Thermodynamic Help?

Yakima 728
Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
CKing

CKing

Well-Known Member
Joined
Feb 14, 2012
Messages
178
Reaction score
15
Location
North Canton
Been thinking more about fermentation temperature control. And was wondering if the following scenario is worth any merit along with how to accurately plan for a specific temperature range:

Chill Wort to 70 degrees (about the best I can do in Summer)
Basement is a constant 65 degrees
Plastic bucket primary placed in larger plastic bin with 10 gallons of water chilled to 60 degrees added (not a full blown swamp cooler set up)
Chilled water maintained with ice a couple times a day, but rising 1-2 degrees per day for the first few days

Without taking samples or using thermometers in the primary, is there a basic way to calculate how effective the heat transfer could be between the chilled water around the primary and it's effect to the temperature of the wort during the first few days of fermentation?
 
CKing,
This is my technique. I have asked the same Q previously (although not in a dedicated thread) but not gotten a definitive scientifically based answer. Best consensus seems to be that the thermal mass will help moderate temps., but there still will probably be a temp. rise in the center mass of the fermenter. I have a fermenator strip on my bucket, and a floating aquarium thermometer in the swamp cooler. Granted neither are terribly accurate, but I can keep fermenting temps about 6-10 degrees below environmental temp, depending on how many ice bottles I use and how often I change them out.
I'm going to keep my eye on your thread to see if anyone answers with a better answer.
 
Yeah, I'm looking more for numbers, factors, equations, etc. that would apply to this method.
9 months out of the year I don't really worry since the tap water is colder and my wort chiller can get the wort down to low-mid 60's
It's the summer months that I want to adjust for.

Cheers!
 
There is obviously some way to calculate this, but I see it being freakin complicated for nothing.

To do it properly, you'll be looking at the water temp and wort/beer temp, the room temp (as that will effect how fast your water warms up, and humidity might even play a role), the material of your fermenter (glass or plastic, and how thick?), the water level etc... There are so many factors to take into account that you'll probably never really be accurate.

Instead, just use the method you want and keep track of the temps in the fermenter. Two or three tries down the road, you'll probably have it dialled in. Sure, the first one won't be perfect, but it'll work.

I did that for my most recent batch. Got wort down to 75 with my immersion chiller, then rotated my frozen bottles regularly. I kept my wort between 62-64 (based on fermometer on my FV) for 5 days without any trouble at all, and after that I slowed down and let it rise to 68. And I kept it there with very, very little effort afterwards.
 
I agree with the previous poster that it's a lot of complication for nothing.

Get a thermowell and a thermometer. Measure and adjust the cooling water by adding ice as necessary.

My intuition (I am an engineer) is that you'll probably be 2-4 degrees higher in the center of the fermenter than your cooling water during days 1-4 of fermentation. Both plastic and glass are good insulators so don't expect a homogeneous temperature.

Since you're dealing with a lot of thermal mass your temp changes will also be relatively slow so record what you do and use it to establish a better base line for next time. You'll probably need a few pounds of ice on each of the first few days, then quickly won't need any as fermentation finishes and you'll be able to remove the fermenter from the water.
 
For the time and effort it takes to reliably calculate this and design a system that achieves your desired results consistently, would it not be simpler (and maybe even cheaper ofer the long haul) to pick up a used refrigerator and buy or build a temp controller?

For $50 total investment, I have been brewing in the same fermentation chamber for 5 years (free used refrigerator and a $50 analog Johnson controller). My results with that are consistently more stable and more repeatable than any of my previous efforts with swamp cooling.
 
The problem with calculating this is the lack of important data. Here are factors you'd need to take into account:
= the exothermic reaction of fermentation. You'd need to know how much energy that produces... which is no doubt variable over time.
= the conductive properties of the materials of the containers and their thicknesses. You can get close answers online.
= an accurate way to measure out cold water or ice to match heat absorption in the ice bath.
= the conductive properties of the concrete or carpet or whatever of your basement floor (assuming that's where it's sitting).
= accurate data for the room temps throughout the day and night.

The math isn't that tricky... it's data collection and constant monitoring that is the hard part. The heat produced by fermentation would be so variable based on so many contributing factors that it would make prediction almost impossible. Not to mention you'd have to constantly be metering out coolant.

My solution was an ice bath that pumps cold water into an insulated bucket holding my fermenter. I use a thermal controller to monitor and regulate the temperature by turning the pump off.

Good luck, man. I tried doing the same calculations.
 
Though if you want to know how the thermal properties of the water around your fermenter work the basic principles of equilibrium will give you a rough idea of what to expect.

The fermenter produces heat from fermentation and your ambient air temps are both getting rid of heat into your water tub. That means eventually that water will become at least 65 degrees given your ambient in your basement. Which means your fermenter (fermentation heat aside) will also eventually reach 65. Your thermal mass of ten gallons means that it will take longer for everything to reach 65. Now add in the energy produced by fermentation and it will get an unknown quantity higher.
 
Cking, if your basement is 65F you don’t need much cooling.

If you have 5 gallons of beer at 70F going into 10 gallons of water at 60F, The system will equalize fairly quickly at about 63 1/3. Then it will drift slowly up to 65. In a ten pound batch there’s enough fermentation heat to melt about 2 ¼ pounds of ice, but that takes place over about three days.

If you’re fermenting at 65, we’re done. If you want it cooler, then obviously you need more ice. How much is hard to figure, too many variables. If you want to ferment at 60-62, I’d guess you need another two pounds of ice per day.

YMMV, but there’s your ballpark.

I have to quibble with schematix a little bit. I ferment with a carboy in a water bath and the beer is always less than 1F warmer than the water. That’s with the water level with the beer.

The thermal conductivity for glass is close to 1, water is .6, plastic .4-.5. In this application that’s pretty much no insulation at all.
 
CKing said:
Been thinking more about fermentation temperature control.
Click to expand...

I took the title literally and though you had a real nasty thermodynamics question (something about the change in Gibbs energy during fermentation or something equally unpleasant). Application of the first law will do here.

CKing said:
Without taking samples or using thermometers in the primary, is there a basic way to calculate how effective the heat transfer could be between the chilled water around the primary and it's effect to the temperature of the wort during the first few days of fermentation?
Click to expand...

No and some of the reasons for this have been pointed out by other posters. To put the first law to work you want intimate thermal contact between the fermenting beer and the cooling water. This could be done by using a metal fermenter (keg - corny or otherwise) inside a larger container through which cool water is circulated/maintained or you could use a plastic or glass fermenter with an attemperating coil of copper or aluminum. As the metal is a good conductor you can assume the temperature of the beer is close to the temperature of the cooling water. As cool beer is denser than warm the beer close to the walls of a metal fermenter or close to the coils in an coil system will sink and this causes circulation so the fermenting beer temperature, while not strictly uniform, won't range over a wide set of values either.
 
First Law? That’s a stretch. This ain’t exactly a closed system. Heat is leaking off to the air.

But to your (AJ’s) second point I completely disagree. If I have a carboy in water, it’s essentially one big system. The thermal conductivity of glass is actually greater than that of water. Whether it’s metal or glass doesn’t make any difference. Plastic would slow it down a bit, but not in any practical sense, I would imagine.

Thus to answer the OP’s musical question,
Without taking samples or using thermometers in the primary, is there a basic way to calculate how effective the heat transfer could be between the chilled water around the primary and it's effect to the temperature of the wort during the first few days of fermentation?
Click to expand...
The answer is yes because they are pretty much the same. If you want to get fancy, dial in 1 Fahrenheit degree as an offset for those first few days. Better a little cool than a little hot.
 
Wynne-R said:
First Law? That’s a stretch. This ain’t exactly a closed system. Heat is leaking off to the air.
Click to expand...

The laws of thermodynamics do not apply only to closed systems. Closed systems are just easier to analyze. The first law says that systems that are at different temperatures will exchange energy in the form of heat until they are in equilibrium (at the same temperature). That's what we want to realize here - beer, which we don't want to instrument, in equilibrium with a mass of cooling water, which, I assume, we don't object to instrumenting. The way to work this is to get the two in close thermal contact i.e. with nothing separating them but something which conducts heat well. Metals are good conductors of heat. Plastic and glass are not. Heat will be interchanged between beer and coolant much faster, and equilibrium reached quicker, if the intervening medium is a good conductor.

Wynne-R said:
Heat is leaking off to the air.
Click to expand...
My intent here was to simplify away that actuality by having the beer in such close thermal contact with the coolant that heat loss/gain from the air, floor etc. would be quickly made up by heat flow through the separating wall from/to the coolant.


Wynne-R said:
But to your (AJ’s) second point I completely disagree. If I have a carboy in water, it’s essentially one big system.
Click to expand...
If I'm understanding this correctly it's a carboy in a water bath. One would model this as a mass of wort containing a heat source separated from a mass of water by either an insulator or conductor but in any case a thermal impedance. To cool the beer the heat generated by the yeast must be carried away and, if the temperature is to be lowered to some value below say room temperature, more heat still must be removed. This would be done by connecting the outer container (or coil as I suggested) to a cool 'reservoir' i.e. one whose temperature is maintained constant by addition of ice, a chiller... If there is heat flow from the beer to the coolant there will be a temperature drop across the separator. If the separator is highly conductive (low thermal impedance) the drop will be small and the temperature of the beer will be close to that of the coolant. If the separator is high impedance the drop will be large and the beer will be appreciably warmer than the coolant. The time taken to reach equilibrium will be much less with a good conductor than a poor one.

In a typical brewery scenario the device I'm describing is a cylindroconical fermenter. In those freezing or subfreezing glycol is circulated through a jacket which surrounds the internal fermentation tank. Because the wall of the tank is metal the thermal impedance is low and heat flows out quickly. It is not intended that the beer reach equilibrium but rather that the beer close to the wall (and coolant) gets quickly cooled causing it to sink and set up the currents which mix it with the rest of the beer and keep the yeast in suspension. The principle of operation would be no different in a system in which a Corny keg, the fermenter, is immersed in a garbage pail full of water which is circulated at some temperature set by external control.

Wynne-R said:
The thermal conductivity of glass is actually greater than that of water. Whether it’s metal or glass doesn’t make any difference.
Click to expand...

The thermal conductivity of copper is about 400 times that of glass and of iron is about 80 times that of glass. Whatever the heat flow the drop across a cooper separator would be 1/400th that across one of glass of the same thickness. That would make quite a difference as long as there is a heat flow. Once the heat load is gone (fermentation complete) then it doesn't much matter unless the loss/gain from the ambient is appreciable. Keep in mind that much, if not most, of the transfer of heat inside the fermenter is by convection.
 
Let me go all AJ on you.

No and some of the reasons for this have been pointed out by other posters.
Click to expand...

Theoretically, but not in any practical sense.

To put the first law to work you want intimate thermal contact between the fermenting beer and the cooling water.
Click to expand...

Actually the first law does not require “intimate contact”.

This could be done by using a metal fermenter (keg - corny or otherwise) inside a larger container through which cool water is circulated/maintained or you could use a plastic or glass fermenter with an attemperating coil of copper or aluminum.
Click to expand...

No “attemperating coil” is necessary.

As the metal is a good conductor you can assume the temperature of the beer is close to the temperature of the cooling water. As cool beer is denser than warm the beer close to the walls of a metal fermenter or close to the coils in an coil system will sink and this causes circulation so the fermenting beer temperature, while not strictly uniform, won't range over a wide set of values either.
Click to expand...

That also happens in the plastic or glass. Not that it needs to, the fermenting wort stirs it up pretty well.

Any insulating properties of the plastic or glass can be safely ignored, because their effects are insignificant.

Everything AJ said is technically correct but pretty irrelevant. It would take a volcanic fermentation to heat up the beer much if it’s setting in ten gallons of water.
 
I'm not sure what the point of all this is and there isn't much point in rehashing what I've already posted so let me pose a gedanken experiment which may help you to see what I'm trying to get across. You have 4 capacitors of equal size. Two are charged to 50 volts and two are charged to 100 volts. You make two pairs each consisting of a 50 volt and 100 volt cap. You connect the negative terminals of each pair to one another with wire and the positive terminals of one pair with a 1 ohm resistor and the positive terminals of the other with a 400 ohm resistor. Measuring the voltage of each pair, which comes to 75 volts faster? Is the size of the resistor irrelevant? You may think the whole problem irrelevant but it isn't. It is an excellent model for what happens when you cool or heat one mass by connecting it thermally to another.

I do understand what you are trying to say (I think). Once things are in equilibrium there is no heat flow and therefore it doesn't matter what the connecting impedance is. It is when you try to change (lower) temperature or when the ferment is producing heat (and in small ferments this isn't a problem - in larger volumes it is that these effects become important.
 
A.J. I’m not sure what else to say either. You’re going on about things that are too small to measure with ordinary instruments and you can’t seem to fathom that I call them insignificant.

I ferment with 5 gallon carboys in approx 6 gallons of water. The temperature difference between thermowell and water bath is usually .5 C at the most (max krausen). If the carboy were copper, what would it be? .46 C?

Your model is transparently flawed. I propose to change the voltage difference to 1 volt, substitute the capacitors for batteries. The internal resistance is 10K ohms. The resistors are 400 and 1.

As for equilibrium, I guess we’ll hit it pretty quick once the battery (fermentation heat) dies.
 
I hate to say anything, but I see what AJ is saying and I believe he is correct. That said I'm merely an engineering student with only a basic grasp of thermo. His concern about intimidating the heat source with the coolant is valid- we want to increase the rate at which the heat energy passes to the water. The metal coils are a clever work around by, ideally, decreasing the temperature of the water closest to the carboy or bucket walls.

I appreciate the discussion... there is much to gain from this.

AJ, I'm very interested in your background.
 
I thought perhaps I didn't understand what the OP was asking so I read it again. Here's what he said:

CKing said:
Chill Wort to 70 degrees (about the best I can do in Summer)
Basement is a constant 65 degrees
Plastic bucket primary placed in larger plastic bin with 10 gallons of water chilled to 60 degrees added (not a full blown swamp cooler set up)
Chilled water maintained with ice a couple times a day, but rising 1-2 degrees per day for the first few days

Without taking samples or using thermometers in the primary, is there a basic way to calculate how effective the heat transfer could be between the chilled water around the primary and it's effect to the temperature of the wort during the first few days of fermentation?
Click to expand...

I interpret that to mean that he wants to cool 70 degree wort to 60 degrees and wants to know if he can monitor the progress of this without taking measurements in the beer IOW he wants to take measurements on the cooling water and wants to know if there is a simple way to interpret those. Guess it depends on how you define simple. If he knows the mass of the beer and the mass of the coolant he can assume conservation (relatively small loss to the ambient) and calculate the drop in wort temperature from the rise in coolant temp. I would consider that a relatively simple thing to do but he might not so what I suggest is that he put a small impedance between the two such that equilibrium is reached quickly at which point he can assume that beer temp and coolant temp are the same.

If you'll look at the last picture in at this page: http://wetnewf.org/pdfs/estimating-mash-ph.html you'll see what happens when you do something like this. When a Congress mash is done the use of metal (as opposed to glass) beakers is specified for the reasons we're discussing. You want the mash to come quickly to the temperature of the water bath. When I do malt titrations I follow this recommendation. The graph shows pH history as the electrode is immersed first in 4 buffer then in 7 buffer and then in malt to which water plus acid is added. The green trace is the temperature. The buffers are in plastic containers. As is clear from the picture it takes several minutes (10?) for the buffers to come to water bath temperature. OTOH the mash in the metal beakers comes to water bath temperature in about a minute. These temperatures are measured with ordinary instruments.

I ferment with 5 gallon carboys in approx 6 gallons of water. The temperature difference between thermowell and water bath is usually .5 C at the most (max krausen). If the carboy were copper, what would it be? .46 C?
Click to expand...

At equilibrium, ignoring paths to the ambient the beer and wort should be at exactly the same temperature regardless of the material because at ambient there is no heat flow. I think this may be what you are missing. OP wants to cool wort, at least initially. Glass has (depending on composition) a conductivity of about 1 W·M^-1·K^-1 and copper 400. A temperature difference of 10 °F is about 6 K so 1200 joules would flow through a 0.5 cm carboy wall of area 1 m^2 per second were the carboy glass but 800 times as much would flow through a copper one (400*2 as the copper vessel would be thinner). Whatever the actual areas and masses the 'time constant' (I assume you are familiar with that term) is 800 times larger for a glass carboy than a metal one. I think you may be arguing the same point that I am - that at equilibrium the temperature of the beer is essentially the same as that of the water bath. But you are forgetting that it takes time to get there and that that time depends on the thermal impedance. If time constant is unimportant then it doesn't matter and once temperature is reached then it doesn't matter any more and perhaps that's your point. In my brewing time constant does matter - I want the wort cooled to proper operating temperature as quickly as possible.

Your model is transparently flawed. I propose to change the voltage difference to 1 volt, substitute the capacitors for batteries. The internal resistance is 10K ohms. The resistors are 400 and 1
Click to expand...

It's an excellent model for the cooling of wort by contact with a cold body (not reservoir) of water prior to pitching. Perhaps you didn't like the numbers. Make them (60-32)/1.8 + 273 volts for the water and (60-32)/1.8 + 273 volts for the OP's wort. Multiply the thermal impedance of the separating wall by the thermal mass of the wort and chose convenient capacitor size and choose the resistors to give the same time constants and the thermal time constants and you have a model so good that it was the way problems like this were solved in the days of 'analog computing'.

Substituting batteries for the capacitors does not make for a good model as they do not discharge/charge the way a capacitor does which is the same as the way in which a thermal mass responds to heat loss or gain. To model the yeast's energy input you would use a current source on the wort capacitor side. The coolant side can be replaced by a voltage source only if the coolant side is a cold reservoir (i.e. a chiller which holds a particular temperature).

With the proper model current from the current source charges the wort capacitor at whatever rate the yeast produce heat and the capacitor's voltage rises. The voltage difference between the wort capacitor and coolant capacitor causes current to flow from one to the other with the rate of current flow being proportional to the voltage difference. In thermal terms, heat flows through the thermal impedance between wort and water at a rate that depends on the temperature difference and the thermal impedance. If that heat (current) flow is greater than the source current (yeast heat production then the wort cools. Otherwise it warms until the yeast current and conduction currents are equal and an equilibrium is reached which is maintained until the yeast current source is turned down (fermentation rate is declining) and eventually off after which time a new equilibrium is reached.
Whether the length of time for equilibrium to be reached is quick or slow depends on the time constant (thermal impedance) and thus the material of which the fermenter is made. As a practical matter the rise is small at home brew scale and, of course, as the yeast 'battery' doesn't get shut off abruptly, unless the brewer drops the temperature dramatically by lowering the temperature of the coolant radically, the rise due to the yeast output is small at the end of fermentation.

To return to the original question as to whether OP can put a carboy of 70 ° wort in a larger container of 60 ° water and deduce the temperature of the wort from the temperature of the water the answer is 'Yes after a couple of time constants'. How long a couple of time constants is depends on the material of the carboy. Can he then 'crash cool' the wort if he wants to and deduce the temperature of the wort from the temperature of the cold water? The answer again is 'Yes, after a couple of time constants.'
 
DSorenson said:
AJ, I'm very interested in your background.
Click to expand...

I am a retired electrical engineer with experience in high power rectifier design (that's where I learned these thermal models), radar, signal processing, estimation, communications, orbit mechanics and rf engineering. Out of practical necessity I've also done diesel mechanic, electrician, refrigeration and plumbing. IOW if it broke in the field I fixed it (if I could). Brewing turns out to be a great hobby for application of lots of the stuff I picked up as an engineer.
 
Ha ha, AJ- I should have guessed. Well done! I think you've made safe assumptions and accounted for significant contributing factors.

One thing I don't understand Wynne-R:
You said fermentation energy could melt 2 1/4 lbs of ice a while back.
Could you explain a little about where that number comes from?
How would you apply that information?

I only ask because a block of ice and ice cubes of the same volume melt at different rates and cool at different rates. The application of this is what I'm most curious about: I use way more than 2.25 lbs of ice in the first 4 days using a very insulated (on all sides, top and bottom) water bath to maintain a delta t of 15 degrees lower than ambient.
 
Dsorenson, I figured 141 kj/kg * 2.2 kg extract, and 80 cal/g heat of fusion for the ice. It was very back of the envelope, guessing that the 17 or so cal/g from heating the melted ice would be offset by other losses.

I didn’t whip out the numbers because most people don’t care and that it was so quick and dirty. It’s a guess.

The application would be to add that ice gradually over the course of a few days, in addition to whatever ice would be needed to pull his delta t on the water bath.

I have a Rubbermaid ice bucket that pumps water through an immersion chiller controlled by an STC-1000. I use about 8Kg of frozen water bottles per day rockin’ about 10C delta.

It’s a tub with 6" of fiberglass wrapped around it. It’s open on top and the bottom is on concrete. My setup is fairly inefficient but the regulation is awesome. I think your water bath losses are probably higher than you think.
 
CKing said:
Chill Wort to 70 degrees (about the best I can do in Summer) Basement is a constant 65 degrees Plastic bucket primary placed in larger plastic bin with 10 gallons of water chilled to 60 degrees added (not a full blown swamp cooler set up)
Click to expand...

I chill my wort to 70 and pitch yeast (usually US-05 or S-04). I'd chill it down further, but I'm too impatient.

I then place an 18G storage tub onto the cold concrete floor of my 65 degree basement, place the bucket / plastic carboy into the storage tub, and use a hose to fill the tub with 60 degree water up to the water line in the bucket or carboy.

If it's a bucket, I'm done. If it's a carboy, I put on a blowoff tube into a mason jar of starsan and cover with a towel to keep sunlight from skunking my beer.

And that's it. No ice, no fan, no tee-shirt. It doesn't take long for the water bath to get up to ambient, and at high krausen it might be 3 degrees above ambient (measured against the side of the carboy), but never more. I'll check it a couple of times per day for the first few days just to make sure everything is okay.

After a week or so, I bring the beer upstairs where it's 70 for a week or so.

One of these days I will build a temperature controlled fermentation temperature, but until then this procedure seems to produce perfectly drinkable ales.
 
AJ, Nowhere in your analyses have you counted for the thermal impedance of the water. That’s why we’re in different universes. I can’t make my way up your airless frictionless plane.

Your magic water apparently has a thermal conductivity of infinity, because the only resistance you’re counting is the container. The heat in effect doesn’t have to travel through the water, it’s just there.

Back in my universe, suppose I have a heat source in a carboy. I turn it on for a few seconds. Then I wait for the thermometer in the water bath to get the news. It takes a while, but eventually I get a stable temp reading. Now I try this again with an equivalent mass of water substituting for the carboy and beer. Theoretically this takes a little longer, but the difference is slight and I don’t notice it with my thermometer and my stopwatch. Now we go for the darkhorse, the copper carboy. And the results are . . . No measurable difference.

Despairing for the First Law, I try it again with a Rubbermaid cooler. We have a winner! That takes forever.

Now if I had lab grade equipment, Copper would have won, Glass second closely followed by Water, with Cooler hardly out of the gate. But I don’t have lab equipment. I make wort in the kitchen, and the beer has the laundry closet. The yeast are happy and I’m tickled spitless.

Still in my universe, I go to work building AJ’s model. Wait, something’s wrong. I’m getting the same goofy results that AJ was. I used the current source, the capacitors and the resistors. Changing the model, I add series resistors to the capacitors to model the thermal resistance of the water. Now the times are getting close but the voltage is high. I put a resistor in parallel with the water bath capacitor to model heat loss to the air. Now we’re talkin’.

Having pondered the strange Physics of an alternate universe, I settle back for a well deserved beer. Maybe a nap.
 
Wynne-R said:
AJ, Nowhere in your analyses have you counted for the thermal impedance of the water. That’s why we’re in different universes. I can’t make my way up your airless frictionless plane.
Click to expand...

No, I haven't, nor would I as it's not necessary to. Besides which in answering the OP's question
...is there a basic way to calculate how effective the heat transfer could be between the chilled water around the primary and it's effect to the temperature of the wort during the first few days of fermentation?
Click to expand...
complicating the model further only makes it clearer that you basically can't do calculations until equilibrium is reached at which time a simple measurement will suffice.

Nevertheless, a conductivity model for the water isn't a good one as heat in this situation is heat from the yeast would be transferred within the wort mostly by convection. But the heat source isn't a piece of current carrying wire down the axis of the fermenter. Is is trillions of little heat sources (yeast cells) distributed fairly uniformly throughout the mass of the wort and the rising CO2 column and the existence of a cold surface (the one between the water and wort) gives rise to currents which mix this stuff up pretty well. This means that one could fairly well model things as a discrete line heat source in a perfectly conducting medium! Never thought of that. Remember, though, that the object here isn't to develop a model but rather to use a model to gain insight. There is no need to model each yeast cell as a separate current source and its volume of wort as a separate capacitor with all the capacitors interconnected by resistors.

Wynne-R said:
The heat in effect doesn’t have to travel through the water, it’s just there.
Click to expand...
In this case as the heat source is distributed that's true, it's pretty much there. Plus if you have ever heated water on a stove you know that heat propagates very rapidly throughout the volume but that is primarily by convection (you can see the currents in the water) not by conduction.

Wynne-R said:
Back in my universe, suppose I have a heat source in a carboy. I turn it on for a few seconds. Then I wait for the thermometer in the water bath to get the news. It takes a while, but eventually I get a stable temp reading. Now I try this again with an equivalent mass of water substituting for the carboy and beer. Theoretically this takes a little longer, but the difference is slight and I don’t notice it with my thermometer and my stopwatch. Now we go for the darkhorse, the copper carboy. And the results are . . . No measurable difference.

Despairing for the First Law, I try it again with a Rubbermaid cooler. We have a winner! That takes forever.
Click to expand...

I'm afraid this isn't very coherent and I can't follow it. Again, probably for the 10th time, I'll state that once equilibrium is reached then there is no flow and it doesn't matter if the container is silver or glass.




Wynne-R said:
Now if I had lab grade equipment, Copper would have won, Glass second closely followed by Water, with Cooler hardly out of the gate. But I don’t have lab equipment. I make wort in the kitchen, and the beer has the laundry closet. The yeast are happy and I’m tickled spitless.
Click to expand...

Lab grade equipment has nothing to do with it. You should be able to do simple experiments to confirm these results with a tub full of cold water and first a carboy full of hot water and then a corny keg full of hot water. See which cools faster.

You might also ask yourself why ASBC/IBC specify metal beakers instead of glass ones for the Congress mash procedure. You might also ask yourself why people measuring mash pH cool the sample in a metal container rather than a glass one immersed in cold water when trying to get the sample cool as fast as possible. This also suggests simple experiments you could do. You could, for example put a bottle of beer and a can of beer in an ice bucket and see which gets cold first.


Wynne-R said:
Still in my universe, I go to work building AJ’s model. Wait, something’s wrong. I’m getting the same goofy results that AJ was.
Click to expand...

A.J. didn't get any goofy results. He showed some real data from a similar situation in the real world (that might or might not be practically applicable here because a few mL of mash in a beaker in a water bath isn't the same as 5 gal of wort in a carboy in a wash tub) and described a model commonly used to explain what that data shows.

Wynne-R said:
I used the current source, the capacitors and the resistors. Changing the model, I add series resistors to the capacitors to model the thermal resistance of the water.
Click to expand...

Yes, that would give you goofy results. You would have to have a huge array of capacitors and resistors and the sizes of those resistors would be much smaller than just the conductivity of water suggests because of the convective effect. This is, IMO, for practical purposes, not modelable except empirically.


Wynne-R said:
Now the times are getting close but the voltage is high. I put a resistor in parallel with the water bath capacitor to model heat loss to the air. Now we’re talkin’.
Click to expand...
No, we aren't. The correct model for loss to the air is an impedance to an ideal (0 impedance) voltage source (the ambient is a cold reservoir).

Wynne-R said:
Having pondered the strange Physics of an alternate universe,
Click to expand...

If you are going to understand this you'll have to stay in this one.
 
Thanks AJ, finally you have sort of addressed what I’ve been saying.

First off, I want to put ‘equilibrium’ in to the corner as the petulant child it is. Is an active fermentation in a chilled water bath in equilibrium? Not in my universe.

Without taking samples or using thermometers in the primary,is there a basic way to calculate how effective the heat transfer could be between the chilled water around the primary and it's effect to the temperature of the wort during the first few days of fermentation?
Click to expand...

Does that sound like equilibrium to you? Really?

My answer to the OP’s question is, I have measured the difference and it’s about a degree Fahrenheit in glass. Perhaps somebody can write in and tell us what it is in plastic. I think it’s going to be pretty much the same. I suppose AJ would predict that in metal it would be vanishly small.

So the magic water is caused by yeast maintaining a completely heterogeneous environment. Sounds like a theoretical construct to me. Is there magic yeast in the cooling water too?

AJ I like you and it’s fun arguing with you. It seems you don’t take things personally. I certainly hope so. Civil discourse is increasingly rare. BUT (saw that coming, didn’t you?) your model is silly and your analysis is largely irrelevant.
 
Wynne-R said:
First off, I want to put ‘equilibrium’ in to the corner as the petulant child it is. Is an active fermentation in a chilled water bath in equilibrium? Not in my universe.
Does that [active fermentation] sound like equilibrium to you? Really?
Click to expand...

As I've noted before there do seem to be differences between the physics in your universe and the one that most of us think we live in but no, strictly speaking an ongoing fermentation is not in equilibrium but it can approximate steady state. For a period of hours the rate of the fermentation will not vary by an appreciable amount and the heat output will be essentially constant. This would allow temperatures throughout the system to approach steady state values asymptotically. And that is, of course, the approach our OP should take. He should monitor coolant temperature until it stops changing appreciably. At that point he could conclude that the wort and coolant are at close to the same temperature. They will never be at exactly the same temperature because steady state is only reached after an infinite number of time constants provided the fermentation is really steady state.



Wynne-R said:
My answer to the OP’s question is, I have measured the difference and it’s about a degree Fahrenheit in glass. Perhaps somebody can write in and tell us what it is in plastic. I think it’s going to be pretty much the same. I suppose AJ would predict that in metal it would be vanishly small.
Click to expand...

Theoretically if heat is being evolved at the time the measurement is being made then yes, the drop across a metal container will be smaller. I just cannot fathom why you cannot understand this. Apparently you have never worked with heat flow problems. This is very fundamental.


Wynne-R said:
So the magic water is caused by yeast maintaining a completely heterogeneous environment. Sounds like a theoretical construct to me. Is there magic yeast in the cooling water too?
Click to expand...
It's a model and one that turns out to be better than even I thought. It actually isn't the yeast though, it's the convection. And yes, convection is responsible for transferring heat through the water to the surface of the coolant where it is transferred to the air. I'd use the electric circuit model again but apparently you are not familiar with that.

Wynne-R said:
AJ I like you and it’s fun arguing with you. It seems you don’t take things personally. I certainly hope so. Civil discourse is increasingly rare. BUT (saw that coming, didn’t you?) your model is silly and your analysis is largely irrelevant.
Click to expand...

I try to help people out where I can not only because I'm a nice guy but it helps to keep my skills up. But there are limits. When one meets resistance no matter how logical or thorough the explanations one realizes that the desire to really learn is not there and that he is wasting his time. Whether you find the model silly or not is immaterial. But there may be others reading this and I don't want them to be misinformed. As most of them have been lost at this point I'm sure this is going to be my final transmission. If you sincerely want to understand this stuff there are plenty of sources you can go to and, as I suggested in an earlier post some simple experiments that you can do in your kitchen with a dial thermometer could help you to see how things work.

So I'm going to sign off with the results of a simple experiment I did in my kitchen (but I didn't use a dial thermometer). I took two containers, one of stainless steel and one of some sort of ceramic (my SO didn't have anything of glass I could use) and put them in a 'water bath' at 13.6 °C. This was the kitchen sink with the drain plug in place and the water running fast enough to keep the water level constant. The two containers were of approximately the same size and I put equal quantities of water in each, put them into the water bath and recorded temperature. The results are in the graph below. The data exhibit the classic model that I have been trying to get you to understand. As they fit the classic model so well we can compute time constants for the two and get 27.5 minutes for the ceramic and 10.6 minutes for the metal. Even if you don't understand the concept of time constants, it should be clear from the data that our brewer is going to have to wait much longer for his wort to chill to close to bath temperature with the non-metallic container than the metallic one. This is the point I have been trying to hammer home. Eventually the water in both containers will come to the same temperature - the model also computes the asymptotes which were within a degree of the bath temperature which is pretty good for this crude data set.

As I mentioned in an earlier post, a complete model of active yeast in the fermenter would predict a temperature gradient from the center to the wall. I didn't mention this because you seem to be confused enough. But it should be there because heat has to flow from the center to the wall and whenever there is a flow, as in a fermenter or my experiment, there will be a drop. My intuition says that in a large fermenter, even with mixing (wort against the wall is cool and falls thus pushing a column of warm wort up the middle) this might be appreciable (a degree or so?) but that in a small carboy it wouldn't be. So all the measurements I made were made on the axis of these small containers but I did check at the walls to see if there was any drop. There wasn't. This says that the heat that is flowing from the center of the container to the wall is not flowing across an impedance large enough to register a drop on my thermometer (which reads to 0.1 °C).

If based on this you still think what I have been telling you is silly I would advise keeping that to yourself. Or you can PM me and tell me that you think it is silly but I wouldn't say so in public. I have to admit I felt a little silly doing it because I knew what the results would be but I didn't know what the actual time constants would be and that was sort of interesting. It is also interesting that the data fit the capacitor through resistor to voltage source model so well. This says that losses to the air (20 °C) were small.

WaterBath.jpg
 
Edit: deleted because I didn't read most recent post, but hey, don't be a dick aj, we're all tryin to learn.
 
The last post is chock full of learning opportunities including real world data and guidance on how to check wort temperature by checking the surrounding water. Just wait until the surrounding water temperature stabilizes.


???
 
AJ, in the beating a dead horse to death dept., I’m going to try one last time to make my point.

I have no trouble accepting the Physics. I don’t disagree with your professorial explanations. I have no problem understanding schematics. Dare I say it, I’m not an idiot.

Where we differ is that you like your model, and I don’t. For one thing, an ideal capacitor has no resistance. Water has quite a bit. A fermentor in a water bath is in no way an ideal system. Your model doesn’t fit.

You can probably guess my objection to the ceramic experiment. One thing baffles me though; You don’t have beer glasses? How do you drink beer?

You keep repeating the obvious, as if I don’t get it. Maybe if you keep hitting me over the head with it, the light will come on. No, you don’t get me. You refined your model, based on my objections. You admitted you were ignoring the thermal resistance of the water when I called you on it. You backpedaled on the heterogeneous heat in the fermenter. Then you act like I don’t understand high school physics.

Maybe it’s the perennial conflict between engineers and technicians. Once you take something off of paper and actually build it, it will have some unexpected behaviors. If you fall in love with the model, it will circle around and bite you in the butt.
 
Wynne-R said:
AJ, in the beating a dead horse to death dept., I’m going to try one last time to make my point.
Click to expand...

I said I was finished because of the dead horse aspect but in the spirit of brewer assisting brewer...

Wynne-R said:
I have no trouble accepting the Physics. I don’t disagree with your professorial explanations. I have no problem understanding schematics. Dare I say it, I’m not an idiot.[/QUOTE

I know you're not and that's why I can't figure out why you can't seem to grasp what I am trying to tell you.

Wynne-R said:
Where we differ is that you like your model, and I don’t.
Click to expand...

The reason I like it is because it works. If you look at the data I put up you will see that even with it's warts (some points don't fall exactly on the curve) the model fits the data very well.

Wynne-R said:
For one thing, an ideal capacitor has no resistance.
Click to expand...
The model, which again I reiterate fits the data very well as you can see is of an ideal capacitor connected to ideal voltage source. The purpose of a model is to accurately represent the real world to the point that one can use it to accurately predict what will happen without having to construct the system and also to give us insight as to how a system behaves. I took measurements every even minute i.e. 0, 2, 4... minutes. Clearly this model tells me very closely what I could expect to measure at odd minute (1, 3, 5..) intervals and it can predict quite accurately what the temperature will be after 60 minutes - way after I stopped taking data. It's a good model. I can see an objection to it if it didn't fit the real world but I don't think you can object just because you don't like it.

You may not like the model (though again I don't know why) but you will find it in any textbook that touches this subject, application notes for semiconductor devices etc. It is a well established model that has been verified over the years by thousands of engineers who have been using it to solve lots of problems and design lots of systems. I didn't invent it.

Wynne-R said:
Water has quite a bit
Click to expand...
.
As my experiment clearly showed conduction is not the means by which heat is transferred within the wort. It is convection. You can mention the thermal impedance of water as often as you want but for me to take it into consideration you will have to come up with some experimental data to show that it is a factor. Measurements that show a temperature gradient between the core of a fermenter and its walls would support your thinking. My experiment, done with a much smaller volume do not support your thinking.

Wynne-R said:
A fermentor in a water bath is in no way an ideal system
Click to expand...
. It doesn't have to be. It only has to be well described by the model.

Wynne-R said:
Your model doesn’t fit.
Click to expand...

The model fits my experimental data very well, as noted above. There is no denying this. You can argue that my small containers don't accurately model a carboy and you might be right. Theoretically there should be a gradient because convective transfer has an associated thermal impedance too. It is too small to see in my kitchen experiment. Perhaps it would not be with a carboy and probably would not be with a 30 bbl unitank.

Wynne-R said:
You can probably guess my objection to the ceramic experiment.
Click to expand...

Yes, but this tells me that you are missing the point. I have no data on the thermal characteristics of ceramics. They can vary all over the place. Adolf Coors used to package semiconductors in ceramics and presumably that packaging had high thermal conductivity but the ceramic container I used had an impedance of about 2.5 times that of the stainless steel one. What you should take from this is, what I think should be obvious, that different materials will present different thermal impedances and that, therefore, for a given heat flow, there will be different temperature drops across them.

Wynne-R said:
One thing baffles me though; You don’t have beer glasses? How do you drink beer?
Click to expand...
I don't! I spend the summer in Quebec where 1) the beers are pretty bad and 2) the ciders are really good. During the summer I drink cider. I think I've had one beer since I got up here. It was advertized as a wit which I love but this was a typical Quebec brew - harshly phenolic.

More to the point - I wanted something that would hold at least a quart or two of water and wanted to containers about the same size. These two things met those requirements. I don't have any metal beer glasses anyway. Were I at home I would have used metal and glass beakers of the same size.

Wynne-R said:
You keep repeating the obvious, as if I don’t get it. Maybe if you keep hitting me over the head with it, the light will come on.
Click to expand...
That's my hope anyway.


Wynne-R said:
You refined your model, based on my objections.
Click to expand...
No, I've stuck with the ideal cap model all along. I said that modeling the conductivity, should you want to do so, would involve a very complex model with thousands of caps to the point that it would not be a feasible model.


Wynne-R said:
You admitted you were ignoring the thermal resistance of the water when I called you on it.
Click to expand...
I did say I was ignoring it and gave reasons why I thought that was reasonable to do so. Experiment verified that it is.


Wynne-R said:
You backpedaled on the heterogeneous heat in the fermenter.
Click to expand...
Again, you didn't read or comprehend what I posted. I said that my experiments proved that convection alone was sufficient to allow us to ignore the effects of conductivity. I did say 'It isn't actually the yeast though, it's convection'. The yeast do, of course, provide the heat distributed throughout the medium so that a homogenous model is clearly the correct one but that, apparently, isn't a necessary condition for the simple model (and remember the OP was looking for a simple model) to work.

Wynne-R said:
Then you act like I don’t understand high school physics.
Click to expand...
These last three things are things you have misinterpreted. What am I to think?


Wynne-R said:
Maybe it’s the perennial conflict between engineers and technicians.
Click to expand...

The laws of physics are the same for engineers and technicians. The thing I always found fascinating about good techs (I mean the really good ones) was that they had models of their own. I had no idea what they were but they did what a model is supposed to do - allow you to predict the behaviour of a system.

Wynne-R said:
Once you take something off of paper and actually build it, it will have some unexpected behaviors.
Click to expand...
That's why a model has to be verified. As I indicated in my post of yesterday I felt a little silly verifying a model this basic but in doing so I found out some stuff I didn't expect to (i.e. that conductivity of water isn't important and that loss through water/air interface is not significant).


Wynne-R said:
If you fall in love with the model, it will circle around and bite you in the butt.
Click to expand...
A large part of my career was spent modeling communications systems so I'm pretty familiar with their (models) strengths and weaknesses. The biggest problem I have with them is the widespread tendency of the model maker to change the model until it give him the answer he wants. One example of this would be Michael Mann tweaking the principal components from his climate data until the medieval warming period disappeared. Another would be you attempts to change the basic RC circuit model to get the results you wanted.

1. If heat is transferred between 2 bodies there will be a temperature drop/rise across the separator given by the product of the heat flow with the thermal impedance of the separator. If heat is flowing, that rise will always be present unless the separator is a perfect conductor of heat.

2. If the bodies are connected by a separator that has twice the thermal impedance of that in another identical system the drop across the separator will be, for a given heat flow, twice what it is in the other system. If the drop across the separators in the two systems is the same the heat flow in the first will be half what it is in the second.

3. In such systems the time constant (relaxation time, time to come to steady state, time to equilibrium...) is proportional to the thermal masses of the systems and the thermal impedance connecting them. If one system has a separator with twice the thermal impedance of another it will take twice as long for temperature to settle out.

Those are the facts. They can be verified in thousands of ways. Accept them or not. I hope you do.
Click to expand...
 
Based on my apparently defective understanding of physics, I draw the conclusion that ‘some sort of ceramic’ is a better insulator than stainless steel based on AJ’s kitchen sink experiment. We don’t know what the thermal properties of the particular ceramic are, so it’s impossible to make a prediction. Is there some other possible conclusion that I’m missing?

It was pretty convincing that you saw less than .1 C across the radius, until I tried it myself. I’m seeing something like1 C. It’s hard to say since it’s a moving target. I boiled 1L of water in a 2 qt sauce pan with a 16cm inside diameter. I used a pretty fast digital thermometer at approx 50 C in a 25C sink waterbath.

We should probably drive a stake in this, having lost the OP a long time ago. I just wanted to toss this in because the ceramic thing is tied to this thread.

I’ll get some better numbers and come back to this in a new post. Straw men and circular reasoning are not to be allowed.
 
Wynne-R said:
Based on my apparently defective understanding of physics, I draw the conclusion that ‘some sort of ceramic’ is a better insulator than stainless steel based on AJ’s kitchen sink experiment. We don’t know what the thermal properties of the particular ceramic are, so it’s impossible to make a prediction. Is there some other possible conclusion that I’m missing?
Click to expand...

The conclusions are two:

1) The thermal impedances between two containers made of different materials (I don't know the properties of the stainless either) and a cool reservoir are different (one is about 2.5 times the other). By considering the amount of the water and the thickness of the walls one could calculate the actual thermal conductivities of the two materials.
2) The rate of heat transfer between the water in both containers and the cool reservoir are well modeled by an ideal capacitor connected to an ideal voltage source through a resistance.

Wynne-R said:
It was pretty convincing that you saw less than .1 C across the radius, until I tried it myself. I’m seeing something like1 C. It’s hard to say since it’s a moving target. I boiled 1L of water in a 2 qt sauce pan with a 16cm inside diameter. I used a pretty fast digital thermometer at approx 50 C in a 25C sink waterbath.
Click to expand...
As it is clear that there must be some impedance it should also be clear that the higher the temperature difference the higher the heat flow rate from core to wall and the higher the drop will be. Note that this impedance, if it is appreciable, will just add in series parallel to the impedance of the wall. This does not mean that the system will no longer be modelable as an ideal capacitor and resistor but rather that the size of the resistor will have to be bigger (and the time constant of the system longer).

Of course if there is an appreciable drop in a real system that raises the question "What is the temperature of the beer? Is it the temperature at the wall or the core?"
 
Ya' all are geeks.... I recommend fermenting at a temp that the beer will taste really good. :drunk:
 
Aj and Wynne: Thanks for the entertainment. I almost want to apologize to the forum for starting this(I was the 1st responder that's suggested that perhaps those with a scientific background would pipe in). But, to me anyways, this was entertaining. Coming from a medical background, I was good in Biology, Biochemistry, and Chemistry, but piss-poor in Physics back in college(30 years ago). However, I was pretty much able to follow your arguments and will be able to take from it what I can use practically.
Hopefully, Thus Endeth the Discussion. :mug:
 
So is the situation in my original post worthwhile?
Will the cooler water around the primary help bring the wort temp down before and during the first few days of fermentation to any noticeable extent?
Perhaps I should have been more straightforward...............anyway interesting 4 pages of "discussion" so far.

Thanks....I think ;)
 
Sure, it sounds like a good plan. Keep the water 2 degrees F below your target. Measure it at mid-height for a good average. With 65 F air it should be pretty easy to maintain a good ferm temp.

If AJ’s right is would be 2.5 degrees, if I’m right, a little less. I’ve come part way from my previous estimate of 1 degree, which is what it would be for glass. I hadn’t accounted for the heat being stirred up in the fermenter. I still think AJ’s wrong, but I was wrong too. The truth lies somewhere in-between.
 
Okay, you'll need the conductivity of the wort, water and air. You also need the thermal conductivity of the bucket and the larger bin. You will also need to the thicknesses of the bucket and the larger bin. Finally, the temps of the wort and the air are enough; however they aren't constant, so good friggin luck.

You'll need a heat transfer graduate student to figure this out.

Don't bother with it.
 
CKing said:
So is the situation in my original post worthwhile?
Will the cooler water around the primary help bring the wort temp down before and during the first few days of fermentation to any noticeable extent?
Click to expand...
Yes, absolutely and it happens pretty quickly if the container is metal and more slowly if the container is glass.

The model really isn't that complicated - until you try to account for convection. For 20L of wort at 30 °C placed in a bucket of 20 °C water with only conduction to explain the transfer of heat from the center of a stainless steel fermenter (shape: cylinder with height 3 radii) to the wall it will take 6.5 hours for half the beer (the beer between the wall radius and 70% of the wall radius) to come to within 1 °C of the coolant and 10 hours for all the beer (from the center of the fermenter to the outer wall) to come to within 1°. If the fermenter is glass it will take 10 hrs and 13 hrs respectively.

If we assume that convection improves the thermal coupling between the concentric cylinders of wort in this fermenter by a factor of 10 the times drop to 0.7 hr and 1 hr for steel and 4.75 hr and 5.0 hr for glass. I have no idea whether convection lowers thermal impedance by as much as or much more than (I suspect the latter) a factor of 10.

This ignores heat production by the yeast. As the 19L stainless steel fermenter model has a thermal conductance of about 1000 watts/°C it is clear that with stainless this is a valid approximation (19 L of fermenting wort doesn't produce anything like a kW and so the rise across the fermenter wall from this is clearly way less than a degree. The thermal impedance of the glass fermenter, otoh, is 160 times greater than this and, thus, a heat production of only 6 watts or so would result in a 1 ° rise in passing through the fermenter wall.

Lots of assumptions went into these numbers and the only way to really find the truth is to do experiments. The model, however, is a simple RC one in which I divide the fermenter's wort up into 100 concentric shells about 1 mm thick, calculate the thermal masses in those shells and the thermal impedance between them. I then start each shell off at 30 °C and calculate the heat transferred between shells (and consequent temperature drops) at time intervals equal to 5% of the time constant of the shell with the smallest time constant.

I've omitted lots of stuff here because I don't think there is interest and I only got this model running recently and haven't really fully shaken it out. Nor have I compared it to real world measurements. This is the validation I referred to in an earlier post.
 
ajdelange said:
Yes, absolutely and it happens pretty quickly if the container is metal and more slowly if the container is glass.

The model really isn't that complicated - until you try to account for convection. For 20L of wort at 30 °C placed in a bucket of 20 °C water with only conduction to explain the transfer of heat from the center of a stainless steel fermenter (shape: cylinder with height 3 radii) to the wall it will take 6.5 hours for half the beer (the beer between the wall radius and 70% of the wall radius) to come to within 1 °C of the coolant and 10 hours for all the beer (from the center of the fermenter to the outer wall) to come to within 1°. If the fermenter is glass it will take 10 hrs and 13 hrs respectively.

If we assume that convection improves the thermal coupling between the concentric cylinders of wort in this fermenter by a factor of 10 the times drop to 0.7 hr and 1 hr for steel and 4.75 hr and 5.0 hr for glass. I have no idea whether convection lowers thermal impedance by as much as or much more than (I suspect the latter) a factor of 10.

This ignores heat production by the yeast. As the 19L stainless steel fermenter model has a thermal conductance of about 1000 watts/°C it is clear that with stainless this is a valid approximation (19 L of fermenting wort doesn't produce anything like a kW and so the rise across the fermenter wall from this is clearly way less than a degree. The thermal impedance of the glass fermenter, otoh, is 160 times greater than this and, thus, a heat production of only 6 watts or so would result in a 1 ° rise in passing through the fermenter wall.

Lots of assumptions went into these numbers and the only way to really find the truth is to do experiments. The model, however, is a simple RC one in which I divide the fermenter's wort up into 100 concentric shells about 1 mm thick, calculate the thermal masses in those shells and the thermal impedance between them. I then start each shell off at 30 °C and calculate the heat transferred between shells (and consequent temperature drops) at time intervals equal to 5% of the time constant of the shell with the smallest time constant.

I've omitted lots of stuff here because I don't think there is interest and I only got this model running recently and haven't really fully shaken it out. Nor have I compared it to real world measurements. This is the validation I referred to in an earlier post.
Click to expand...

I'm curious how far you got with heat transfer. Are you in it now? Grad student, professor or heat transfer pro?
 
I propose calling this the “AJ effect”. It ‘s where the liquid gets so mixed up it forgets what it’s doing (it’s thermal properties). I propose this to honor the man who repeatedly beat me over the head with it.

From rough measurements I took in an ongoing fermentation, it seems this effect starts from equilibrium, and then builds into max krausen, peaks, and fades into equilibrium. The beer is pretty well mixed in about 18hours. The cooling water is still showing some stratification until max krausen at about thirty hours.

So if you want to apply your model to an actual fermentation you need to reflect the times involved. The AJ effect kicks in gradually and then subsides. I suppose you could model the AJ factor with a couple of differential equations (the beer and the water) and then integrate the whole thing over time.
 
petep1980 said:
I'm curious how far you got with heat transfer. Are you in it now? Grad student, professor or heat transfer pro?
Click to expand...
Not very far. This is quite simple stuff where the discretized resistor/capacitor network is accurate largely because of the symmetry (coaxial cylinders) of the problem. I probably first encountered this in the GE SCR manual many years ago (i.e. when SCRs first hit the market). Never studied it in school, never applied it beyond the level of the current discussion. I was an electrical engineer and we often had to keep things at a regulated temperature.

You are thinking of the really hairy stuff where an engineer needs to know, for example, the distribution of temperature throughout a turbine blade in a jet engine in order to calculate thermal stresses. The basic concepts may apply but the solution of such problems is orders of magnitude more difficult than the solution of this one.

Unless you try to model the convection in the fermenter. That brings fluid mechanics into it and I wouldn't even know where to begin.
 

Similar threads

D
Fermentation Temperature Control
Replies
21
Views
2K
Arbe0
Arbe0
Closet Fermenter
Don’t do this, (with a chiller)
Replies
1
Views
664
JohnSand
JohnSand
Langerz
First Fermentation with a Tilt - Questions
Replies
20
Views
2K
AZ Maverick
AZ Maverick
Uniique
My mead has achieved a SG of 0.992 in 14 days, and i need some help with it.
Replies
3
Views
676
Uniique
Uniique
BeerNsteadofH2O
New to the show
Replies
1
Views
306
BeerNsteadofH2O
BeerNsteadofH2O
BrewSwag 300

Latest posts

Back
Top