Keep the Heat B/Thermodynamics C

[quizbowl]

Gentleman, you can't have a heated conversation in here! This is the thermodynamics thread!

[chalker7]

I want to chime in here in defense of the students that have posted. Each concern regarding the initial temperature of the water posted is valid and many of the replies include the word "suspect". This event has never been run anywhere, and is just now being tested in schools throughout the United States. As Regional Director, I have already heard from two other schools with the same concern about the initial temperature. The initial temperature needs to be taken as the device is closed. My students have thoroughly tested this 'transfer' problem and the average time of transfer is 30 seconds (can't safely be done faster than that...we do not want high schoolers pouring nearly-boiling water twice in less time than that) and the average temperature difference is about 12˚C, but is larger with higher initial bath temperatures. As science teachers/scientists, we work to eliminate variables and should only be testing one at a time. The goal of the event appears to be two-fold: 1. Build a device out of natural materials, including "granular organics" (this needs defining, by the way) to retain as much heat as possible, and 2. Be able to accurately predict the final temperature, which is IMPOSSIBLE without accurately knowing the initial temperature. There are at least 8 variables that have some effect on the initial temperature, all of which would be eliminated by taking the initial temperature once the water is in the device. As for difficulty for event supervisors to measure these initial temperatures individually, with staggered start (Rule 4.a.iv), there will be no problem for event supervisors, even if they have only one thermometer (most of which will have more than one).

You stated at least 3 times in your replies that you "suspect" an outcome. The difference between plastic and glass appears to be minimal, but the other two, it is significant. The problem with "suspecting" something is that it has not been tested. It has now been tested and there is a problem. The rules have been written to allow "luck" and "guessing" to be a factor...not good science. Please change the rule to state the initial temperature will be taken once the water is IN the device. [By the way, the state of Kansas just announced this morning that is has already decided that the initial temperature will be taken as the lid closes.]

The problem is that everyone is focusing on just one aspect of this event. While I agree that it would make prediction easier if the "initial" loading temperature were known, it is nearly impossible to control that initial temperature for all the teams. It's a fairness issue, if every team does not have a single baseline to start with, we will receive numerous, justified arbitrations. As a Regional Director, I know you want to avoid those as much as possible. I would strongly suggest NOT running the event as you described to specifically prevent the complaints from the teams and coaches. If I were a coach for a team that competed at a tournament that provided varying initial temperatures for different teams, I would assuredly file an arbitration. The way the rule is currently written, it is fair for all teams and rewards those who practice.

We DID test this event extensively (I personally built a device back in April), but cannot predict every eventuality. We did discuss the issue of cooling, but also knew that it was reproducible and if a team prepared by practicing their procedure and taking the initial cooling period into account, then it would not be a completely unknown variable as you suggest.

I'm not entirely sure attacking people who "suspect" something is productive to this conversation.

[Frogger4907]

The problem is that everyone is focusing on just one aspect of this event. While I agree that it would make prediction easier if the "initial" loading temperature were known, it is nearly impossible to control that initial temperature for all the teams.

No, it does not make the prediction easier, it would make it POSSIBLE. And the initial temp is going to be different for all teams Anyway!

I'm not entirely sure attacking people who "suspect" something is productive to this conversation

If you were to suspect everything then you wouldn't have any real or valid information. So therefore everytime you "suspect" something we have to assume that you have no idea as to what you're talking about, and that you have not collected and valid or relovent data, which clearly seems to be the case.

all I have left to say is:
Without the initial temperature being used for the prediction, this event is absolutly terrible. A simple mistake has turned this event into a complete mess, and it will remain that way until a certain group of people stop worrying about their pride, admit the error and fix the problem. The goal of an event shouldn't make it a guessing game because of all of the variables, it is absolutly out of the question for any of the processes to move the water to the device to be safe and reproducable.
Please just grow up and admit the error.
Edit: AND FIX IT!

[JSGandora]

There is a certain degree of respect one must show towards people in authority. Nothing will get done if you disrespect them like that - know your place. Reasonable arguments can be presented without insults being thrown around. If you're going to argue, then do it respectfully. Any thought of changing the rules chalker might have would be eliminated because of how you're presenting your argument. You can listen to me or not, I'm just putting it out there.

[Schrodingerscat]

According to my understanding, the idea is not to change or remove any other part of the safeguards to ensure fairness in the temperature of the water that delivered to the teams, such as a constant temperature water source. If I am correct, the idea is to provide the teams with the data of the water temperature after the beaker has been inserted into the device and it is fully closed. If this number varies between teams, as it should based upon the rate of water cooling at these temperatures, it will also vary with or without the additional procedure of measuring the water temperature for the competitors at a particular step. This would allow teams to ignore the numerous variables and to focus on how there device insulates with only the three variables of room temperature, the volume of the water, and the temperature of the water once loaded into the device.

[Frogger4907]

There is a certain degree of respect one must show towards people in authority. Nothing will get done if you disrespect them like that - know your place. Reasonable arguments can be presented without insults being thrown around. If you're going to argue, then do it respectfully. Any thought of changing the rules chalker might have would be eliminated because of how you're presenting your argument. You can listen to me or not, I'm just putting it out there.

I can't give anyone respect anymore when i get an answer involving all the variables magically canceling out

If it was one variable I would try to account for it. But it is many more than that. If they measured initial temp than it would be a workable situation.

How do you know it isn't a workable situation? Have you run multiple repeat tests? For all you know the various other variables you are concerned about tend to cancel each other out, allowing you to rather accurately predict the final temperature based upon initial conditions.

Regardless, this is a slippery slope you are going down. Let's say we do measure the initial temp once the beakers are in the devices. I can imagine someone then saying we need to take into account variables such as whether the device is close to a window (with the sun shining in), or directly under a HVAC vent (increasing the local air flow), or sitting on a wooden table versus a lab bench, or for exactly how long the supervisor left the temperature probe in the beaker, or what the temperature of the probe was before it was inserted, etc. etc. etc.

The bottom line is there is always a struggle between theory and practice. We have to create an event that is practical to run, which means there are tradeoffs to be made regarding the theory that is being tested.

If you can't even slightly admit there is an error on mistake with something you are involved in creating. I have no respect for you, no matter what your authority. Authority must be challenged by the "governed" to eliminate "corruption". I tried being respectful, but I've gotten nothing but answers that DO NOT solve THE PROBLEM.

[hmcginny]

Frogger I see why you would think this is an error in the rules, but it really isn't. Every team is provided with water at the same temperature and if you account for how long it takes to get the water into the insulation device, you can test and thus predict your temperature fairly accurately. The variables don't magically cancel out, they're just an unavoidable source of error in this lab that you can account for by testing. The movement of water from the bath to your device is just part of the event, yes it may be slightly more random than the rest of it, but it is still testable. If you were going from initial temperatures, everyone would start at a different point which is a. very complicated for the supervisors and b. unfair for the competitors. Instead since everyone starts at the same temperature, you can determine the initial temperature when you close your insulation if you test for it. Also show chalker some respect, he doesn't have to come onto the forums and debate about the rules and insulting him isn't the best way to get him to listen to your argument.

[chalker]

This event has never been run anywhere, and is just now being tested in schools throughout the United States.

This is untrue. If you look at the trial events page on soinc.org (http://soinc.org/trial_events), you'll note there are not 1, not 2, but THREE versions of this event listed (Keep the Heat, Keep the Heat 94, Keep the Cool). This event has been run in a variety of forms for more than 15 years all across the country. Even in recent years I believe Minnesota ran it as a trial event. The published rules do NOT differ from any of the trial rules with regards to the initial temperature issue we're discussing. While we were working on updating the rules we solicited feedback from a variety of sources (including local supervisors AND coaches) and this issue was never mentioned.

My students have thoroughly tested this 'transfer' problem and the average time of transfer is 30 seconds (can't safely be done faster than that...we do not want high schoolers pouring nearly-boiling water twice in less time than that) and the average temperature difference is about 12˚C, but is larger with higher initial bath temperatures.

Great! Can you share the reproducibility of the results? What I've been asking for all along is for someone to test not whether or not there is a significant temperature drop (we know that), just whether it's reproducible / calculable if you take into account easily obtained major factors (e.g. room temperature, time of transfer)

As science teachers/scientists, we work to eliminate variables and should only be testing one at a time. The goal of the event appears to be two-fold: 1. Build a device out of natural materials, including "granular organics" (this needs defining, by the way) to retain as much heat as possible, and 2. Be able to accurately predict the final temperature, which is IMPOSSIBLE without accurately knowing the initial temperature. There are at least 8 variables that have some effect on the initial temperature, all of which would be eliminated by taking the initial temperature once the water is in the device.

My perspective is obviously different. The goal of the event is 1. to build and understand the performance of a device that doesn't rely on 'obvious' insulating materials (such as commercial insulation, foam, etc etc) and 2. Be able to respond to a variety of questions related to the field of thermodynamics. Note the prediction score is only worth about 17-18% of the total possible event score.

People keep using the word IMPOSSIBLE with regards to predicting the final temp, but I've yet to see any evidence anyone has actually put serious effort into. Everything I've heard so far sound to me like "it appears this is too hard a problem to excel at, please make it easier for us so we can get a really good prediction score and thus focus our energies on other aspects of the event" You say you've identified 8 variables.. Great! Now you can conduct testing to hold each of them constant and see what impact each has on the final result!

As for difficulty for event supervisors to measure these initial temperatures individually, with staggered start (Rule 4.a.iv), there will be no problem for event supervisors, even if they have only one thermometer (most of which will have more than one).

My brother does a good job in another note of summarizing some of the 'problems' event supervisors might face measuring individual initial temperature. It's disingenuous to claim there is no additional burden or difficulty for them to implement this.

You stated at least 3 times in your replies that you "suspect" an outcome. The difference between plastic and glass appears to be minimal, but the other two, it is significant. The problem with "suspecting" something is that it has not been tested. It has now been tested and there is a problem. The rules have been written to allow "luck" and "guessing" to be a factor...not good science.

All events (and real world science and engineering) have some element of 'luck' or 'guessing', particularly with building events. Just check out some of the other threads here. For example, some people are concerned about how 'level' the tower test structures are. Regardless, we try to minimize those types of factors, and I haven't seen any evidence presented yet that they are a major factor in this event.

Please change the rule to state the initial temperature will be taken once the water is IN the device. [By the way, the state of Kansas just announced this morning that is has already decided that the initial temperature will be taken as the lid closes.]

One of the great things about SO is that it's a very decentralized organization. Each State handles things a bit differently. Kansas is welcome to make that change. It'll be interesting to see what impact it has on both the competition and the resulting scores. From a National standpoint though, we have the following philosophy: We know the rules aren't perfect, but they are published at this point. We hesitate to make major changes / clarifications to them unless we are convinced that is some aspect we overlooked or erred one that provides an unfair advantage to some competitors or completely makes the event unworkable. I have yet to see any sign of that here.

Regardless, my previous offer still stands: provide a draft of the changes you'd like (that take into account the various factors I've mentioned) and I'll be happy to discuss and consider them, if not for this year's rules, definitely for next year's rules.

[chalker]

I'm going to multiquote a lot in this post, to avoid cluttering up this thread with a lot of individual replies.

No, it does not make the prediction easier, it would make it POSSIBLE. And the initial temp is going to be different for all teams Anyway!

Again, show me some evidence that it's impossible (versus just difficult) to predict right now. I haven't seen any presented.

If you were to suspect everything then you wouldn't have any real or valid information. So therefore everytime you "suspect" something we have to assume that you have no idea as to what you're talking about, and that you have not collected and valid or relovent data, which clearly seems to be the case.

There's a big difference between having an educated 'suspicion' and making a 'wild ass guess'. As indicated in other posts, we did conduct some experiments, have a lengthy history of this event being run this way, and have had numerous people with advanced engineering degrees in a variety of disciplines be involved in the rules making process. We aren't some random joes off the street pontificating about this....

Without the initial temperature being used for the prediction, this event is absolutly terrible. A simple mistake has turned this event into a complete mess, and it will remain that way until a certain group of people stop worrying about their pride, admit the error and fix the problem. The goal of an event shouldn't make it a guessing game because of all of the variables

A factor that potentially impacts AT MOST 18% of the max score makes this event a complete mess? This has nothing to do with 'pride', or 'errors'. In fact I've clearly stated in this thread and elsewhere that we aren't perfect - no large human endeavor is. And I'm the first to admit to true errors when they are pointed out (for example look earlier in this thread about the confusing wording of adding ice water to the beakers.) What we have here in not an 'error', but rather a potentially different approach to the running of the event.

There is a certain degree of respect one must show towards people in authority. Nothing will get done if you disrespect them like that - know your place. Reasonable arguments can be presented without insults being thrown around. If you're going to argue, then do it respectfully. Any thought of changing the rules chalker might have would be eliminated because of how you're presenting your argument. You can listen to me or not, I'm just putting it out there.

Well said, however please don't think I sit in an ivory tower 'looking down' upon the general SO competitors. I (and my brother) have thick skin and understand people can be passionate about things, as we are (and have it come off a bit harsh via online postings). I respect, value and encourage any and all feedback (and have made that very clear many times here). But it does make it easier for all of us if we maintain a conversational instead of confrontational tone. What's more important to me is that I'm not seeing responses to my specific questions or offers. Just a generic demand based upon some initial perceptions. Technically I don't have the power to make any official changes after the rules have been published. What I do have power to do is strongly recommend a change be made, but I'm going to receive even greater questioning and pushback than you see me giving here, so I need to be prepared for that if I'm going to do it.

According to my understanding, the idea is not to change or remove any other part of the safeguards to ensure fairness in the temperature of the water that delivered to the teams, such as a constant temperature water source. If I am correct, the idea is to provide the teams with the data of the water temperature after the beaker has been inserted into the device and it is fully closed.

That is my understanding as well as to what we are discussing.

I can't give anyone respect anymore when i get an answer involving all the variables magically canceling out

You are either deliberately twisting my words or not paying close enough attention to my posts. Whichever it is, if you want to have a serious conversation about this issue I recommend not doing that. I've repeated asked for more data points, and was simply proposing that some of these variables might be minute enough or have net combined effects that they don't impact the final result significantly enough.

If you can't even slightly admit there is an error on mistake with something you are involved in creating. I have no respect for you, no matter what your authority. Authority must be challenged by the "governed" to eliminate "corruption". I tried being respectful, but I've gotten nothing but answers that DO NOT solve THE PROBLEM.

No, you have gotten lots of questions / requests you've ignored. Again, scan through the history of my postings on here. I readily admit mistakes or errors. This is not such a situation. You propose that there is a problem that is so significant it makes the event completely unworkable. My current opinion is that is not the case, but I've asked for additional information in an attempt to see if I'm missing something.

Frogger I see why you would think this is an error in the rules, but it really isn't. Every team is provided with water at the same temperature and if you account for how long it takes to get the water into the insulation device, you can test and thus predict your temperature fairly accurately. The variables don't magically cancel out, they're just an unavoidable source of error in this lab that you can account for by testing. The movement of water from the bath to your device is just part of the event, yes it may be slightly more random than the rest of it, but it is still testable. If you were going from initial temperatures, everyone would start at a different point which is a. very complicated for the supervisors and b. unfair for the competitors. Instead since everyone starts at the same temperature, you can determine the initial temperature when you close your insulation if you test for it.

Thanks for conveying what I've been trying to convey in much better terms.

Also show chalker some respect, he doesn't have to come onto the forums and debate about the rules and insulting him isn't the best way to get him to listen to your argument.

Insults aside (thick skin remember;), I'm listening, but also trying to speak. If you want a change made, I can't make it, but I'm the best advocate to 'the powers that be' that could make the change. I recommend trying to win me over to your point of view with relevant and pertinent responses to my questions, instead of trying to browbeat me into agreement. I can assure you I'm doing the same in reverse right now;)

[Schrodingerscat]

Although I do not at this moment have the means to conduct some tests, I do have a few cooling curves from some initial prototypes to collect some initial data. I do not have any data to use to support this, but I will make a prediction based upon the initial water temperature, volume, and room temperature, a competitor is able to predict the temperature of the water once inserted into the device fully to plus or minus one degree. At a time of 39.0 seconds the water had a temperature of 71.4°C. At a time of 111.0 seconds, the water temperature inside the device had cooled to 69.4°C. This gives a range of potential temperatures of 2.0°C in a time variation of 72.0 seconds. I will also make the assumption due to my current unavailability of actually using different starting temperatures in this device that I can predict by analyzing temperatures with this difference in time. At a time of 1200.0 seconds (20 minutes), the water had a temperature of 48.7°C. At a time of 1272.0 seconds, the water had a temperature of 47.8°C. In the same uncertainty in cooling time created by the hypothetical difference in the uncertainty in the inserted water temperature, there is a difference of 0.9°C. Assuming a median value this will be simplified to 0.45°C error in prediction. According to the rules, this would result in a 0.2 point deduction. However, this result is dependent upon my predicted transfer accuracy of plus or minus 1°C. I can neither say whether or not this is indicative of what could be attained by the numerous potential variables in the water transfer nor if my method of predicting will the affect on the prediction score from initial temperature variation due to uncertain water transfer variables.

I will also as a second test hypothesize that I can only predict with an uncertainty of plus or minus 3°C. At 39.0 seconds, the water had a temperature of 71.4°C (same data point). At 234.0 seconds, the temperature was 65.4°C. The change in temperature was 6.0°C in 195.0 seconds. At the time of 1200.0 seconds, the water had a temperature of 48.7°C (same data point). At the time of 1394.0 seconds, the water had a temperature of 46.4°C. This resulted in a temperature change of 2.3°C in the 195.0 seconds. I will again assume a middle value and assume the prediction error is up to 1.1°C. According to the rules, this would result in a 0.6 point deduction.

Finally, I will predict what would happen if the competitor did not think about the cooling during water transfer and assumed the water reached the device at the water bath temperature of 70 degrees, but had actually cooled to 58 degrees. (A reasonable cooling according to the tests that we did conduct earlier.) At a time of 93.0 seconds the water had a temperature of 70.0°C. At a time of 569.0, the water temperature was 58.0°C. This is a change of 12.0°C in 473 seconds. At the time of 1200.0 seconds, the temperature was 48.7°C (same data point for consistency). At the time of 1672.0 seconds, the temperature was 43.7°C (last data point recorded by probe). The prediction error was 5.0°C, which would be penalized 2 points.

These theoretical predictions appear to show the transfer variables will barely be significant at all, but will reward teams who attempt to account for them. However, I am not even certain about my method of predicting these from a single graph, and more tests specifically controlling the temperature of the water once in the beaker may need to be performed.

Major Edit 1: Fixed calculation error for third predicted score causing large overestimation of point deduction and adjusted conclusion accordingly