DominionSeraph
Diamond Member
- Jul 22, 2009
- 8,386
- 32
- 91
In the Department of Engineering Science at University of Oxford where I was working for the month of January. If anyone should know how to brew tea in an intelligent, safe manner, I would think they would be the ones, but perhaps you know better. They had both free-standing kettles and wall mounted designs, both of which were closed systems with thermal controls similar to those I mentioned.
TuxDave
Lifer
- Oct 8, 2002
- 10,571
- 3
- 71
Born2bwire
Diamond Member
- Oct 28, 2005
- 9,840
- 6
- 71
Bah, forum nuked my original response so here's take two.
All the air pots and kettles that I have seen that have temperature controls on them will always boil the water first and then let it cool down to temperature and then use the heating element to maintain the temperature as needed. This is particularly necesary for areas where it is recommended or required that you boil water prior to consumption. My only complaint is that my air pot loses heat so slowly that I have to open it up for the water to drop down to temperature in a reasonable amount of time.
Sure, the one that I have features both:
http://www.panasonic.hk/ha/english/feature_585.aspx
http://www.sonicelectronix.com/item_21235_Panasonic+NC-HU401P.html
VIP is their catch-phrase for vacuum insulation. There are other air pots that I have seen in the store that are vacuum insulated too. I really like this one, just wish it could take in both 220 and 110 V.
All the air pots and kettles that I have seen that have temperature controls on them will always boil the water first and then let it cool down to temperature and then use the heating element to maintain the temperature as needed. This is particularly necesary for areas where it is recommended or required that you boil water prior to consumption. My only complaint is that my air pot loses heat so slowly that I have to open it up for the water to drop down to temperature in a reasonable amount of time.
Sure, the one that I have features both:
http://www.panasonic.hk/ha/english/feature_585.aspx
http://www.sonicelectronix.com/item_21235_Panasonic+NC-HU401P.html
VIP is their catch-phrase for vacuum insulation. There are other air pots that I have seen in the store that are vacuum insulated too. I really like this one, just wish it could take in both 220 and 110 V.
William Gaatjes
Lifer
- May 11, 2008
- 22,233
- 1,419
- 126
That makes sense.
Don't yoke about it, but when i always see ( in my inner mind) something that has to be heated and cool down, as a electrical capacitor but instead of holding electrical charge it stores heat energy. The thermal resistance equals an external series resistance. The internal series resistance of the capacitor is analog to the heat transfer capability of the material. The inner leakage of the dielectric is analog as how fast the material radiates it's heat off. An external parallel resistor connected to the capacitor equals any heat transfer through conduction or convection.
By Born2bwire
I can imagine that water must boil first , E.coli strains that are not harmless or legionella could be in the water in too large quantities.
This is basically a control problem that falls under non-linear control theory, which involves extensive modelling, analysis, and math.
The easiest way would be to implement a fuzzy logic controller. Something like:
if temp < setpoint and difference is 50 deg. or more
add tons of heat
else if temp < setpoint and difference is >= 15 degrees
add medium heat
else if temp < setpoint and difference is < 15 degrees
add low heat
else if temp > setpoint
turn off heat
The system would bounce around the set point of the system very slightly, but due to the thermal mass of the heating element, kettle, and water, the oscillation will be minimal if the logic above is optimized.
#2 first. An electric heater immersed in water is ALWAYS 100% Efficient. That is, whatever electrical energy is actually delivered to the element will ALL go into the water. Now, you do have to ensure that the wiring and connections in the system don't waste energy outside the element, but that's not hard. By far the largest impact on "Efficiency" in this system is NOT Efficiency of the heater; it is heat loss from the hot body (kettle), which depends on how much insulation surrounds it. Regarding calculating and predicting from water mass, etc just how much electrical energy will be required, there are just SO MANY other factors and places where heat can flow that you CANNOT make an accurate prediction. This is a Feed-Forward type of plan. By far the better alternative is a simple classic Feedback plan - you MEASURE the result of the controller's work (water temperature) and use that to change the energy input rate (controller's output action) as the water temperature changes.
Now #1. You certainly don't need a full PID Controller, although sometimes it's easier to use one and ignore some of its features to get what you really might need - a P controller only. In those acronyms, P stands for Proportional, I is for Integral, and D is for Differential. In a PID controller the output is composed of three terms. The Proportional part takes the Deviation of [(Manually-Entered Setpoint, aka SP) - (Measured Process Variable, aka PV)], multiplies that by the negative Gain value (the Proportionality constant), then adds a Zero Offset constant. Gain and Offset normally are manually entered, too. Note that Deviation is NEGATIVE when the PV is LESS THAN the SP, which is why you have a NEGATIVE Proportinality constant (Gain). So, as the Deviation increases (that is, the NEGATIVE value gets smaller!) the output decreases. This is where system response time comes in. If the sensor's measurement of PV is slow in responding to the heater power input, the controller may be slow to reduce that power as PV (temperature) increases, resulting in overshoot of PV. The second term, Integral, constantly samples and adds up over time the Differential value above, multiplies it by its own negative constant, and adds this term to the first. In this way a system that is constantly under the setpoint (that is, all the terms adding up are negative) will get an extra boost of output to get it closer to setpoint. The third term, Differential, attempts to prevent rapid ovrshoot in either direction. It takes a sort or "derivative" of measured value versus time by sampling the Deviation and calculating the DIFFERENCE from current value to previous. It multiplies this by a negative constant and adds the result to the previous two terms; this means a rapid rise in PV produces a reduction in output in anticipation that the PV's rapid rise would continue too quickly if there were no Differential action.
For OP's original task, the I and D terms of a PID loop are not necessary. A simple Proportional controller is probably sufficient. Assuming the Gain, Offset and Damping parameters were adjusted properly, it would produce good heating to Setpoint and maintenance within a reasonable control band.
In fact, as many others already have said, even that may be overkill. A very simple On / Off controller can do a good job. Such a system certainly will produce overshoot in the initial heating phase, and there will be a "dead band" between how low the temperature must fall before it turns the heater on, and how high it must reach before it is turned off. That "dead band" will be mostly a property of the switching mechanism, because the speed of temperature change in the water is slow, and the delay time until the sensor detects the temperature correctly can be optimized by sensor placement. A very simple mechanical connection from thermal sensor to switch contacts may have a wider dead band because the sensor must exert mechanical force to move the contacts. A better system that uses minimal force to move small contacts, but then uses those light contacts to control a relay to switch the heavy heater current, could be made to yield a smaller dead band.
Now #1. You certainly don't need a full PID Controller, although sometimes it's easier to use one and ignore some of its features to get what you really might need - a P controller only. In those acronyms, P stands for Proportional, I is for Integral, and D is for Differential. In a PID controller the output is composed of three terms. The Proportional part takes the Deviation of [(Manually-Entered Setpoint, aka SP) - (Measured Process Variable, aka PV)], multiplies that by the negative Gain value (the Proportionality constant), then adds a Zero Offset constant. Gain and Offset normally are manually entered, too. Note that Deviation is NEGATIVE when the PV is LESS THAN the SP, which is why you have a NEGATIVE Proportinality constant (Gain). So, as the Deviation increases (that is, the NEGATIVE value gets smaller!) the output decreases. This is where system response time comes in. If the sensor's measurement of PV is slow in responding to the heater power input, the controller may be slow to reduce that power as PV (temperature) increases, resulting in overshoot of PV. The second term, Integral, constantly samples and adds up over time the Differential value above, multiplies it by its own negative constant, and adds this term to the first. In this way a system that is constantly under the setpoint (that is, all the terms adding up are negative) will get an extra boost of output to get it closer to setpoint. The third term, Differential, attempts to prevent rapid ovrshoot in either direction. It takes a sort or "derivative" of measured value versus time by sampling the Deviation and calculating the DIFFERENCE from current value to previous. It multiplies this by a negative constant and adds the result to the previous two terms; this means a rapid rise in PV produces a reduction in output in anticipation that the PV's rapid rise would continue too quickly if there were no Differential action.
For OP's original task, the I and D terms of a PID loop are not necessary. A simple Proportional controller is probably sufficient. Assuming the Gain, Offset and Damping parameters were adjusted properly, it would produce good heating to Setpoint and maintenance within a reasonable control band.
In fact, as many others already have said, even that may be overkill. A very simple On / Off controller can do a good job. Such a system certainly will produce overshoot in the initial heating phase, and there will be a "dead band" between how low the temperature must fall before it turns the heater on, and how high it must reach before it is turned off. That "dead band" will be mostly a property of the switching mechanism, because the speed of temperature change in the water is slow, and the delay time until the sensor detects the temperature correctly can be optimized by sensor placement. A very simple mechanical connection from thermal sensor to switch contacts may have a wider dead band because the sensor must exert mechanical force to move the contacts. A better system that uses minimal force to move small contacts, but then uses those light contacts to control a relay to switch the heavy heater current, could be made to yield a smaller dead band.
Wrong.
A steam maker will have a low "efficiency" according to your definition. The better the steam maker design, the lower the "efficiency". Please don't get me started on ideal engines.
It is amusing how flexible layman usage of words are.
Last edited:
So will someone drink boiling water if they set it to a low temperature and dispense the water too quickly?
No. You've decided that science is somehow a matter of opinion, that everyone else here is a layman, and that you already know the answer to a design question which has many answers.
Canis, yes, the term "PID Loop" is used widely to refer to a feedback control loop with these three terms in its operation - Proportional, Integral and Differential. However, that form of controller is SO common that the term colloquially has been used as if it were simply a synonym for any feedback controller. It also is a type so widely sold and supported that almost nobody makes a less-capable "just Proportional" controller any more because including the other two functions in a mass-produced controller makes more sense that producing and stocking separate products. Besides, it is very easy with any modern PID controller to simply not use the I and / or D functions and achieve a "just P" control if that's what you need.
My father was an instrumentation specialist back when controllers were analog pneumatic devices and the most common type was "just P". On them one adjusted the Setpoint and provided a pneumatic Process Variable signal (3 - 15 p.s.i. air pressure) and it produced a 3-15 psi output air pressure signal to route to the actuator device. Most had manually-adjustable settings in them for Gain and Damping, and they often were built as a recording device, too, with pens writing on strip or circular charts. The addition of Integral and Differential capabilities (not easy with pneumatic air pressure as your means of calculation over time) was real progress and got a lot of attention. As the transition to electronic controllers, sensors and actuators replaced those systems, it just made sense to replicate the full capability of the PID systems.
Regarding Efficiency, it happens that's an old discussion point dating back to my first-year university Physics labs. Any simple resistive electrical heating element IS 100% Efficient - ALL of the electrical energy supplied to it, being the product of voltage x current x time, is released into the surroundings as heat - there is NO other energy flow anywhere else. I quite understand that the overall device MAY have places where electrical energy is wasted - for example, in a poor connection in the plug of the kettle cord, leading to heat wasted there and not delivered to the water - but if you examine ONLY the energy that actually does arrive at the heater terminals, it ALL gets converted into heat in the kettle's water. Thereafter, much is lost outside the kettle, but that is an impact on the overall system operation, not on the "efficiency" of the heater itself.
Now, in the case of a "steamer", I presume you mean a device which heats water to boiling and then releases that hot steam into the air or into some further process. The fact that the hot steam that leaves the steamer vessel takes heat energy with it does not reduce the device's efficiency. That hot steam is part of the output and therefore must be included in the efficiency calculation of (energy out) / (energy in). Moreover, the "energy" in hot steam is not just in its temperature, but includes also the Heat of Vaporization it carries by virtue of having been converted from the liquid to the gaseous physical state.
My father was an instrumentation specialist back when controllers were analog pneumatic devices and the most common type was "just P". On them one adjusted the Setpoint and provided a pneumatic Process Variable signal (3 - 15 p.s.i. air pressure) and it produced a 3-15 psi output air pressure signal to route to the actuator device. Most had manually-adjustable settings in them for Gain and Damping, and they often were built as a recording device, too, with pens writing on strip or circular charts. The addition of Integral and Differential capabilities (not easy with pneumatic air pressure as your means of calculation over time) was real progress and got a lot of attention. As the transition to electronic controllers, sensors and actuators replaced those systems, it just made sense to replicate the full capability of the PID systems.
Regarding Efficiency, it happens that's an old discussion point dating back to my first-year university Physics labs. Any simple resistive electrical heating element IS 100% Efficient - ALL of the electrical energy supplied to it, being the product of voltage x current x time, is released into the surroundings as heat - there is NO other energy flow anywhere else. I quite understand that the overall device MAY have places where electrical energy is wasted - for example, in a poor connection in the plug of the kettle cord, leading to heat wasted there and not delivered to the water - but if you examine ONLY the energy that actually does arrive at the heater terminals, it ALL gets converted into heat in the kettle's water. Thereafter, much is lost outside the kettle, but that is an impact on the overall system operation, not on the "efficiency" of the heater itself.
Now, in the case of a "steamer", I presume you mean a device which heats water to boiling and then releases that hot steam into the air or into some further process. The fact that the hot steam that leaves the steamer vessel takes heat energy with it does not reduce the device's efficiency. That hot steam is part of the output and therefore must be included in the efficiency calculation of (energy out) / (energy in). Moreover, the "energy" in hot steam is not just in its temperature, but includes also the Heat of Vaporization it carries by virtue of having been converted from the liquid to the gaseous physical state.
Think of thermal conductivity and the wires attached to the element.
OK. Please tell me the "efficiencies" of the following:
a.) Ideal Kettle with 1500 watts heat input and 0 watts heat loss
b.) Kettle with 1500 watts heat input and 750 watts heat loss.
c.) Ideal Steam Maker with 1500 watts heat input and 1500 watts heat loss via steam.
I maintain in ALL cases the "Efficiency" of the heating element is 100% - that is, if you accurately measure the voltage being supplied to the heater terminals, the current flowing through it and the time it is powered on, ALL of that energy will be converted into heat in the water. BUT, if we consider instead the "Efficiency" of the entire kettle - voltage, current, and time measured at the kettle plug's blades - then my reply to your cases is:
a.) Zero heat loss in any form, it's 100% Efficient. That ultimately will change, though, once you get all the water to 100C and start converting water liquid to stream - you will either lose steam from the system right away, or will build up pressure in a closed container until something blows up!
b.) 1500 W input, 750 W heat loss rate, the system is 50% "Efficient" in terms of delivering heat to the desired point (hot water still in the kettle ready to use); half the energy input is "lost" to surroundings and not usable.
c.) Steam Maker using 1500 W power input with zero heat loss from itself, but allowing a continuous flow of steam to exit, carrying 1500 W heat flow with it: that also is 100% "Efficiency" because the heat ALL is being delivered to the intended point (hot steam for use elsewhere) and none is being lost to surroundings.
a.) Zero heat loss in any form, it's 100% Efficient. That ultimately will change, though, once you get all the water to 100C and start converting water liquid to stream - you will either lose steam from the system right away, or will build up pressure in a closed container until something blows up!
b.) 1500 W input, 750 W heat loss rate, the system is 50% "Efficient" in terms of delivering heat to the desired point (hot water still in the kettle ready to use); half the energy input is "lost" to surroundings and not usable.
c.) Steam Maker using 1500 W power input with zero heat loss from itself, but allowing a continuous flow of steam to exit, carrying 1500 W heat flow with it: that also is 100% "Efficiency" because the heat ALL is being delivered to the intended point (hot steam for use elsewhere) and none is being lost to surroundings.
OK. What is the "efficiency" of:
1.) An ideal steam machine with a power input of 1500watts and 1500watts lost via steam. The steam is used to humidify and warm up a room.
2.) A kettle with a power input of 1500watts and 1500watts lost via steam. Being a poorly designed kettle, it consequently humidifies and warms up a room.
3.) A dual purpose kettle (twice the bang for your buck
) with 1500watts power input and 750watts lost via steam. The steam is used to humidify and warm up a room, and the hot water is used for tea.
Last edited:
Now it comes down to what you want to do with the device, which will lead to what your definition of "Efficiency" is. I'm suggesting that if "Efficiency" generally means (Energy delivered to an intended use) / (Energy input to the machine), the "intended use" part is the key element in the definition.
Now, IF I accept that I want humidity AND heat delivered to the room, PLUS I also contend that hot water to make tea is a desired use, then every one of your cases is 100% efficient in delivering energy to where I want.
In cases 2 and 3, suppose we are in a very warm room and are using an air conditioner to cool it down, but we really do want the humidity because this is Arizona with hot dry air outside. In those cases you could take two views. One might be to say that the heat delivered to the room is NOT part of our desired use, and hence "wasted energy". That will lead to an "Efficiency" less than 100%. Alternatively I can take the view that Efficiency still is 100% in terms of converting all electrical energy into heat energy that enters the room; however, I have an additional but separate task to reduce the room temperature which will involve use of another system (an air conditioner) requiring further expenditure of energy.
This is a good example of a general principle. People use words and language to communicate concepts. When there is complete agreement on word definitions and language structure, the communication can be "100% Efficient". However, lots of times that complete agreement is missing BUT presumed to exist, and mis-communication results. My own pet sub-topic of this is, in fact, the use of "percent". I ALWAYS want to know, "% of what?, because the item after "of" is the denominator in calculating the % number. In the current discussion, we also are clarifying what goes into the numerator in order to understand each other's true meaning.
Now, IF I accept that I want humidity AND heat delivered to the room, PLUS I also contend that hot water to make tea is a desired use, then every one of your cases is 100% efficient in delivering energy to where I want.
In cases 2 and 3, suppose we are in a very warm room and are using an air conditioner to cool it down, but we really do want the humidity because this is Arizona with hot dry air outside. In those cases you could take two views. One might be to say that the heat delivered to the room is NOT part of our desired use, and hence "wasted energy". That will lead to an "Efficiency" less than 100%. Alternatively I can take the view that Efficiency still is 100% in terms of converting all electrical energy into heat energy that enters the room; however, I have an additional but separate task to reduce the room temperature which will involve use of another system (an air conditioner) requiring further expenditure of energy.
This is a good example of a general principle. People use words and language to communicate concepts. When there is complete agreement on word definitions and language structure, the communication can be "100% Efficient". However, lots of times that complete agreement is missing BUT presumed to exist, and mis-communication results. My own pet sub-topic of this is, in fact, the use of "percent". I ALWAYS want to know, "% of what?, because the item after "of" is the denominator in calculating the % number. In the current discussion, we also are clarifying what goes into the numerator in order to understand each other's true meaning.
OK. But for case #2, if there is no hot water for tea, how can you say it is 100% "efficient" (your definition)? Being a kettle without hot water to dispense? The hot water being the "intended use" of the kettle?
What I am getting at is your definition of "efficiency" is contradictary. The exact same physical model can have varying "efficiencies" based on whim, desires, feelings, and judgements.
Last edited:
Right! You got it! "Efficiency" is not an absolute term. YOU have to define what your end goal is. For Case #2, if you find real value in the heat and humidity, you have full efficiency. If you think those are just wasted and your ONLY goal was hot water for tea, then efficiency is roughly zero.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter