topbanner.gif
Login
Username

Password




Not a member yet?
Click here to register.

Forgotten your password?
Request a new one here.
Shoutbox
You must login to post a message.

10/21/2021 2:41 AM
muckymuck, mckinleyharris and williamjonex welcome cup

10/19/2021 6:22 AM
Welcome tarunk!

10/17/2021 12:40 PM
Ploni and nader fouad, Welcome!

10/15/2021 2:19 AM
merlot85, maycondelpiero and hoeltz, Welcome !

10/14/2021 10:06 AM
Thanks for the addition to the group. Seriously considering building a drum roaster along the lines of oldgrumpus's. Love the design and craftsmanship.

In Memory Of Ginny
Donations

Latest Donations
JackH - 25.00
snwcmpr - 10.00
Anonymous - 2.00
Anonymous - 5.00
Anonymous - 5.00
Users Online
Guests Online: 28

Members Online: 0

Total Members: 7,369
Newest Member: williamjonex

View Thread

Who is here? 1 guest(s)
 Print Thread
Immersion Brewing
ChicagoJohn
Background

In the past three years, Ive gone through several roasting modifications, but recently Ive focused more upon brewing. I want to share my current mental model of immersion and how Im now implementing it. Any comments would be appreciated as Im still very much in the learning process and open to experimenting.

Brewing Models

The two primary distinctions I see are infusion, where fresh water is continuously added to the coffee grounds, and immersion, where the grounds are exposed to a single quantity of water and subsequently separated. Examples of the former would be espresso, pour-over, and drip coffee makers, and examples of the latter Aeropress, French press, and Clever.

For infusion, the water composition is constant while that of the grounds constantly changes, while water and coffee changes interact for immersion. However, it seems to me that for a given grind, water temperature, residence time, and final coffee-to-water ratio, immersion has the potential to be more repeatable; less subject to variation from such effects as channeling, bloom, and uniformity of water application to the grounds mass than may occur with infusion processes.

Evolution of My Brewing Process

The Aeropress was my first transition from a Coffee-Mate and store-bought ground coffee, and I continued to use it as I ventured into roasting using a modified popcorn popper. As my consumption increased to 16 oz and then 32 oz each morning, I built a larger capacity air roaster and transitioned to an alternative brewing method in which I stirred grounds in a container of hot water and then filtered them through a paperless, stainless steel cone.

Most recently, based upon comments by allenb regarding water temperature, difficulties in dialing in some new single origins to the flavor profile I like through roasting, and recommendations on paper filters from snwcmpr (which I came to learn also remove cafestol, a substance which elevates LDL cholesterol, which I have), I decided to do some reading and experimenting to learn more about effects of immersion brewing parameters, and Im writing this to share where Im at now and hopefully to learn more from readers responses.

Immersion Brewing Parameter Effects

Per allenbs suggestion, I increased water temperature. I had been using 175 185F water as Aeropress recommends. Now I bring it to full boil, split it, and drop the immersion portion down to 195-200F before adding the grounds. Per snwcmprs suggestions Im now using a Hario 02 washed natural paper filter in my stainless steel cone. This should remove most or all cafestol according to what Ive read and I think it actually promotes the speed of filtration and produces a cleaner product.

I also learned that the flavor profile of the extraction can be significantly affected by the ratio of water to grounds. This, it turns out, is well reflected in Aeropress recipes, something Id not previously realized. As an experiment, keeping all other variables the same, when I doubled the water-to-grounds ratio Id been using for a Yirgacheffe I like, I obtained the same strong, aromatic flavors I dislike when I try making an espresso with this coffee. Clearly changing the brewing ratio dramatically altered the flavor profile. It turns out that osmotic pressure is a significant factor in extraction: The lower the water to grounds ratio, the faster the reduction in osmotic driving forces over time as an equilibrium is reached, a change that appears to be solute-specific and therefore capable of changing the flavor profile, not just the strength or intensity: By varying this ratio, Ive been able dial in the flavor profile to my preference. The ratio of coffee to total water used (ultimate dilution) can be controlled independently.

To illustrate, Ill use where I finally ended up in seeking an optimum result. Starting with a target full water to coffee ration of 18:1, an initial water volume of 500 ml (500 gm) would require 27.8 gm of coffee. Next I calculate amount of water to use for the target brewing ratio, which for me is about 7:1, requiring about 200 ml of immersion water. So get 500 ml of water boiling in a pyrex measuring cup in the microwave, pour all but 200 ml into my mug, stir that 200 ml for about 30 seconds to drop the temperature to around 195F, dump in the 28 gm of grounds, stir vigorously for about 45 seconds, and dump the slurry into the cone with washed filter paper. (I also experimented with grind size and wound up with a 25 setting on my Rancillio Rocky grinder now as opposed to the 50 I had been using. The slurry appears to reach equilibrium very quickly, at least in 30 seconds, and Ive tried continuing the stirring for 90 seconds without making a noticeable difference.

Ive also found that for other single origin roasts that had some objectionable aspect using my previous method, altering the brewing ratio from 7:1 up or down can often move the result into a region of flavor profile I like much better. And its nice that its repeatable.

Also, I had been using 48 gm of coffee to make my same ~ 16 oz drink (or two) each morning. Now Im using 28 gm, 60% as much, and enjoying the result more with no noticeable difference in strength.

So as of right now, thats pretty much where Im at with it, and any comments or suggestions would be appreciated. Im sure theres a lot more to learn.
So many beans; so little time....
snwcmpr
Interesting read.
--------------
Backwoods Roaster
"I wish I could taste as well as I wish I could roast."

As Abraham Lincoln said "Do not trust everything you read on the internet".
BenKeith
I've said many years ago, you can't really learn how to roast coffee beans until you learn how to properly brew the coffee. There's a whole lot more to brewing than just passing some water through it, no matter what process you use.
I started back in 2000 and it took me a few years to really learn the brewing process, and still regularly experiment with it. After I really learned how to brew it, it was amazing how bad all that stuff I had been roasting tasted. That's when I really started learning how to roast it, and 18 years later, I'm still learning.
The more I learn, the more I realize I don't know.

Also, the Bonavita 1.7L Variable Temperature Kettle is probably one of the best things you can have. I've used one since they first came out, and would buy another instantly if the one I have failed. It's PID controlled temp and will hold the temp for up to one hour, and you set it to the exact degree you want the water. It also heats the water very quickly if you just pour in the amount you plan to use.
Edited by BenKeith on 08/27/2018 7:04 AM
ChicagoJohn

Quote

BenKeith wrote:

I've said many years ago, you can't really learn how to roast coffee beans until you learn how to properly brew the coffee.

Also, the Bonavita 1.7L Variable Temperature Kettle is probably one of the best things you can have.


Good point. Not having a reproducible brewing method makes it difficult or impossible to evaluate roasting variables.

Currently I can only roughly control my immersion temperature within about 5F, and I would like to be able to control it more precisely. I have checked the 1.7L Bonavita variable temperature kettle being offered on Amazon, and while it has over 1800 reviews, over 10% complain about rusting and leaks developing in a few months. So I'm wondering if it is the same model you have -- sometimes manufactures try to cut costs on a successful item. The one they are selling is BV382518V.
So many beans; so little time....
renatoa
Is more about keeping temperature, than control...
I am using a simple stove Joe Frex 600 ml kettle:
http://barshaker....ackjoefrex
... and amazed how small is the temperature loss during the first minute, under 2C degrees.
Removed the cover knob, is just a M4 screw, and inserted a very precise and fast digital thermometer, kind of "Eric" used for E61 group adapters, I can bring manually the temperature exactly enough in the 85-88 C temperature range I am using for brew.
To be better than this, another kettle should have the same small footprint, is very tight in my coffee corner, be mobile powered or somehow able to keep temperature by itself, or have the base very handy, in the close proximity of my brew place.
ChicagoJohn

Quote

renatoa wrote:

Is more about keeping temperature, than control...
I am using a simple stove Joe Frex 600 ml kettle:
http://barshaker....ackjoefrex
... and amazed how small is the temperature loss during the first minute, under 2C degrees.


I think you raise a good question regarding temperature stability. I'm doing full immersion, not pour over. So I heat a quantity of water to a full boil and then do the immersion in a small fraction of it. For example, I heat 500 ml and only use 200 ml for the extraction. The water is currently in a 500 ml heavy pyrex measuring cup, and when I stir the 200 ml portion it drops 14F or 8C in one minute. I think perhaps I should to the extraction in a well insulated container like a dewar type vacuum flask.

But I guess my main concern is that whatever it is, it's repeatable.
========================
Turns out I have a 16 oz vacuum flask so I just tried it out. After transferring the boiling 500 ml into it, the temperature levels off at 197.5F / 91.9C. After pouring 300 ml out, leaving 200 ml in it, the temperature then only decreases at a rate of 4F/2F per minute, and the residence time of my grounds is only about 45 seconds, so that should be fine.

Thanks for bring up the issue of temperature stability. ThumbsUp
Edited by ChicagoJohn on 08/27/2018 1:57 PM
So many beans; so little time....
renatoa
Yeah, and we are already in the realm of personal preferences... Grin

You wrote above that for you the most complete extraction is by immersion... well, not for me... I am sold to pour over.
Reason: trying to avoid by all means the bitter part of this completeness, rejecting even the reversed method of Aeropress, for example.
If Sivetz said "keep beans moving"... my saying is: keep water flowing Grin
ChicagoJohn

Quote

renatoa wrote:

Yeah, and we are already in the realm of personal preferences... Grin

You wrote above that for you the most complete extraction is by immersion... well, not for me... I am sold to pour over.
Reason: trying to avoid by all means the bitter part of this completeness, rejecting even the reversed method of Aeropress, for example.
If Sivetz said "keep beans moving"... my saying is: keep water flowing Grin


If you use a cupping ratio of water-to-coffee of say 16:1 or 18:1, in immersion, that may be the case. But what I'm finding is that if you greatly reduce that ratio, say to 4:1 to 7:1, then there is a limiting phenomenon, primarily driven by osmosis mechanisms, that minimizes over-extraction and makes that less subject to other variables -- such as extraction time. Continuous introduction of fresh water (infusion) would preclude this rate limiting phenomenon from happening because the concentration of all solutes in the new water is always zero.
So many beans; so little time....
renatoa
12:1 is my typical brew ratio.

So you suggest to prepare a 6:1 by immersion, then dilute by two ?
ChicagoJohn

Quote

renatoa wrote:

12:1 is my typical brew ratio.

So you suggest to prepare a 6:1 by immersion, then dilute by two ?


If you're in the mood to experiment. I think you may find it is different. But what I was comparing was more like full immersion in 12:1 with full immersion in 6:1 then diluting to 12:1. The 12:1 immersion came out much closer to the espresso I did with the same light Yirgacheffe roast while the 6:1 diluted to 12:1 came out much more like what I enjoy in coffee flavor profile. My wife, who only drinks espresso, also disliked the aromatic flavors (not bitter or acid) that came out in both the 12:1 immersion and espresso but not in the 6:1 immersion then diluted.

I really haven't done pour over so if you try it as an experiment to compare with that, I'd be quite interested in what you you see.
==========================
Postscript - Well, I decided to get a goose neck kettle so I'll be able to try pour over too. Conceptually I have a problem with it since it now seems clear that the amount of extracted solutes in water affects how effect that water is in further extracting fresh coffee. So it were possible to stratify the coffee in the cone after a pour over is completed, I would think there would be a gradient of extraction with the top layer over extracted and the bottom under extracted -- whereas all particles see essentially the same aqueous environment, albeit a changing one, in immersion. But we'll see since the criterion is going to be flavor profile.
Edited by ChicagoJohn on 08/27/2018 6:04 PM
So many beans; so little time....
renatoa
To avoid this layering, in the second half of extraction the grounds are mixed in the water.
I have a "smart" coffee machine who stop the pour at 2 minutes for 30 seconds, to allow me open the lid and homogenise the mixture.

Tried today 8:1 in Aeropress reversed, then diluted to 12:1. Bitter was predominant in both ratios :(
ChicagoJohn

Quote

renatoa wrote:

To avoid this layering, in the second half of extraction the grounds are mixed in the water.
I have a "smart" coffee machine who stop the pour at 2 minutes for 30 seconds, to allow me open the lid and homogenise the mixture.

Tried today 8:1 in Aeropress reversed, then diluted to 12:1. Bitter was predominant in both ratios :(


Well, maybe I'm completely wrong about the concentration effect. Maybe it's more of a question of water temperature. When I dump 30 gm of room temperature grounds into 200 ml of 195F water, there is going to be much more of an immediate temperature drop than if I drop that same 30 gm into 500 ml of 195 water. So maybe it's more the temperature than the water to coffee ratio.

I'll have to see if I can think of a way to keep temperature constant during the actual extraction process. One way to do that might be to start out with a higher initial temperature for the 200 gm process than the 500 gm process so that both reach the same temperature during extraction. I'll try some calculations involving approximate specific heat of coffee and water and see if I can do that experiment.
=============================================
So running the numbers, solely on the basis of specific heat, adding 30 gm of coffee grounds to 500 gm of 100C water should drop the temperature by 3.03C, and adding 30 gm of grounds to 200 gm of 100C water should drop it 7.37C for a delta of 4.34C or 7.81F.

While it strikes me as unlikely that a difference this small could be the sole explanation for the very substantial difference in flavor profile I observed in immersion of 30 gm in 200 gm then diluting to 500 gm total after filtering versus immersion of 30 gm in 500 gm before filtering, I'm going to do the experiment tomorrow anyway, starting at about 5-10C higher for the 200 gm water to try and compensate. We'll see what happens.
Edited by ChicagoJohn on 08/28/2018 8:21 AM
So many beans; so little time....
snwcmpr
I do not have much to add.
But when you mentioned temperature drop I was thinking pizza stove in an oven. The pizza stone takes longer to heat up the oven, but it heps maintain the same temp if/when the door is opened.

The container that your 200 ml of water is in. Does it have enough mass to hold the temp? Or is it a thin glass beaker?
Just another factor to add to your equation.

Ken
--------------
Backwoods Roaster
"I wish I could taste as well as I wish I could roast."

As Abraham Lincoln said "Do not trust everything you read on the internet".
renatoa
Should rename thread: the Pharmacy of the Brew Grin
ChicagoJohn

Quote

snwcmpr wrote:

I do not have much to add.
But when you mentioned temperature drop I was thinking pizza stove in an oven. The pizza stone takes longer to heat up the oven, but it heps maintain the same temp if/when the door is opened.

The container that your 200 ml of water is in. Does it have enough mass to hold the temp? Or is it a thin glass beaker?
Just another factor to add to your equation.

Ken


Excellent question. The container is a heavy pyrex measuring cup (670 gm) which the entire 500 ml (500 gm) of water was in during the 4 minutes on "high" in the microwave. I'm pretty confident that since the water reaches a very rapid boil at 3:30 and remains boiling as it is removed (in fact it is actually super-heated, the Joules of heat contained in the glass maintain the glass temperature at a very low gradient with respect to the water, resulting in negligible heat flow from one to the other.

So, I remove it from the microwave, quickly pour out 300 ml into the mug, and then dump the 30 gm of grounds into the remaining 200 ml, stir, and dump that into the cone / paper (ambient), where the aqueous portion has full separated in less than one minute.

The only difference is in dumping the 30 gm into 200 ml before filtering or into 500 ml. When I do this tomorrow, I'll time the 200 ml filtration and collect the 500 ml for equally as long, discarding the remainder to hold constant any additional extraction time that may have otherwise taken place in the cone due to a longer residence time there. I can only fit about 200 ml into the cone anyway, so it's probably going to work out the same for both.


Maybe something other than temperature and extraction ratio is involved. We'll see, I guess.
So many beans; so little time....
ChicagoJohn
The results of experiments this morning which control for temperature and time are convincing to me that water-to-coffee ratio is a factor that can be used significantly affect full immersion results.

I used an exposed junction, thin-wire K thermocouple with a calibrated pyrometer measuring in 0.1C increments to prepare four samples pairing brew ratios of 6.7:1 and 16.7:1 with extraction at 90C mean slurry temperature for 40 seconds mix time and 80C for mix times of 90 seconds. After preparation, the filtrate from the 6.7:1 brew ratio samples was diluted to the same initial water volumes used to prepare the 16.7:1 samples so that evaluation would take place at comparable total water to coffee ratios.

After cooling to ambient temperature, these four samples were evaluated for taste and pH value (hydrogen ion concentration, an index of carboxylic acid extraction).

Taste comparison in the 90C/194F samples at 40 seconds mix time before filtration indicated a good flavor balance in the lower water ratio. The 16.7:1 ratio was quite acidic and bitter and these characteristics were lingering and made the coffee undrinkable, in my opinion. The pH measurements were 6.65 and 6.38, respectively, and with pH values being the negative exponents of power of 10, this translates into 1.9X the acid in the 16.7:1 ratio coffee, assuming the dissociation constants of the acids involved are similar in both cases. That is, the higher water ratio extracted almost twice as much carboxylic acid with other variables controlled.

In the 80C/176F (a temperature recommended by Aeropress) trials with much longer mix times before filtration, the 6.7:1 ratio was a bit better, I thought, but the 16.7:1 was equally bad and had the same pH of 6.37. The pH of the 6.7:1 sample was 6.46, indicating that perhaps the longer mixing time even at lower temperature extracted more acid than in the higher temperature trial. However, I didn't find this objectionable. Additionally, with the longer mix times, the 16.7:1 ratio sample picked up the same objectionable aromatic flavors we find for a light roast like this when we've tried to make it in espresso -- definitely not floral or fruity, more "vegetable-like", I'd say. Maybe almost carrot like without being sweet -- not like carrots but along those lines. Not a flavor I like in coffee in any case.

So in summary, controlling for extraneous variables as best I could, I found that drastic changes in brewing ratio can result in drastic changes in flavor profile. In some of the previous trials I did where I was not measuring variation in temperature and time specifically, I think I observed more subtle differences in the range of 4:1 to 7:1 in flavor profile.

So anyway, all this leaves me convinced, now, that this is a real phenomenon and one that can be used to advantage in combination with roasting profile to achieve a result that is more to the personal preferences of a coffee drinker.
=======================================
Postscript -- The coffee used in the above is a Yirgacheffe natural process Gedeb from Happy Mug roasted with the same light profile I've always used for Yirg. In my air roaster et-up, 1C starts around 385F at about 9 minutes and I initiate cooling at about 408F at 12 minutes. In the same roaster I do a Brazil 19+ vanilla from Happy Mug for my wife's espresso out to 430F where it just starts second crack. So that may give you an idea of what kind of roast I was using here relative to what you are doing. I would think that my immersion results would have been different for medium or dark roasts, but I'd guess the brew ratio would still be a significant factor.
Edited by ChicagoJohn on 08/29/2018 10:28 AM
So many beans; so little time....
chaff
Thanks for this informative thread. Your conclusions make sense to me in that pulling espresso seems to be similar, a fast shot tends to be a lot different from a shot pulled on the same charge tamped down hard to make a slower extraction.

i wasn't so happy to learn about about cafetol, I read further to find the claim that the cafetol is in the fines; now I have the dilemma, less espresso or double on the statins. I think I will chose 'B'.
ChicagoJohn

Quote

chaff wrote:

Thanks for this informative thread. Your conclusions make sense to me in that pulling espresso seems to be similar, a fast shot tends to be a lot different from a shot pulled on the same charge tamped down hard to make a slower extraction.

i wasn't so happy to learn about about cafetol, I read further to find the claim that the cafetol is in the fines; now I have the dilemma, less espresso or double on the statins. I think I will chose 'B'.


According to the articles I've read, cafestol is extracted in all of it, coarse or fine. I've read that a paper filter takes virtually all of it out.

I would think that, like most things, this is dose dependent. My wife drinks a double shot every morning and has great LDL values. I was using six times the amount she is, though not espresso, and my last LDL was inexplicably elevated. So if you're not drinking a lot of espresso's every day, and particularly if your LDL levels are within range, there's probably nothing to be concerned about. In my case, my LDL was high and so was my consumption. So I'm trying a year of paper filters and we'll see what I get next physical :)

Espresso is an infusion process since fresh water is continually applied. The immersion method is different in that you add the grounds to a fixed amount of water and the extraction components increase over time until they approach an equilibrium level based upon the water to coffee ratio you use (along with temperature and other factors).

So in espresso and pour over, particle size distribution of the grounds is important for exactly the reason you cite -- it dramatically affects the rate at which fresh water passes through the grounds. In the immersion method, it only matters in how rapidly the equilibrium value is reached (and, I suppose, how rapidly the slurry passes through the filter subsequently, a variable which does not change the extraction significantly relative to the initial immersion stirring).
So many beans; so little time....
chaff
https://www.sciencedirect.com/science/article/pii/S0963996917304477

By fines I meant the solid particulate which I presume the filter blocks better than the PF.
ChicagoJohn

Quote

chaff wrote:

https://www.sciencedirect.com/science/article/pii/S0963996917304477

By fines I meant the solid particulate which I presume the filter blocks better than the PF.


Thanks! Good to know. Until your explanation, I didn't understand how the filter removed the cafestol. Now it makes more sense.
Edited by ChicagoJohn on 08/31/2018 3:47 PM
So many beans; so little time....
Jump to Forum:

Similar Threads

Thread Forum Replies Last Post
Brewing home-roasted coffee Roasting Coffee 8 10/03/2020 7:59 AM
Is oxygen necessary for the brewing water? Preparing Coffee 19 02/04/2020 3:12 PM
Immersion Style Fluid Bed Concept Fluidbed Roaster 26 11/13/2019 11:31 AM
Immersion brewing temps and ratios? Preparing Coffee 7 03/18/2019 12:07 AM
Ew! Weird Film - Pourover Brewing Preparing Coffee 3 07/29/2017 8:46 AM
Homeroasters Association Logo, and all Content, Images, and Icons © 2005-2016 Homeroasters Association - Logos are the property of their respective owners.
Powered by PHP-Fusion Copyright © 2021 PHP-Fusion Inc
Released as free software without warranties under GNU Affero GPL v3
Designed with by NetriX