PDA

View Full Version : Thermal Dynamics Heat Transfer Model



pyro
03-15-2017, 10:42 PM
I boil on a 14 gauge SS pan, and have had an interest on how much improvement I would see if I went to a ~20 gauge pan. A new pan is fairly expensive so I wanted to understand if I would see a significant improvement. Therefore I started researching some thermal dynamics basics and would like to share what I learned as I haven't seen it presented before. This thread is for the engineering nerds out there, but I hope everyone can learn something as well.

Our goal is to get energy transferred from the fire to sap which will convert the liquid to steam. To do this we need a lot of energy. But its not just energy alone we are concerned with. We don't want to boil all day and night, so we want to do this as fast as possible which is why we need to talk about the rate of this heat transfer called the "heat transfer rate" or H. Heat transfer is (Eq 2) H = T/R.

We can model the heat transfer rate of two different temperatures across a medium by equating it to a basic electrical circuit with two voltages across a resistance. One temperature is the fire temperature say 1000 F. The other side is boiling sap at a constant of 212 F (I know what your thinking, but its not significant). To transfer energy from one side to another it must go across the pan which has some unknown thermal resistance. Again equate heat transfer to current in an electric circuit. In design phase, our goal is to reduce the resistance to increase the heat flow (or current in electrical analogy).

16082

Thermal resistance is defined as (Eq 3) R = L/kA. The thermal resistance is made up of three variables. To reduce the thermal resistance we have only three options:
1. Increase area A
2. Increase thermal conductivity k (by switching materials)
3. Decrease pan thickness L

Substituting Equation (3) into Equation (2) we get Equation (4) which shows the total energy transfer of the system. It proves that decreasing the pan thickness L by 50% is equivalent to doubling the surface area A! It is this that is why the pan thickness is significantly important.

It also shows that if we found a different material such as copper it could increase k and significantly improve energy transfer. For example, copper has a thermal conductivity (=1/R) of about 15-30x more than 304 stainless steel. That means the thermal resistance is 15-30x less and would improve thermal energy transfer 15-30x!

Well not so fast. why are the few people with copper pans not raving about their 15x improvement? Why are the manufacturers not using them? I suspect there is something I am missing but I can't find what it is. We can find the losses in the system, but I still don't think it could explain where the 15-30x improvement is lost. Maybe someone has ideas?

Let's investigate losses anyways since it will help settle the sap boil depth question. The most significant loss is due to sap depth. The sides lose heat to ambient air. We can model this using the same thermal resistance. However the temperature delta is not as large as the bottom of the pan since the outside is not 1000 but ambient outside air. As an example, I have compared 1" to 3" boiling depth at 1000 degree F fire temperature, and 73 degree F ambient air on the side of the pan. I made up 73 degrees F (room temp) since the evaporator heats up the surrounding area and I have never measured it directly. These value are percentage of energy lost to the sides. Less is better.

For a 2x3 flat pan:
1": 5%
3": 15%

For a 2x4 flat pan:
1": 4%
3": 13%

For a 2x6 flat pan:
1": 4%
3": 12%

So based on all that math that I believe, I still sometimes convince myself I see a better boil with 3" deep, rather than 1". But that likely is more due to the correlation between the fact that I never have 1" of sap in when the fire is ripping.

I appreciate any comments, especially if you can prove or disprove using thermal equations.

RileySugarbush
03-15-2017, 11:04 PM
Here are some things to consider:

The heat transfer is not limited only by the conduction through the pan material. You also have to consider conduction through any soot layer Also, the convection from flue gas to the bottom surface and from the pan interior to the sap. Both of those are functions of fluid properties, velocities and turbulence . One more is the radiative heat transfer from fire/coals to pan bottom.
Some of these act in series and some in parallel. There is a lot going on.

ecolbeck
03-17-2017, 06:29 AM
If you believe there is error in the calculated efficiency gain from changing to a copper pan might there also be significant error in the change from thicker to thinner stainless?

bigschuss
03-17-2017, 08:55 AM
I boil on a 14 gauge SS pan, and have had an interest on how much improvement I would see if I went to a ~20 gauge pan. A new pan is fairly expensive so I wanted to understand if I would see a significant improvement. Therefore I started researching some thermal dynamics basics and would like to share what I learned as I haven't seen it presented before. This thread is for the engineering nerds out there, but I hope everyone can learn something as well.

Our goal is to get energy transferred from the fire to sap which will convert the liquid to steam. To do this we need a lot of energy. But its not just energy alone we are concerned with. We don't want to boil all day and night, so we want to do this as fast as possible which is why we need to talk about the rate of this heat transfer called the "heat transfer rate" or H. Heat transfer is (Eq 2) H = T/R.

We can model the heat transfer rate of two different temperatures across a medium by equating it to a basic electrical circuit with two voltages across a resistance. One temperature is the fire temperature say 1000 F. The other side is boiling sap at a constant of 212 F (I know what your thinking, but its not significant). To transfer energy from one side to another it must go across the pan which has some unknown thermal resistance. Again equate heat transfer to current in an electric circuit. In design phase, our goal is to reduce the resistance to increase the heat flow (or current in electrical analogy).

16082

Thermal resistance is defined as (Eq 3) R = L/kA. The thermal resistance is made up of three variables. To reduce the thermal resistance we have only three options:
1. Increase area A
2. Increase thermal conductivity k (by switching materials)
3. Decrease pan thickness L

Substituting Equation (3) into Equation (2) we get Equation (4) which shows the total energy transfer of the system. It proves that decreasing the pan thickness L by 50% is equivalent to doubling the surface area A! It is this that is why the pan thickness is significantly important.

It also shows that if we found a different material such as copper it could increase k and significantly improve energy transfer. For example, copper has a thermal conductivity (=1/R) of about 15-30x more than 304 stainless steel. That means the thermal resistance is 15-30x less and would improve thermal energy transfer 15-30x!

Well not so fast. why are the few people with copper pans not raving about their 15x improvement? Why are the manufacturers not using them? I suspect there is something I am missing but I can't find what it is. We can find the losses in the system, but I still don't think it could explain where the 15-30x improvement is lost. Maybe someone has ideas?

Let's investigate losses anyways since it will help settle the sap boil depth question. The most significant loss is due to sap depth. The sides lose heat to ambient air. We can model this using the same thermal resistance. However the temperature delta is not as large as the bottom of the pan since the outside is not 1000 but ambient outside air. As an example, I have compared 1" to 3" boiling depth at 1000 degree F fire temperature, and 73 degree F ambient air on the side of the pan. I made up 73 degrees F (room temp) since the evaporator heats up the surrounding area and I have never measured it directly. These value are percentage of energy lost to the sides. Less is better.

For a 2x3 flat pan:
1": 5%
3": 15%

For a 2x4 flat pan:
1": 4%
3": 13%

For a 2x6 flat pan:
1": 4%
3": 12%

So based on all that math that I believe, I still sometimes convince myself I see a better boil with 3" deep, rather than 1". But that likely is more due to the correlation between the fact that I never have 1" of sap in when the fire is ripping.

I appreciate any comments, especially if you can prove or disprove using thermal equations.

My only comment is that my biggest concern about equations when it comes to boiling is whether I should buy 2 or 3 six packs for my buddies that come over. Three buddies + myself x 3 beers each = two six packs.

pyro
03-17-2017, 11:20 PM
bigschuss: Your equations are more critical to boiling efficiency, so spend the effort to make sure those are correct!

ecolbeck: Yes, any errors would be the same for both material type and pan thickness.

RileySugarbush: It took me a while, but I agree. My error is I am assuming an ideal source for temperature. The fire is not ideal and is energy limited. I can prove this by dumping in a lot of cold sap and seeing the flue temperature drop. In fact I can use this to backwards calculate an equivalent model of the fire.

I will update the model to better understand what it does for pan thickness vs area. I'll post the results if anyone is interested.

bigschuss
03-18-2017, 07:55 AM
bigschuss: Your equations are more critical to boiling efficiency, so spend the effort to make sure those are correct!



Funny! I wasn't trying to be snarky. Hope you didn't take it that way. I'm a science educator and I do love stats and data and a good problem! For me though sugarin' is a hobby and all I want to do is throw some wood in the firebox, drink a few beers, and boil some sap with family and friends.

Vtmbz
04-08-2017, 08:19 AM
Would there be a measurable improvement in heat transfer if a stainless flat bottom pan had a sheet of copper soldered to the underside? if so, would a 1/4" copper sheet be better or worse than a 1/8" sheet?

BCPP
04-08-2017, 08:49 AM
Would there be a measurable improvement in heat transfer if a stainless flat bottom pan had a sheet of copper soldered to the underside? if so, would a 1/4" copper sheet be better or worse than a 1/8" sheet?
No because you increase the total thermal resistance. You will often find high end cooking pots have a layer of copper sandwiched between SS on the bottom. The reason is because of coppers superior heat conductivity it spreads the heat more evenly so the pot does not have hot spots on the bottom where food burning might occur.

For evaporating pans three thoughts on why SS is used over copper. 1)copper is much softer so copper flat pans would likely need to be thicker to support the weight of sap without buckling or warping. 2)I know of more than one farmer from my youth in Quebec who had an evaporating pan run dry resulting in solder joints on their copper pans melting. Once that happens pan is scrap as it is virtually impossible to repair. SS tig welding does not suffer this fate! 3)SS is easier to clean / more wear resistant than copper.
Rileysigarbush, while those all seem reasonable things to consider do they not cancel out when doing the comparison of Ss thickness as pan thickness would not impact them?
Doug

RileySugarbush
04-08-2017, 10:55 AM
While those other sources of thermal resistance do not change for various materials, they are still there. Since they can be significant, they can mask the improvement of thinner and more conductive pan material.

An analogy: Your car's gas milage at 60 MPH is affected by many things as well. Tire rolling resistance, aerodynamic drag, engine efficiency and driveline friction. Take any one of them and make it perfect, with zero losses, and your milage does not go to infinity. It is still held back by the others.

Same way with your pan. Cut the conductive loss in half and you can get a noticeable improvement, but not proportional to the improvement in conduction.

BCPP
04-08-2017, 04:04 PM
Agreed John, I was thinking of only the relative comparison of the pan thickness. You are correct, all those other factors could have a much larger impact that could make the absolute effect much less, possibly even not noticeable!

One other thought, as you decrease the SS thickness at some point there will be structural issues where the pan will be very susceptible to warping and unable to maintain a level pan. Not sure how you factor that in!
Doug