Homeroasters - Home Roasting Coffee Community » THE ART OF ROASTING COFFEE » Roasting Chemistry
Who is here? 1 guest(s)
Page 2 of 2: 12
 Print Thread
endothermic reversal point
Dan
Top  #41 Print Post
Posted on 08/06/2013 10:15 AM
What I've noticed on this and other forums is that some people, but not BDL above, confuse exothermia and exothermic reaction. They are not the same and I've seen where this leads to misunderstandings.

An iron skillet on the stove is exothermic (gives up its heat) when the gas flame under it is turned off. And exothermic chemical reaction is when chemicals combine to produce heat; two-part epoxy is an example, and so is wood burning in a fireplace.
1 pound electric sample roaster, 3 pound direct-flame roaster, both handmade; modified Mazzer Mini grinder, LaSpaziale Vivaldi II automatic espresso machine. When the electricity goes out I make vacpot coffee from beans ground on my Zassenhaus hand grinder, and heat the water with a teakettle on the gas range.
 
Web
eric
Top  #42 Print Post
Posted on 09/25/2013 7:37 PM
Sorry for delay, will reply in full in the other thread:
http://forum.home...ad_id=1885

Specific points above:

Quote

Dan wrote:

This might help. Raemy and Lambelet measured coffee energies in their study and found that roasting becomes markedly exothermic around 195?F. as seen in this graph.
Is this an artificial environment where different chemical reactions will occur if the pressure is allowed to rise, which I am presuming would happen in a sealed container? (I have not read the paper you referenced). Just does not seem relevant to coffee roasting where we are concerned with overall exothermia, I guess - any exothermic chemical reactions less specific heat and latent heat of phase changes.


Quote

boar_d_laze wrote:

Quote

eric wrote:



I will take exception to one of your statements - the folklore around exothermic reactions at 1C - I am sure there is no such thing. I have never seen it with my setups that should have shown it: very low thermal mass air roasters with good thermometry can easily detect any exothermic reaction as a rise of bean temp above env temp.


Bad hypothesis. Exothermia doesn't mean that BT would be greater than ET. Exothermia would mean a faster rate of rise for BT AND ET (because the beans would be heating the environment by being net emitters of energy) than would otherwise occur absent exothermia.

The hallmark of science isn't making or using measurements which can be expressed with numbers, it's "scientific method" which is just a way of thinking things through and testing the guesses.

BDL
In an air roaster ET measures the air temp before it gets anywhere near the beans and airflow is one way from the ET measurement to the beans, so its difference with the bean pile temp is a good indication of beans overall absorbing or releasing heat. Please see the other thread above for details.

Should have raised this in the other thread, ooops. :-)
Eric
 
eric
Top  #43 Print Post
Posted on 09/25/2013 9:21 PM
Following on from the sweetness thread:
http://forum.home...ad_id=3421
I know this has been discussed at length in many forums. I guess I am firmly in the not exothermic during 1C camp.

Quote

Dan wrote:

Scott, I know there's been a lot of talk about this on this forum and other forums over the years, but I thought by now everyone had come to understand the simple science of the Maillard reaction. Yes, these graphs are scientific proof of an exothermic reaction. Here is the abstract for that article which clearly states coffee roasting is exothermic and even warns that it is so intense that it can start a fire:

"The technique of heat flow calorimetry was used to determine the specific heat of coffee and chicory products and to study their thermal behaviour above 20?C. Intensive exothermic reactions were particularly evident when measurements were made with sealed cells. The data obtained help in the understanding of how exothermic reactions (and self-heating) occurring in such foodstuffs can bring them above their minimum ignition temperature."*

The notion that roasting coffee is not exothermic is a falsehood. Granted, there are roasters who disagree, but until they run their own calorific tests and get it published in a peer-review journal, I'm going to stick with what these two dudes are saying.

Burning hydrocarbons are indeed releasing energy and it is therefore an exothermic reaction, but then so is mixing a two-part epoxy. If you've ever mixed too much in a paper cup you know it gets warm. That's an exothermic reaction, too.

*Raemy A, Lambelet P. A calorimetric study of self-heating in coffee and chicory. Int J Food Sci & Tech, 1982;17(4):451?460.
IMHO a sealed container calorimeter measurement is not relevant to coffee roasting as roasting does not happen in a sealed container. Chemical reactions are going to be different.


Many years ago when I read the Schenker 2000 dissertation (posted here in 2009 without any significant comment, still available free at http://e-collecti.../eth:23461) one of the takeaways of mine was that 1st is not exothermic and I have watched for it ever since. Now if you go to page 66 of the PDF (page 54 of the paper) you will see that he cites quite a list of authors who vote for exo at some point while there are only two at that time (including Illy '95) who say no exo.

Schenker did go to the trouble of drilling 250um holes in beans and installing thermocouples in individual beans for measurement of bean core temp. If you look at all his plots, the core temp ALWAYS lags the bean pile temp (what we call BT). He is quite adamant in his conclusion of no exothermic behaviour in both the laboratory and commercial roaster setting WITHIN normal roasting temperature ranges. Don't know how he could measure bean core temp in the middle of the 300kg Probat flying saucer roaster! Pity he did all isothermic air roasting in the lab, but I guess that is closer to commercial commodity roasting.

My air roaster controller has a "Cal" ET profile that I run after every rebuild where I put a full load of those glossy lumps of charcoal that used to be beans into the roaster and the controller locks onto a constant ET of 130C and waits for BT to stabilise (RoR<0.25C/min), records BT then ramps up to 230C ET and does the same measurement. Takes close to 30 min including auto cool down cycle. This gives me two offsets that I linearly interpolate and apply to the roaster screen ET plots relative to the BT to get more useful information than ET: the difference between the two plots is a reasonable proxy for rate of power flow into the beans, yes it depends upon the fan setting and current RoR as well but all the roaster specific errors are nullified. With an air roaster there always is the problem of the huge temp gradient in the air coming off the nichrome or from the flame - this needs to be well mixed before measuring ET. The other source of error nullified is the heat loss through the walls of the roaster so the effective ET-BT represents power used to roast the beans and heat up the roaster. I have also focused on minimising the thermal capacity of the roaster, as best I can within all the other contraints. It is a very complex optimisation problem.

Recently I added a field onto the roast screen on the roaster for adjusted ET minus BT so that I can see it while roasting (the graphing only happens afterward in review of a roast). Across many different beans and profiles I have never seen the power flow into the beans reverse and so I conclude endothermic throughout. I try to avoid 2C, and I must say if I accidentally run into a rolling 2nd then the power inflow to the beans drops significantly. Most roasts the power flow into the beans just goes down through first and on down up to 2nd. Sometimes it can look like it is heading for a cross but I know to kill the roast well before because I am well into 2nd. Never just left it to see - never have tested the fire detector cut out either!

Some numbers from my experience: when ET probe placement and mixing are working well ET-BT @ 130C is 3C and 5C at 230C so I am pretty well insulated. Physics says this should be linear, I have not checked, at these low levels linear should be close enough. The adjusted ET - BT peaks at BT around 180-185C with 100C (so my MET is often 285C without tipping or any other problems) and ramps down as RoR is dropped and then thru 1st and can go down just below 10C into a rolling 2nd.

Perhaps one could run a dummy roast with with those inert beans and record the ET-BT and use that as an offset to see how much is just going into the roasting, excluding some of the specific heat of the beans.
Eric
 
Dan
Top  #44 Print Post
Posted on 09/26/2013 10:13 AM
Thanks for Schenker, I'll take a look, it is very long. There was an earlier test using thermocouples inserted in the bean, but I can't find it.
1 pound electric sample roaster, 3 pound direct-flame roaster, both handmade; modified Mazzer Mini grinder, LaSpaziale Vivaldi II automatic espresso machine. When the electricity goes out I make vacpot coffee from beans ground on my Zassenhaus hand grinder, and heat the water with a teakettle on the gas range.
 
Web
boar_d_laze
Top  #45 Print Post
Posted on 09/26/2013 5:35 PM

Quote


eric wrote:
In an air roaster ET measures the air temp before it gets anywhere near the beans and airflow is one way from the ET measurement to the beans, so its difference with the bean pile temp is a good indication of beans overall absorbing or releasing heat.


ET is taken in some place before the airflow gets "anywhere near the beans, and aiflow is one way.." You can't have it both ways. Since ET is measured far from the beans, we have no way of knowing what the ET is near the beans; and therefore, the ET as measured cannot be used as a reliable indicator of what's going on with the beans.

Since exothermic reactions are a widely observed, well reported, peer-reviewed, and accepted part of coffee roasting, I think you carry a burden of producing compelling evidence if you want to deny it. With great respect, I don't think you approach that standard.

BDL
Edited by JackH on 09/27/2013 11:23 AM
USRC 1lb Roaster, Chemex+Kone, Espro, Various FPs, Royal Siphon Vacuum, Yama Ice Drip Tower, Bunnzilla, La Cimbali M21 Casa, Ceado E92.
CookFoodGood
 
JETROASTER
Top  #46 Print Post
Posted on 09/27/2013 11:07 AM

Quote

boar_d_laze wrote:



With great respect, I don't think you approach that standard.

BDL


Care should be taken in all our conversations to inspire more conversation. Personal statements should be replaced by questions to foster more discussion.
Simply making a statement of "all due respect" is not the same as being respectful. Carefully crafting a response is respectful.
-Cheers, Scott
The culture at HRO is far more important than any thread or opinion.
 
eric
Top  #47 Print Post
Posted on 09/27/2013 7:59 PM
More thoughts: all I think one can conclude from Schenker's work is that there are no exothermic effects in isothermal or commodity roasting scenarios. I do believe that one needs to measure the scenario that one is interested in and check for exothermic effects in that particular scenario.

I did want to ask Dan what wonderous flavours he gets when roasting in a pressure cooker? Does pregrinding the beans before roasting so that the usual 10+bar core pressure is avoided create even better flavours? Sorry could not resist - that is what I think you are proposing by submitting those papers on exothermic coffee roasting. Please come back if you think that is not correct!

Just after posting above I realised that I already had all the information I needed to completely take the roaster out of the equation - no one offered this suggestion: since we have a step function in ET all one needs to do is fit the exponential to the BT, get the time constant, work out what fraction of the specific heat load is roaster vs coffee and split appropriately. Then the delta T attributable to the RC is RoR * tor (units are C). In the case above even though the beans look charcoal I took them as pure cellulose - they have probably been at 220C+ for a couple of hours and have had all their water and oils and other volatiles evaporated out and no further chemical reactions taking place (checked roast loss 37%!) - cellulose has 3x the specific heat of stainless and given my low mass RC 15% of the heat was going into the RC, the rest into the beans. So I have something for the next software rev. I really don't care what the scale of the dial for power into the beans is, just that it is linear and has no offset.

I have been running stall and tipping protection on my controller for ages, it is simple ET control that overrides the much more complex BT control if power into the beans drops to zero (force adj ET - BT > 0) or if ET exceeds a preset, currently set to 300C to also protect the roaster, though all its parts are rated to 400C. If I ever went exothermic there would be a runaway and that has not happened. This protection has meant that through all the R&D I have not had one undrinkable roast, despite some pretty unstable BT control!

Given that so many people use the exothermic idea to help them with their roasting, I still believe that it is the coming down from a very high effective specific heat prior to 1C that is what is happening, not that the beans go exo during or just before 1st. I previously attributed the high power requirement for the 180-185 band as peaking in the Maillard reactions (core temps would be a bit lower) , and that is probably a contributing factor, however I hypothesise that it is the most heat demanding process in roasting that is the cause: latent heat of vapourisation of water. It has been noted that pressures with a bean can rise to 15 bar during roasting, - perhaps that is a tad above for if we say 10-12 bar then the last of the water turns to steam at 185C and all of a sudden at that point the apparent specific heat of the beans drops significantly. I believe it this effect that is the cause. I have not read it anywhere else, so I cannot support it but that does not mean that it is not correct.

Quote

boar_d_laze wrote:

Quote


eric wrote:
In an air roaster ET measures the air temp before it gets anywhere near the beans and airflow is one way from the ET measurement to the beans, so its difference with the bean pile temp is a good indication of beans overall absorbing or releasing heat.


ET is taken in some place before the airflow gets "anywhere near the beans, and aiflow is one way.." You can't have it both ways. Since ET is measured far from the beans, we have no way of knowing what the ET is near the beans; and therefore, the ET as measured cannot be used as a reliable indicator of what's going on with the beans.

Since exothermic reactions are a widely observed, well reported, peer-reviewed, and accepted part of coffee roasting, I think you carry a burden of producing compelling evidence if you want to deny it. With great respect, I don't think you approach that standard.

BDL
Hey BDL that is not relevant as if there is nothing to cool the air between the ET measurement and beans then the reading is meaningful. The great thing about the way I measure it above all these errors are taken out anyway. Even bad mixing, as long as it is not too bad.

I know that the numbers I mentioned above for my last rebuild are showing a slight problem in the mixing as I usually get delta of 1-2C @ 130C and the problem I know is that I accidentally slightly misplaced a baffle in the mixing chamber, but that is not a problem!

BTW that is a pretty new roaster! ThumbsUp
Eric
 
Lawnmowerman
Top  #48 Print Post
Posted on 09/28/2013 11:02 AM
I may be oversimplifying this. If you can kill the heat and keep the beans agitating, and an increase in temperature is observed, then wouldnt thst make them exothermic? This has been reported to me by my bean supplier, a probat user, and i have confirmed it@ home with my breadmaker roaster.
Bad coffee prevails when good coffee roasters stand by and do nothing.
 
Page 2 of 2: 12
Jump to Forum: