egg boiler using electric kettle

[elfcurry]

I've had in mind for a while using an old electric kettle with a failed automatic cut-off to boil eggs.

A kettle is designed to boil the water quickly using 2-3kW element and then switch off. Usually they take quite a few seconds of vigorous boiling before you hear a 'click' as this happens. My little project will be different.

The idea is that a microcontroller controls the power and detects the boiling and then either reduces the power or just turns on and off repeatedly to keep the water just on a gentle rolling boil until the set time elapses when (as a stylish robotic extra) a mechanical mechanism could lift the eggs out of the water. The power reduction or cycling is to avoid the full 2-3kW being wasted just to fill the kitchen with steam! Once boiling the power needed to overcome heat losses will be quite a lot less.

The main problems to overcome are:

1. electronics safely controlling a mains current (say 10A) at 240V AC
2. detecting the boiling
3. the lifting mechanism (this is optional and can be ignored for now)

For reducing the effective power I think it makes sense to use timing, whether for quite short periods triggering a Triac or much longer periods of a few seconds. I don't think phase delay timing within cycle makes sense due to electrical noise so I need to find out how to trigger a chunky Triac. Maybe use a solid-state relay.

Detection of the boiling is the other thing. Temperature detectors which can interface to the electronics and stand being boiled and not be a hazard to the food and be cleaned from egginess seems a tall order. Also I'm not sure measuring temperature is very useful as it won't really change once it's reached boiling point so it won't respond quickly.

So my idea is to use a small mic to detect the sound of the bubbling. But what signal would the μC see?

Your input would be welcome but I should say that I usually take a long time to do things and often don't finish projects so please don't invest a whole lot of time or effort.

Thanks.

 

[picbits]

I've always been taught to put the egg in cold water, bring the water to the boil, turn off and wait 7 minutes for a soft to medium boiled egg or 10 mins for a hard boiled egg.

Obviously depends on the volume of water and quantity of eggs being cooked but as a rule of thumb seems to work.

A thermistor to detect 100 degrees C - or a car temperature sender would also do - some of the radiator fan switches have a 95-100 degree cut off point. Interface to a relay, one shot switch and time and job done.

 

[Reloadron]

Given a choice I would just use an SSR for handling power to the heating element. A SSR with a 3 - 32 volt trigger. The heating element is a resistive load and there is no reason to get fancy. That said, I would use a thermistor embedded in epoxy as a sensor, get creative and make your own sensor housing, a drinking straw might be a good start. I would not use any form of microphone to detect boiling. Why? The eggs are in their shells and there are plenty of non-toxic temperature sensor solutions. While I would not bother with a uC and likely use a simple comparator circuit On/Off with hysteresis if you want to use a uC then have at it. Configure the thermistor in a voltage divider configuration and run the output into an analog input of your uC.

Water boils at5 sea level at 100C so write your code so 100C is the SP (Set Point) and actual temperature is your PV (Process Variable). When the PV is below SP the output is 100% or code it however you wish. If you want to pulse the element write code so when the PV is say 90C put it in your code. The bottom line is the water at sea level will never get hotter than 100C at which point you are making steam. Using a uC you have the benefit of programming your control however you wish. Just a matter of your creative imagination. Use another DO from the uC to raise/lower your basket.

<EDIT> PC Bits beat me. </EDIT>

Yeah, I start with the eggs in cold water, bring to a boil, wait and then place the eggs in cold water.

Just My Take...
Ron

 

[elfcurry]

I've always been taught to put the egg in cold water, bring the water to the boil, turn off and wait 7 minutes for a soft to medium boiled egg or 10 mins for a hard boiled egg.

Obviously depends on the volume of water and quantity of eggs being cooked but as a rule of thumb seems to work.

A thermistor to detect 100 degrees C - or a car temperature sender would also do - some of the radiator fan switches have a 95-100 degree cut off point. Interface to a relay, one shot switch and time and job done.

Thanks, it's worth looking at alternatives. I know there isn't just one method to boil eggs so if I build it my way, I could still try your method - boil from cold and let it cook in the water's heat for a known period. That period to cook as desired must depend on the amount of water present though, and the ambient temperature. I still prefer my cooking method as it's repeatable and you can play with the parameters to see what's best, probably moreso than doing it 'manually' unless you're very methodical.

The thermistor is something I hadn't thought of. I'd need to read its resistance using analogue input but that's ok. My concern is that measuring temperature won't be so much use to keep it boiling properly as it'll be at the same temperature all the time until it goes off the boil. The idea of going by the sound it makes is because I suspect the vigour of the boiling will change with power while temp won't nearly so much. But I'm not sure. I'll keep both in mind for now.

Thanks.

 

[elfcurry]

Given a choice I would just use an SSR for handling power to the heating element. A SSR with a 3 - 32 volt trigger. The heating element is a resistive load and there is no reason to get fancy. That said, I would use a thermistor embedded in epoxy as a sensor, get creative and make your own sensor housing, a drinking straw might be a good start. I would not use any form of microphone to detect boiling. Why? The eggs are in their shells and there are plenty of non-toxic temperature sensor solutions. While I would not bother with a uC and likely use a simple comparator circuit On/Off with hysteresis if you want to use a uC then have at it. Configure the thermistor in a voltage divider configuration and run the output into an analog input of your uC.

Water boils at5 sea level at 100C so write your code so 100C is the SP (Set Point) and actual temperature is your PV (Process Variable). When the PV is below SP the output is 100% or code it however you wish. If you want to pulse the element write code so when the PV is say 90C put it in your code. The bottom line is the water at sea level will never get hotter than 100C at which point you are making steam. Using a uC you have the benefit of programming your control however you wish. Just a matter of your creative imagination. Use another DO from the uC to raise/lower your basket.

<EDIT> PC Bits beat me. </EDIT>

Yeah, I start with the eggs in cold water, bring to a boil, wait and then place the eggs in cold water.

Just My Take...
Ron

Thanks Ron - same cooking method as picbits. I also start with the egg in cold water but time it from boiling and keep it just boiling. That's what I'm trying to automate.

I agree that a solid state relay is probably the way to go but I still prefer to keep it on the boil rather than change cooking method. This means ongoing measurement and power cycling.

You also suggest a thermistor and I'm starting to wonder if my relying on the change in sound makes sense.

Otherwise it's pretty much what I was thnking. In a past project I used a μC to control a seedling tray heater (15W) which was dirt cheap but got too warm. I used a Dallas DS18B20 digital sensor chip and had a small LCD display to show the current temp and SSR, with LED to show when the heater was on.
I think this is conceptually fairly similar except it controls more power and it's too hot for a DS18B20.

 

[ccurtis]

My concern is that measuring temperature won't be so much use to keep it boiling properly as it'll be at the same temperature all the time until it goes off the boil. The idea of going by the sound it makes is because I suspect the vigour of the boiling will change with power while temp won't nearly so much.

I'm thinking that measuring the vapor pressure developed would be a better indicator, and a fast responding process variable as well. Pressure sensor could be mounted in the cover of the kettle.

 

[Reloadron]

Hi Ya elfcurry,

OK, let me clear up a little of what I mentioned. My suggestion to use a SSR to control your power to the heating element was simply to control the power. You could build your own triac circuit but I see an off the shelf SSR as a simple solution that can use a nice 5 volt control voltage to turn your 220 VAC on and off. As to the sensor? A thermistor is but one of several ways to go about sensing the water temperature. For example a thermistor like this one has a cost of about $1.95 USD and is pretty inexpensive. Looking at the data sheet it has an upper range of -40 to 125 degrees C which is adequate. When it comes to the sensor I figure it this way, both my micro-wave oven and my kitchen gas stove oven have probes that can be inserted into meats (or other foods) that are totally non toxic and leave no residue in the foods. If I can determine when the internal temperature of a roast beef is 140 degrees F (I like medium rare beef) it can surely determine when water reaches 100 degrees C. Anyhow, the sensor is your call, I merely suggested a thermistor but there is no shortage of other sensor options out there that would do just fine. A Google of "thermistor arduino" will bring up a dozen hits or more about interfacing a thermistor to for example an Arduino uC.

As to using any form of for example PID (Proportional Integral Derivative) control I would just run with On/Off control. Assuming around sea level water will boil at 100C. At 300 meters above sea level that only will drop by a few degrees. Matters not how much heat our elements are supplying the hottest the water will get id the boiling point. If we turn the elements off the water will stop a rolling boil pretty quick. So given a choice using a uC I would start my eggs in cold water and apply the heat. Once the water temperature reaches 100C start a timer function where the digital out of your uC remains on for about another 15 or 20 min. My attempts at hard boiled eggs always suck. Truth is if I want to hard boil 5 eggs I start with 10 or more just to get 5 good eggs. My wife can make perfect easy to peel eggs every time but mine are a disaster! So after a prescribed time turn off the heaters. This is typically called a ramp and soak cycle in heating applications, we ramp the water temperature up to 100C and then start a timed soak cycle. When the cycle times out the digital out to the SSR goes low and that is about it.

Actually the DS18B20 has an upper range of 125C I believe and it could be used in conjunction with a uC. Back to choosing a sensor.

Ron

 

[Reloadron]

I'm thinking that measuring the vapor pressure developed would be a better indicator, and a fast responding process variable as well. Pressure sensor could be mounted in the cover of the kettle.

Seems more complex to me. As water is heated it expands before turning to a gassious sate at 100C. No problem when open to atmosphere. But also no pressure to detect when vented to atmosphere. There can be a big problem when it is not vented and pressure builds up. This is why a pressure cooker has a relief valve and even those things scare me.

Ron

 

[ccurtis]

Seems more complex to me. As water is heated it expands before turning to a gassious sate at 100C. No problem when open to atmosphere. But also no pressure to detect when vented to atmosphere. There can be a big problem when it is not vented and pressure builds up. This is why a pressure cooker has a relief valve and even those things scare me.

Ron

Yes, Ron, granted, somewhat more complicated. I didn't mention that I was also thinking of placing a vent hole in the cover to limit the maximum pressure and use a sensistive differential pressure sensor to measure only the difference between the pressure inside and outside the cover. Just my personal two cents. If it's the degree of boil I'm interested in, then that's what I want to measure. I would probably still installed a temp sensor just to cut everything off should the water happen to boil off completely and even as a secondary process variable.

 

[Reloadron]

Funny, as when you start really thinking about this, there is no limit as to what can be done. Reminds me of some of the projects over the years at work. Someone wants to do something and you say OK, we can make that happen. So you start kicking around a few ways to make it happen. Then someone comes along and says it would be nice if it also did this or that, so the basic design has a few new features added. Finally you look at the project and remember what it was originally and how it evolved, which is pretty cool.

You start with a plan for a basic egg cooker and begin adding features. Each feature is another "how to" like you just mentioned for example, what if I lose my water? What if all my water boils off? How do I sense if this happens? What sensing method to use? Ah, a fail safe smart egg cooker.

I should share the story of how once, when I was much, much younger I set 6 eggs on the stove to boil and promptly fell asleep. This was after a night that involved several pubs. I never knew 6 eggs could make so much horrible smoke and stink so bad! It was ugly!

I hope some other forum members chime in with some ideas as to making a good, reliable, egg cooker. I would like to see some other thoughts on how to go about it beyond my thinking.

Ron

 

[()blivion]

The potential problem that I see with using just a temperature sensor to control something like this is that if you aim for a set point of say 100c, and you are actually at a higher altitude, or the barometric pressure changes a large amount, you run a modest risk of *NOT* properly detecting boil, and thus not shutting down the heat. This happens because water boils at whatever temperature it boils at... and no higher... for a given pressure.

This same physics is why you can boil water in a paper cup.

 

[elfcurry]

consider latent heat of vaporisation

I'm thinking that measuring the vapor pressure developed would be a better indicator, and a fast responding process variable as well. Pressure sensor could be mounted in the cover of the kettle.

Nice thought and in a sealed container this could work but my kettle is plastic with a far-from-airtight lid. I don't want to make it airtight and if I did it would affect the boiling point and complicate things.

The potential problem that I see with using just a temperature sensor to control something like this is that if you aim for a set point of say 100c, and you are actually at a higher altitude, or the barometric pressure changes a large amount, you run a modest risk of *NOT* properly detecting boil, and thus not shutting down the heat. This happens because water boils at whatever temperature it boils at... and no higher... for a given pressure.

This same physics is why you can boil water in a paper cup.

Well, thinking back to school science lessons, I think there's a bigger problem than altitude or atmospheric pressure: latent heat of vaporisation.

I don't see the point of measuring the temperature for this project*. If I wanted to keep it at, say, 70°C, where the water isn't changing phase from liquid to vapour, then measuring temperature would be fine but this is a special case: boiling point. For a normal kettle it's easy - just detect boiling (with a bi-metal strip which allows some latitude) then switch the power off. Job done.

My aim is to keep the water at boiling point for several minutes (period determined by me) without wasting power or boiling the water away or filling the kitchen with steam! The problem is to detect when it's boiling, and turn off the heat briefly until it starts to lose vigour and then turn it on again.

From www.britannica.com latent heat

For example, when a pot of water is kept boiling, the temperature remains at 100 °C (212 °F) until the last drop evaporates, because all the heat being added to the liquid is absorbed as latent heat of vaporization and carried away by the escaping vapour molecules.

Since the temperature can't change, it can't be the main input to the control system*. That's why I proposed using the sound of the boiling - though admittedly details are hazy! Maybe my thinking is wrong but if so, I can't see it yet.

I should also say that I'm not talking about hard-boiled eggs to be eaten cold later. I mean normal, soft-boiled eggs eaten hot at breakfast. Cooking time is critical, hence the control system and attempt to make it just right!

Thanks all, for your help so far. Don't give up on me!

*Measuring temperature would be helpful as a safety input.

 

[elfcurry]

Hi Ya elfcurry,

OK, let me clear up a little of what I mentioned. My suggestion to use a SSR to control your power to the heating element was simply to control the power. You could build your own triac circuit but I see an off the shelf SSR as a simple solution that can use a nice 5 volt control voltage to turn your 220 VAC on and off.

Hi Ron, I've had a look at SSRs on ebay and found one which can control 25A at 240V with a logic input. It didn't say whether it's resistive load but as mine is, it's fine. I do want to learn about triggering a triac to control mains but that can wait. SSR will do for this.

As to the sensor? A thermistor is but one of several ways to go about sensing the water temperature. For example a thermistor like this one has a cost of about $1.95 USD and is pretty inexpensive. Looking at the data sheet it has an upper range of -40 to 125 degrees C which is adequate. When it comes to the sensor I figure it this way, both my micro-wave oven and my kitchen gas stove oven have probes that can be inserted into meats (or other foods) that are totally non toxic and leave no residue in the foods. If I can determine when the internal temperature of a roast beef is 140 degrees F (I like medium rare beef) it can surely determine when water reaches 100 degrees C. Anyhow, the sensor is your call, I merely suggested a thermistor but there is no shortage of other sensor options out there that would do just fine. A Google of "thermistor arduino" will bring up a dozen hits or more about interfacing a thermistor to for example an Arduino uC.

A thermistor is something I know nothing about so it's another thing to learn if I go that way. But see my post above for why temperature may not be the best thing to measure. Toxicity wasn't a major concern, just something I thought I should not ignore.

As to using any form of for example PID (Proportional Integral Derivative) control I would just run with On/Off control. Assuming around sea level water will boil at 100C. At 300 meters above sea level that only will drop by a few degrees. Matters not how much heat our elements are supplying the hottest the water will get id the boiling point. If we turn the elements off the water will stop a rolling boil pretty quick. So given a choice using a uC I would start my eggs in cold water and apply the heat. Once the water temperature reaches 100C start a timer function where the digital out of your uC remains on for about another 15 or 20 min. My attempts at hard boiled eggs always suck. Truth is if I want to hard boil 5 eggs I start with 10 or more just to get 5 good eggs. My wife can make perfect easy to peel eggs every time but mine are a disaster! So after a prescribed time turn off the heaters. This is typically called a ramp and soak cycle in heating applications, we ramp the water temperature up to 100C and then start a timed soak cycle. When the cycle times out the digital out to the SSR goes low and that is about it.

Actually the DS18B20 has an upper range of 125C I believe and it could be used in conjunction with a uC. Back to choosing a sensor.

Ron

I wasn't aware that DS18B20 could go past boiling but I am now.

Also see my previous post about soft-boiled eggs. Firm white with runny golden yolk, eaten with an egg spoon and served with hot buttered toast* and a pot of tea. Mmmmm!

*My toaster modifications can be a separate thread.

 

[Reloadron]

The potential problem that I see with using just a temperature sensor to control something like this is that if you aim for a set point of say 100c, and you are actually at a higher altitude, or the barometric pressure changes a large amount, you run a modest risk of *NOT* properly detecting boil, and thus not shutting down the heat. This happens because water boils at whatever temperature it boils at... and no higher... for a given pressure.

This same physics is why you can boil water in a paper cup.

I agree with that to a point and touched on it earlier. Water boils at 212 degrees F or 100 degrees C at 1 atmosphere (sea level). The standard atmosphere (symbol: atm) is a unit of pressure equal to 101.325 kPa ("kiloPascals") or 1013.25 millibars or hectopascals. It is equivalent to 760 mmHg (torr), 29.92 inHg, according to the trusty Wikipedia. Keep in mind that this assumes pure water and tap water is not pure water. Not only does the atmospheric pressure change daily but as mentioned changes with altitude.

The OP is on the South Coast of the UK. While my UK geography leaves much to be desired if I look at London I see that London can vary from a high point of 245 meters (804 ft) in the Bromley area to a low point of about 100 meters (328 ft) in the Highpoint Hill area. The difference between sea level and an altitude of 457 meters (1,500 feet) is just about negligible. At sea level water boils at 100C or 212F and at 609 meters (2,000 feet) water boils at 97.78C or about 208F.

Since I am applying constant heat from a cold start the water will eventually boil and when it does boil that is as hot as it gets. If I start my soak cycle time I mentioned earlier at for example when the water reaches 97C I can pretty much figure my eggs will cook just fine. However, I could calibrate my temperature sensor for my location. If I want to account for barometric pressure I could include a atmospheric pressure sensor in the scheme and calibrate that too. I mentioned adding features earlier also. Just a matter of where I stop spending money. I can build a $20 USD egg cooker or I can build a $5,000 USD egg cooker I guess.

<EDIT> All bets are off, I have been thinking hard and not soft boiled eggs! </EDIT>

Ron

 

[elfcurry]

Funny, as when you start really thinking about this, there is no limit as to what can be done. Reminds me of some of the projects over the years at work. Someone wants to do something and you say OK, we can make that happen. So you start kicking around a few ways to make it happen. Then someone comes along and says it would be nice if it also did this or that, so the basic design has a few new features added. Finally you look at the project and remember what it was originally and how it evolved, which is pretty cool.

You start with a plan for a basic egg cooker and begin adding features. Each feature is another "how to" like you just mentioned for example, what if I lose my water? What if all my water boils off? How do I sense if this happens? What sensing method to use? Ah, a fail safe smart egg cooker.

I should share the story of how once, when I was much, much younger I set 6 eggs on the stove to boil and promptly fell asleep. This was after a night that involved several pubs. I never knew 6 eggs could make so much horrible smoke and stink so bad! It was ugly!

I hope some other forum members chime in with some ideas as to making a good, reliable, egg cooker. I would like to see some other thoughts on how to go about it beyond my thinking.

Ron

Re: no limits - yes, I agree that there are all sorts of projects possible around the home or garden or some workplaces (though safety regulation may complicate things ....)

I like the simple application of technology to solve simple problems. Not expensive, sophisticated systems where you (I) spend money but someone else has the fun of designing and making it but simple inexpensive things to make life easier or better. I almost always want to involve a microprocessor as it means you can make 'intelligent' decisions and record events. It's not always necessary though.

Re: six over-cooked eggs: - I'll take care not to emulate your drunken error.

"... ideas as to making a good, reliable, egg cooker" Agreed. Once working I can tinker with the process or try things. For instance, is boiling temperature needed or does it just make normal human cooking easy as you know the temperature? Would 90C or 95C do just as well (but take longer)?

Another thing I wonder is whether cooking is endothermic process. ie does cooking actually take in some energy to change the food physically from raw to cooked state, or does it just require the presence of the heat but it's all still there as heat at the end?

Many years ago (>30) I had a work placement period of six months between years as a physics undergraduate. Among other tasks a food packaging company wanted me to calculate the energy requirements for several commercial scale cooking processors to cook prepared food in sealed plastic pouches. Each 'cooker' was a cylinder several metres long and diameter. One used hot water and lost it every cooking cycle, another was similar but saved and re-used the water while another used steam to cook. I asked for all sorts of measurements including whether energy was needed to actually cook the food. They didn't know and I've never found out. Maybe I can now!

 

[ccurtis]

Though the boiling temperature of the water can vary significantly, what we do know is that the temperature of the water no longer increases when boiling is reached. That can be used as an indicator for boiling.

 

[()blivion]

Good idea ccurtis. I agree, and it is very simple.

 

[()blivion]

By the way, the process of cooking is basically "Denaturation", which is an endothermic process.

 

[atferrari]

They found a way to look more scientific, why not?

 

[elfcurry]

Though the boiling temperature of the water can vary significantly, what we do know is that the temperature of the water no longer increases when boiling is reached. That can be used as an indicator for boiling.

True. How can this be used? Heat is added and the temperature rises until a plateau is reached, then you know it's boiling. It could be a method for determining the boiling point. But if I had a temperature sensor on my egg-cooking kettle I'd turn on the heat anyway as soon as cooling is noticed.

By the way, the process of cooking is basically "Denaturation", which is an endothermic process.

Excellent, thanks! I'd suspected so but it's nice ot have it confirmed.