as much water in the kettle today. A smaller mass does not hold temperature as well as a larger mass, so itā€™s certainly possible that this led to a larger roll-off over the duration of the brew than Iā€™m expecting.

However, I theorize that thereā€™s a second factor in play, and thatā€™s ambient temperature. Itā€™s fairly cool and drafty in here today (66 F) because I donā€™t have the heat on, but Iā€™m almost positive it was around 71 F yesterday when I made coffee. Now, Iā€™m no chemist and donā€™t know if this matters much in practice; itā€™s just a hunch, and the only reason I even thought of it was from technical reading Iā€™ve done on refractometry. In refractometry, ambient temperature is monumentally important to the extent that a difference of just 1 degree C has a substantial effect on the refractive index of water. To be clear, Iā€™m not saying R.I. is related to heat loss, only that knowing this fact made me suspect ambient temperature might be a hidden factor. After all, most pour over devices have a large water surface area exposed to air and lose heat rapidly.

Thereā€™s more to ambient air factor than just temperature though. Velocity, turbulence, humidity, water stir, etc., are also factors. Right now youā€™re probably thinking Iā€™m going to recommend buying a $9,000 wind turbulence meter or something, but actually no, I donā€™t think we need to measure any of this. The measurement we care about is the brew temperature in the cone, so just being aware that these factors could influence brew temperature enough to make a difference is enough. The only way that can possibly be visible to us is by reading the slurry temp in real time.

Itā€™s possibleā€“even probableā€“that someone has written about this before, but Iā€™ll reiterate that I donā€™t see anyone out there recommending you monitor slurry temps in real time when making pour over coffee. My suspicion is that most of the hocus pocus recipe stuff doesnā€™t matter and isnā€™t reproducible for a reason: our kettle is boiling, we count off 20 Mississippis, click our heels together three times, spin around, then proceed to do five 50 mL pours at predetermined intervals while saying our Bloody Marys in between. And sometimes itā€™s good and sometimes itā€™s bad, and thatā€™s surprising because we did exactly the same thing each time, but what we donā€™t realize is that maybe our average slurry temps were varying wildly each time.

what are you using to measure the temp in the cone? I have a pretty accurate digital cooking thermometer but wrangling that while pouring seems like a nightmare. I have a frothing thermometer that clips on the side of a milk pitcher and would not require another hand but I think itā€™s barely more accurate than guessing and it doesnā€™t go high enough anyway.

Thermapen Mk4. I think itā€™s +/- 0.7 F with 2-3 second read time and NIST calibration. I hold that with my left hand while pouring with my right which, yeah, itā€™s not ideal. Currently brainstorming a better solution. The accuracy is one thing, but I think the read speed is equally if not more important. The newest model has a read time of 1 second and +/- 0.5 F accuracy. Thermoworks also make larger base units that you can plug wired probes into, and thatā€™s probably easier to rig up as a hands-free solution. I havenā€™t looked at options beyond that.

I tried this a year or two ago to develop a feel for how to pour to keep the temp steady. Might have even written about it in here, but I remember it was fairly frequent pouring. What Iā€™m realizing now is that the day to day variation on that seems to be too large to recommend a blind pouring technique to achieve consistent temperature.

looking at what they have, the models with the probes do look like they would be perfect for this except they all seem to have slow read speeds (looks like theyā€™re intended to be left in while youā€™re smoking or whatever so the read speed isnā€™t important).

Yeah, I need to do more research on this part.

Iā€™ve been down the rabbit hole of temperature probes most of the day and finally found one that has technical specs better than the Thermapen One. But first, you have to understand this time constant thing. You need five time constants to get a full temperature read. I recommend reading this to get a lay of the land if youā€™re so inclined (I learned some things):

https://www.thermoworks.com/thermometry101-basic-concepts-speed/

That means the Thermapen One has a time constant of 0.2 seconds since it can provide a read in 1 second. There are many different types of thermocouples, and then also there are RTDs (the needle probe things) and thermistors. Iā€™ve mostly been looking at K-type thermocouples that will plug into any standard handheld thermoreader that accepts a K-type. Here is the fastest probe Iā€™ve located:

However, Iā€™m not exactly sure what youā€™re imagining as the setup. This isnā€™t a needle probe but rather a small metal piece at the end of a wire. I think Iā€™d prefer this to a needle since I donā€™t intend to put the needle deep into the coffee bed. Iā€™m afraid that wonā€™t be as accurate and also may affect the water flow. So if this particular sensor can be submerged in liquid (Iā€™m not sure it can), then may plan would be to affix it in a way that the sensor part reads at the top of the coffee bed, perhaps buy snaking it through the finger loop and securing it there.

*I forgot to mention that the speed of the whole system will include any latency in the reader itself. Itā€™s not clear to me how to determine such latency though. Thermoworks doesnā€™t seem to indicate such a time unless itā€™s in the [[[deep manual]]] or whatever. Also, there are bendable thermocouples that might work well for this application, but Iā€™ve yet to see one with a super fast time constant.

Iā€™m starting to think the reason no one does this is because itā€™s not the easiest thing to put together.

1 Like

Iā€™ve spent the last 3 years studying energy and thermodynamicsā€¦ so Iā€™ve thought quite a lot about it.

Generally. The science says to look for the big variables. In particular

  1. The energy going into the system

  2. The mass of the different elements.

  3. The temperature of the different elements.

  4. Conductivity of the elements.

I think the kettle is where a lot of variation is happening. Dont think of the kettle as a mass of water at boiling point. Instead its.

  1. Water at boiling point
  2. A container with itā€™s own mass, conductivity and possible temperature variation.
  3. A heating element or surface which at the moment you switch off, will be hotter than boiling point, and will continue to transfer heat to the water, then drop below boiling and conduct heat away from it.

When you take the kettle off the boil. All the different elements continue to interact. I think your main variable here is the mass of the water.

I have a kettle with an inbuilt thermometer. I put it back on the stand between pours to reheat and tell me the temp. When the kettle is full. It holds it temperature well between pours. When its half full, the temperature has normally dropped by 2 degrees C just after the blooming pour. I would imagine it would drop further by second or third pour if I wasnt topping it back up.

I think you will be better off getting really tight on the water going in rather than trying to measure it later.

1 Like

ā€œcoldripā€ seems like a suboptimal name

first thing I think of is a sinus infection

yeah this absolutely makes sense.

This is all true but Iā€™m skeptical that it translates to slurry temperature behavior. Check out the variability that this guy gets using a full kettle and one single pour which should be far more predictable than what Iā€™m doing:

https://express.adobe.com/page/0ohz7EVO3oras/

this is one of those things that makes me want to just throw everything into the river and go back to maxwell house

1 Like

Whatā€™s weird is that there seems to be an attenuation effect on temperature in the slurry. Straight up boiling water probably isnā€™t going to sit higher than 195 F, whereas I can definitely get 190 F with 200 F water. But also you can see in that guyā€™s results that the thermokinetic profiles are different, not just peak temperatures. The time to peak, peak duration, and also descents donā€™t seem all that predictable.

The kettle temperature is only one component since we have people using different techniques. Multiple small pours that completely drain before starting a new one is obviously a very different temperature profile than pouring all of the water immediately.

There are some good arguments for not just dumping all of the water in and instead keeping the level just slightly above the top of the coffee bed. Thatā€™s more of less what I aim for. The thing I noticed doing this is that the slurry temp drops much faster with the smaller mass, but frequently introducing new hot water from the kettle allows me to keep the temp up.

Actually scrap that. I went through some calculations and think this is incorrect. It seems to ignore thermal conductivity entirely. LOL Perger.

fukken zuccā€™d