[HN Gopher] Receive push notifications from your rice cooker
___________________________________________________________________
Receive push notifications from your rice cooker
Author : ColinWright
Score : 46 points
Date : 2024-04-02 03:50 UTC (19 hours ago)
(HTM) web link (shkspr.mobi)
(TXT) w3m dump (shkspr.mobi)
| eternityforest wrote:
| I remember when Fuzzy Logic was an everyday tech term that seemed
| to be the big new hotness. I think people were excited about how
| they used it in the mass combat simulator software for LoTR.
|
| Now it seems to be mostly associated with rice cookers. They must
| do an amazing job, since it still seems to be people's favorite
| rice cooking tech!
| morder wrote:
| https://www.youtube.com/watch?v=RSTNhvDGbYI Great Technology
| Connections video about how it works.
| gifvenut wrote:
| Your video explains how traditional rice cookers work.
| However it doesn't explain how fuzzy logic in a rice cooker
| works.
| 0cf8612b2e1e wrote:
| I thought that was all marketing nonsense and they actually
| work by detecting when the temperature spikes because the water
| has been consumed.
|
| If I head to head my zojirushi vs a bargain rice cooker with
| temperature detector, would I be able to tell the difference?
| MisterTea wrote:
| The cheap cookers use a mechanical setup as there are no
| electronics. I have the cheapest Zojirushi 3-cup cooker where
| you just push down a lever. That lever mechanically latches a
| spring loaded mechanism that closes the heating element
| switch. The temperature will stay relatively stable as the
| water absorbs the heat keeping it at around 100 C. Once the
| water is evaporated from the bottom the temperature rises
| which trips a bimetallic mechanism releasing the spring
| loaded lever opening the heater switch. Stupid simple.
| kjellsbells wrote:
| Tangential, but one reason zojirushi is a buy it for life
| product is because you can get spare parts for a very long
| time. I have a 20 year old zj with no fundamental failures
| beyond the cookpot getting old and dented, and zj part
| replacements are easy and cheap.
| yborg wrote:
| Try replacing the clock battery on a unit that has one :/
| gomox wrote:
| Disappointing aspect indeed, but the battery is only
| relevant to people that use the clock to set cooking
| times a lot AND ALSO keep their rice cooker unplugged.
| The intersection of those 2 groups is likely minimal, and
| the slight inconvenience of setting the time for the
| aforementioned group is not that bad.
| invalidator wrote:
| Mine consistently produces better rice than my old mechanical
| one. It's also slower. I think it does a more gentle
| temperature ramp than the mechanical one did.
| gomox wrote:
| I replaced a crummy spring/latch one with a Zoji induction 3
| cup and I would never go back to the old one. Rice is always
| evenly cooked and never stuck/toasted at the bottom, keep
| warm with tracking of elapsed time, countdown to being done,
| plus flexibility for lots of different type of rice and
| grains.
|
| Could you do the job with a cheap one? Sure, but you can
| always make rice on the stovetop as well if savings/space are
| a big priority. This is a convenience appliance, and the
| convenience of the fancy ones is a significant improvement
| over the basic ones.
| orbisvicis wrote:
| When I cook rice I purposefully take the lid off and add 10
| extra minutes to get that delicious crispy bottom.
| gravescale wrote:
| I've always assumed it was a name for a clever way to improve
| on limitations of simple bang-bang control with on-off sensors
| and actuators plus some 7400 series-type logic. Nowadays you
| can have an STM8 with two dozen 10 or 12-bit ADCs and enough
| grunt to do hundreds of PID loops at once for 10 cents.
| jerf wrote:
| My understanding is that there was a period of time where
| computer science was very excited about fuzzy logic, because it
| was thought it would prove to be more powerful than
| conventional logic.
|
| https://en.wikipedia.org/wiki/Fuzzy_logic
|
| However, it was demonstrated that it had no additional power
| over conventional logic, so interest faded as quickly as it
| rose. Which is not to say it is useless, but that rather than a
| revolution it became just another tool in the belt. It can be
| more convenient to formulate certain systems in fuzzy logic
| rather than binary logic, but it doesn't really create new
| possibilities. And programmers in the trenches were doing this
| sort of thing already anyhow, less formally, but generally
| effectively.
|
| It was weird how loud it was for sure. New forms of logic don't
| generally get the red carpet treatment like that.
|
| I don't know why a rice cooker would particularly care about
| "fuzzy logic" other than a bizarre marketing spandrel [1].
|
| [1]: Since this seems an obscure usage of an already-obscure
| term: https://www.scienceabc.com/pure-sciences/what-are-
| spandrels....
| jbverschoor wrote:
| Control theory is very interesting and in my opinion more
| relevant than before
| avidiax wrote:
| Fuzzy logic sounds like it has been superseded by neural
| networks. It has much the same principle, i.e. gates/neurons
| that are tuned to provide a non-digital output. But with
| fuzzy logic, presumably the parameters and network are hand
| tuned.
| sunshowers wrote:
| I don't think "superseded" is the right word, in the same
| way that ordinary relativity [1] hasn't been superseded by
| special or general relativity. Fuzzy logic is a more
| straightforward model that works well enough here, so why
| throw in the additional complexity and illegibility of
| neural nets?
|
| [1] Also known as "classical mechanics". Despite widespread
| belief that relativity is a 20th century idea, the first
| person to describe relativity wasn't Einstein -- it was
| Galileo. Einstein himself saw his work on relativity as
| extending what Galileo had done.
| barbazoo wrote:
| > Step 2 - Install Home Assistant on a Raspberry Pi
|
| In hindsight, I wouldn't want to install HA for the sole purpose
| of running one smart plug. HA is one of the best pieces of
| software I've worked with but it has a learning curve and it does
| require some tinkering every now and then.
| edent wrote:
| I should have mentioned, I have HA for all sorts of different
| things - not just this plug.
| TeMPOraL wrote:
| On the other hand, HA tends to act as an attractor for this
| kind of stuff. I installed mine first to play with Hue lights,
| then to try and add my A/C units to it. With the latter turning
| into a spectacular success (infinitely better than vendor's
| garbage app, plus actually liked and used by my wife), HA now
| also runs hacked IKEA air quality monitors, floor heating, and
| most recently, reports when the washing machine is done. It's
| now pretty much critical infrastructure for us.
| frereubu wrote:
| The alternative to this is to buy an expensive rice cooker like a
| mug (me) and after a few months realise that the last 20-odd
| minutes on the display go down _much_ faster than 20 minutes.
| Fuzzy logic indeed.
| freeplay wrote:
| My rice cooker switches to "keep warm" mode once it's done
| cooking. I think most do this. Is the drop in power consumption
| significant enough to reliably know when its done?
| esel2k wrote:
| I think the author says it's about rice beeing overcooked. Even
| if it turns to keep warm its better to directly open the lid
| and go eating.
| edent wrote:
| Yes. It has been reliable so far. Keeping warm uses less
| electricity than cooking.
| geor9e wrote:
| Haha, I read the headline and my brain went "I bet it's a Energy
| Monitoring Smart Plug on Home Assistant on a Raspberry Pi".
| Honestly a simple automation with device trigger: smartplug
| "power changes" with below: X watts might have worked too if it
| only crosses in that direction once per cook. But it's nice that
| a fancy appliance monitoring add-on exists - some appliances
| probably have a few weird power cycles per cook.
| pvelagal wrote:
| beep beep beep. They have alarms for that ! :) But nice fun
| project !
| vundercind wrote:
| And keep-warm. Most of the reason I use a rice cooker is so I
| _don't_ need to monitor it or try very hard to time the finish
| right.
| djhworld wrote:
| Neat!
|
| I was always under the impression that rice cookers worked on a
| timer basis, but from the article it sounds like it's more of a
| thing where the appliance will tell you when the rice is cooked -
| is that right?
|
| I'm intrigued to get one now, although it's unclear to me what
| the difference is between a PS40 one and the one mentioned in the
| post (which looks to be about PS100 now on Amazon)
| doctoring wrote:
| I love rice cookers!
|
| Rice cookers (usually) make clever use of 1) alloys whose
| magnetism depends on temperature and 2) the fact that boiling
| water occurs at a fixed temperature. With a "trigger"
| temperature just above the boiling point of water, the rice
| cooker automatically turns off the heating element when all the
| water is gone (and thus the temperature starts to rise above
| boiling point).
|
| More detail here: https://youtu.be/RSTNhvDGbYI
| SamBam wrote:
| Those are the older ones. The new ones like the "Fuzzy Logic"
| line are more akin to a PID controller trying to make the
| temperature/time graph match a specific curve, where the
| temperature changes at different points in the cooking, if I
| understand correctly.
| mauvehaus wrote:
| I'd like to see a patent that actually confirms this. The
| Curie temperature of the common ferromagnetic metals is well
| above the boiling point of water, which raises some questions
| as to what one would make the alloy out of.
|
| Canadian nickels were made of nickel until 1981. Assuming you
| have a magnet that's heat resistant, you can do a cool demo
| with a torch or gas stove: pick up the coin with the magnet
| and dangle it in the flames. You'll know when it hits the
| Curie temperature. It'll fall off the magnet.
| LorenPechtel wrote:
| The fancier rice cookers monitor what's going on inside and
| turn off when the last of the liquid is gone. At the lower end
| I do not like them because that ends up somewhat overcooking
| the rice next to the heating surface--I prefer putting it in
| the microwave using settings I have determined by experiment to
| work. Perfect every time.
|
| I presume the more expensive ones are better at avoiding
| overcooking but it seems to me hard to conceive of a
| conduction-heat system that doesn't suffer this problem to some
| degree.
| hunter2_ wrote:
| > putting it in the microwave using settings I have
| determined by experiment to work. Perfect every time
|
| Does that involve a humidity sensor in your microwave?
| Because if you're just talking about power level and time,
| then how do you account for variation of water temperature? I
| guess you could use room temperature water (or a fridge
| pitcher) which would be plenty stable, but tap temperature
| varies wildly based on the season, at least where I live.
| fsckboy wrote:
| I was taught by a chinese restaurant chef his technique for
| how to cook chinese rice in a normal pot on the stove at
| home. The technique consists of (after washing away most of
| the starch) starting out by submerging the rice under an
| extra couple of centimeters of water (you use your
| hand/knuckles to measure, it's not super precise) and high-
| heat boil uncovered till the water level drops to even with
| the surface of the rice. Now cover and put heat down very
| low and let the rest of the water absorb while you make
| your stir fry. Before serving, turn all the rice over a
| time or two with a paddle and let it resettle uncovered,
| this helps the rice dry a little and be less sticky, but
| keeps it cluster-chop-stick-able.
|
| I explained all that to say, I adapted that technique for a
| microwave and it worked great right away. Microwaves vary a
| lot in power and timing, so I can't really give a how-many-
| minutes recipe, but the technique just works.
| morepork wrote:
| There are different ways of doing it, but when I microwave
| rice I boil the water in a kettle first
| hunter2_ wrote:
| Right. Before all water at the bottom has evaporated, the
| temperature is 100C. After all water has evaporated, the
| temperature rises uncontrolled due to lack of latent heat. So
| basically it just shuts off when the temperature exceeds 100C,
| and the classic mechanism is to formulate a magnet that loses
| its magnetism at that temperature (all permanent magnets have
| an operating temperature, and it can be tweaked during
| manufacturing) and at that point it's just a leaf spring switch
| that the magnet conditionally attracts.
| Cyph0n wrote:
| The more expensive fuzzy logic-based rice cookers are more
| "accurate" in the sense that they have a better idea of when
| your rice is actually done cooking. This makes your rice
| cooking more deterministic.
|
| As to whether this is worth the extra cost, it depends on your
| budget and how frequently you cook rice.
| instagib wrote:
| I got an expensive cooker and can tell the difference. Cheap
| 3 cup cooker to 10 cup bottom burner to Zojiriushi induction.
| The same amount of rice cooked would not fit in my rice saver
| container for stir frying without being smushed down because
| it was more fluffy.
|
| It takes 3 times longer to cook however yet has not burned my
| rice when sitting on warm yet. I get happy when I smell the
| fresh cooked rice several times per week so I'm definitely in
| the worth it category.
| entropie wrote:
| I do the same with my washer. I monitor the power usage, save it
| in a variable and run a trigger when it drops from a certain
| amount below another.
|
| Super simple, works good.
| LorenPechtel wrote:
| Hey, big companies--there's a market for this! Why is there no
| commercial product that does something of the sort? Power drops
| below X watts for Y minutes, send notification.
| SkyMarshal wrote:
| Don't tip them off, they'll make something that requires a
| cloud service and records every time you cook rice.
| ranger207 wrote:
| That doesn't scale. What, you sell someone a product, they're
| happy with it, it fulfills the problem they had, they don't
| send you another penny until you spend the money to actually
| make a new product? That sounds too expensive, nobody would
| fund that
| fngjdflmdflg wrote:
| This idea doesn't sound like that though. Current rice
| cookers are already what you describe. A rice cooker with an
| app would be more likely to have a subscription model.
| kjellsbells wrote:
| Alternative implementation idea: pi with a mike feeding audio
| signals to a detector. If waveform matches Twinkle Little Star
| (aka sound of rice cooker), send notif("rice is done").
|
| Extensible to other important sounds like smoke alarm, oven timer
| beeping, and a catch-all that sends notif("weird sound in
| kitchen, best investigate").
|
| Plugin architecture to allow other contributions for waveforms to
| recognise different brand gear, eg the beeps from GE being
| different from Zojirushi.
| mjdiloreto wrote:
| Energy use for constant audio monitoring would make that
| implementation impractical compared to simply monitoring the
| energy draw from the device.
| tnmom wrote:
| A Pi draws somewhere under 10W at peak, probably
| significantly under that most of the time. That's about 87kWh
| per year, worst case.
|
| Something like $8-16 dollars depending on how incompetent
| your power company is - well under the threshold for anyone
| who knows how to do this kind of work to worry about. Again,
| worst case; I'd bet you can do constant listening half that.
| syntaxing wrote:
| Home assistant uses some weird "code in YAML" syntax. Chances
| are, the author could have written a virtual binary sensor by
| making one that turns on off according to the meter. I found that
| chatGPT makes writing these yaml files significantly better and
| faster.
| mrkpdl wrote:
| The talkie toaster from Red Dwarf comes to mind... "Does anyone
| want toast?"
| blt wrote:
| In my Cuckoo rice cooker it's better to let the rice sit in "keep
| warm" for a few minutes anyway. When the cooker detects "done"
| there is no water left on the bottom but the rice grains still
| aren't finished absorbing all the water on their surface. I
| haven't noticed any big quality dropoff, even 1hr is fine.
| carabiner wrote:
| The person who invents a fully automated home rice cooker (rinses
| the rice, puts it in a pot, adds water and cooks) will be a
| billionaire. I want to start my rice cooker while I'm driving
| home from a big hike and have that steaming pot ready when I get
| home.
| chewxy wrote:
| A good thing about living in a small house. I can hear when my
| rice cooker is done (plays a tune), when my washing machine is
| done (plays a tune), when my dryer is done (plays a tune). :)
| (note that that hasn't prevented me from hooking them up to smart
| plugs the way terrence did)
___________________________________________________________________
(page generated 2024-04-02 23:01 UTC)