Freeing a Xiaomi humidifier from the cloud
138 points
1 day ago
| 8 comments
| 0l.de
| HN
piskov
1 day ago
[-]
On a tangent note: don’t use ultrasonic humidifiers. Unless distilled water is used, they create a shit-ton of pm2.5 particles.

Use evaporative humidifiers (just disks with myriads of small notches for water to cling on and a fan): https://us.smartmiglobal.com/pages/smartmi-evaporative-humid...

reply
numpad0
4 hours ago
[-]
Don't use evaporative humidifiers(the motorized wet towel). I don't know if it actually cause legionellosis, but it's not very sanitary, and the sanitizing additives for those are known to be actually harmful.

Use boiling type humidifiers (basically just electric tea kettles).

reply
piskov
56 minutes ago
[-]
Water evaporates alomost completely and is changed daily (I rinse both container and disks in the tub with high-pressure water from rotated showerhead), so I don’t think any contamination is likely.

Been doing this for years.

reply
wnevets
19 hours ago
[-]
> On a tangent note: don’t use ultrasonic humidifiers. Unless distilled water is used, they create a shit-ton of pm2.5 particles.

Not according to my uHoo air quality monitor. I have had one running a few feet from the monitor for over a week and there hasn't been any notable increase in PM2.5 particles.

reply
piskov
8 hours ago
[-]
Last time I checked (brought my sensor to the office to one room with ultrasonic) it showed 101/105 ug/m3 for pm2.5/pm10

In the next room (where there were none) it was 6.

Depends on the water, I guess.

reply
wnevets
6 hours ago
[-]
How large is the space with the humidifier? My space is wide open and that might be the biggest reason for my lack of increased PM2.5 readings.
reply
klatchex_too
4 hours ago
[-]
I think the parent poster is correct that it depends on the water.

I have a couple ultrasonic humidifiers, if tap water is put in them it immediately sends the AQI on my air quality monitor into the "Dangerous" level. I have the monitor upstairs and it detected it when my girlfriend put tap water in the humidifier downstairs.

Purified or distilled water works fine. I bought a counter top water distiller because it was a pain lugging 15 gallons of water into the house every week all winter. You can see the residue of whatever was in the tap water at the bottom of the distiller after it runs and it doesn't look like anything I would want to be breathing.

reply
Elucalidavah
18 hours ago
[-]
> any notable increase in PM2.5 particles

What's your PM2.5 baseline, and did you measure TDS in the water?

reply
wnevets
18 hours ago
[-]
Since Nov it's been fluctuating between 5-7 ug/mg, with a few spikes hitting 9 and 10. At this moment its a 6.

I haven't checked the TDS but when I used a water test strip for an aquarium early this year it was in the hard water range.

reply
wintermutestwin
8 hours ago
[-]
Thanks to this post, I checked my ultrasonic filled with tap water. With it running all night in a bedroom with an open door, morning pm2.5 readings are ~30 and the meter is in the kitchen.
reply
kjkjadksj
4 hours ago
[-]
The ones I had for a bit basically fogged out the apartment immediately and left white (i’m guessing salt) deposits all over everything. I know you are supposed to use distilled but it’s cost prohibitive at the rate these blow through water unless you also have a home distillery.
reply
marcinpieczka
10 hours ago
[-]
Are pm2.5 particles a problem if they are water soluble? After entering the body they will just dissolve.
reply
piskov
1 hour ago
[-]
Depends whether you want that stuff in kidneys and what have you.
reply
esaym
23 hours ago
[-]
> Use evaporative humidifiers

You don't have to buy one either. A suspended wet towel with a fan blowing on it will work very well. If you want to get fancy, have the last inch or two of the towel sitting in a tray of water.

reply
abraae
6 hours ago
[-]
My brother's house in Perth, Australia has an antique air conditioning system in the roof space that works in exactly this principle. 4 blankets that wick up water and have air drawn through them and into the house by a fan. It's in disuse now but I understand they were common and quite effective in the day.
reply
loloquwowndueo
22 hours ago
[-]
But then I have to buy the towel and the fan, the tray, something to suspend the towel at the right height …
reply
dependency_2x
21 hours ago
[-]
This is cheaper, and the towel and fan can be repurposed. Just buy a 3D printer for making the the suspension part (/jest)
reply
jgalt212
10 hours ago
[-]
That's a permanent bachelor design aesthetic.
reply
Etheryte
8 hours ago
[-]
Turns out young me was on to something, I wasn't just leaving wet towels laying around, I was fixing the air humidity problem.
reply
jborichevskiy
1 day ago
[-]
Alec from Technology Connections also has a great video comparing humidifiers here

https://www.youtube.com/watch?v=oHeehYYgl28

reply
kccqzy
23 hours ago
[-]
Distilled water isn’t strictly necessary. I use mine with purified water with a reverse osmosis purifier. I periodically test the TDS of the water to confirm it is low. It’s fine.
reply
piskov
8 hours ago
[-]
I’ve run dyson ultrasonic humidifiers for a few years. The whole appartment was in white dust.

Especially evident on some black leather bags in the wardrobe.

reply
kccqzy
2 hours ago
[-]
You can buy a cheap tester for TDS (total dissolved solids) for <$20. Test the water before pouring it into any ultrasonic humidifier.
reply
piskov
1 hour ago
[-]
I just switched to evaporative ones like smartmi I’ve mentioned.
reply
nvch
22 hours ago
[-]
The best solution I've found a few years ago is one Venta LW 45 for every 30 m² of space. That's enough to run them on the lowest speed while maintaining acceptable humidity and CO₂ levels.

Higher speeds are too noisy. Smaller machines evaporate less.

For sub-zero outside temperatures, it's necessary to add at least 5 g of water to each cubic metre of air coming from outside.

The recommended ventilation rate of 30 m³/h per person requires to evaporate 4 liters of water per day.

reply
dalyons
9 hours ago
[-]
That’s a lot of refilling. You might want to look into a whole house humidifier, I added an aprilaire 700 evaporative to my hvac ducts, it costs a few hundred $. Plumbed in, automatic. So much less screwing around
reply
gessha
8 hours ago
[-]
What sort of maintenance do you have to do on it?
reply
dalyons
2 hours ago
[-]
Once a season / 6 mo of running you replace the evaporator panel, which is ~$50 for branded or $10 for amazon copies.
reply
noitpmeder
22 hours ago
[-]
Which venta are you referencing here?
reply
mercnet
22 hours ago
[-]
They are referencing https://www.venta-air.com/en_us/product/lw45-original-humidi...

I also have one and love how easy it is to clean.

reply
SauntSolaire
21 hours ago
[-]
For $500 it better clean itself
reply
InsideOutSanta
16 hours ago
[-]
I bought two Ventas well over a decade ago, and they still work as well as the day I bought them. They're an expensive initial investment, but IMO worth it over the long run.

They are also mechanically simple, so I trust that if they ever break, I will be able to repair them.

reply
neilv
1 day ago
[-]
That Smartmi model seems to have toxic IoT in it.

I'm currently using the Vornado EV100 non-IoT evaporative humidifier, and my only complaints are relatively minor, as humidifiers go (consumable wick, fan noise, insanely bright blue LEDs). https://www.vornado.com/shop/humidifiers/evaporative/ev100-e...

reply
piskov
23 hours ago
[-]
You don’t need to connect it — works completely offline.

Also no consumable parts there: just plastic disks which you clean with couple of spoons of citric acid dissolved in water from time to time.

reply
NewJazz
23 hours ago
[-]
Couldn't you just use vinegar?
reply
kjkjadksj
4 hours ago
[-]
Drying clothes indoors is also effective. When I set up my laundry rack rh can surge by 30%. I imagine setting up a tray of water under a ceiling fan might be similarly effective.
reply
dheera
19 hours ago
[-]
I found this too. I wonder why they don't just accept a PUR water filter on the input side.

I also wonder why mini-split heating systems drip and pool water outdoors instead of pumping that distilled water back indoors for humidification.

reply
throwawaysoxjje
10 hours ago
[-]
You’d need a reverse osmosis system, not just he charcoal filters like that

And the condensate water from an AC evaporator coil is not anything like distilled water, dust/bacteria are also in it.

reply
somat
18 hours ago
[-]
It's funny in an ironic way because the original purpose of air conditioners was to remove humidity from the air, the mechanism used was to cool the air down thus forcing some of the moisture out. The general public quickly caught on that having cool air was nice in it's own right and that is the main purpose these days. however the dehumidifying function is still sometimes used, people are surprised when their air conditioner turns on at the same time as the heater (why are they fighting each other?) but that is because the system is trying to remove moisture from the air before it is heated. Mainly seen in cars so the windows don't fog up.

Probably something wrong with me but I just find it humorous trying to add moisture to a system designed to remove it. Really a reasonable request however, depending on where you live the air can get quite dry.

reply
Ekaros
11 hours ago
[-]
During summer dehumidification is needed in most areas.

And it is really not distilled water. It gathers dust and so on from air. Distilling in more so closed circuit, where as those are very much open.

reply
ErroneousBosh
14 hours ago
[-]
> I also wonder why mini-split heating systems drip and pool water outdoors instead of pumping that distilled water back indoors for humidification.

Do you want Legionnaire's Disease? Because that's how you get Legionnaire's Disease.

https://en.wikipedia.org/wiki/1976_Philadelphia_Legionnaires...

reply
mytailorisrich
22 hours ago
[-]
How did we survive the last 3.5 billion years?
reply
gpm
22 hours ago
[-]
We didn't have access to modern technology... like ultrasonic speakers?

Also we died at a young age. Everyone dying at 40 isn't incompatible with the species surviving but it's what advice like that is usually trying to avoid (and even less extreme outcomes).

reply
SauntSolaire
22 hours ago
[-]
The concept of everyone dying at 40 is a myth/misunderstanding anyways - the reality was a lot more bimodal than that.
reply
gpm
21 hours ago
[-]
Eh, here it's more of a simplification than a myth as used in my comment. There are two effects:

1. We've reduced infant (and childhood) mortality. My comment isn't talking about this effect but it did drag down average life expectancy substantially. Including this effect life expectancy at birth in the stone age might have been as low as 20... but as you say the bimodality means this is a deceptive statistic when used this way.

2. We've made it so you on average live longer even if you survive childhood, my comment is really just about this part of the effect. It's still a simplification because saying "on average if you survive childhood you die at 40" isn't the same as "everyone dies at 40" but closer to "adults die at all ages in a reasonable smooth monotonic curve and 40 is about the average age they live to but some get lucky and live to 80 or whatever". But then "don't use ultrasonic dehumidifiers" is like this too, using one won't kill you at some specific age, it will just slightly increase your chance of death every year for the rest of your life however long that ends up being.

The number 40 was picked out of a hat, too. It should be right for some areas at some times just by coincidence though and since I was non-specific that makes me right ;)

reply
somenameforme
11 hours ago
[-]
The age 40 includes childhood mortality! It's difficult to get records from prehistoric humans for obvious reasons, but as early as Ancient Greece you had the upper class living about as long as we do now a days. A study of men of the time found the average life expectancy to be 71.3 years. [1]

And while the Bible includes plentiful mythological components, it also includes many historical and contemporary accounts. And this verse is certainly of the latter: "The length of our days is generally seventy years, or eighty years if one is strong, yet even the best of these years are filled with toil and sorrow, for they pass quickly and we fly away." That is part of the Old Testament (Psalms 90:10) that is believed to have been written somewhere from 1400-1200BC.

If you want more contemporary stuff that's completely indisputable you can also take random selections of people of renown. For instance the main Founding Fathers are a great example because they all were relatively young when their names become inexorably etched into history, yet their final life expectancy is again well into the 70s. The youngest major founding father to die was Hamilton, in a duel - at 49. Then Hancock died at 56 - likely of gout which can be caused by things like excessive indulgence. Next up was Washington who died at 67, probably more of the cure than the disease - he was leeched to the point of being pale as a ghost on his death bed. Jefferson, Franklin, Madison, Sam Adams, John Jay all lived to their 80s. John Adams made it to his 90s.

---

I am not trying to claim these samples are representative. These were wealthy individuals who would be relatively immune to famine, war, and other such factors that could have a catastrophic effect on lower classes. But when speaking of life expectancy, I think we are implicitly asking the question 'how long could somebody reasonably expect to live xxxx years ago without access to modern medicine and technology.' And that's what this sampling of people answers.

[1] - https://pubmed.ncbi.nlm.nih.gov/18359748/

reply
Cpoll
7 hours ago
[-]
> The length of our days is generally seventy years, or eighty years if one is strong

Which translation is that? Or are you paraphrasing?

Most translations don't include "generally," and therefore read more as an upper bound "if one is strong" than an average.

reply
somenameforme
6 hours ago
[-]
Just a poor memory translation. Yeah, generally is incorrect - though I think the correct phrasing also implies an average age of natural death, rather than an upper bound. There were certainly plenty of people living past 80. In the aforementioned study of Ancient Greeks, there were at least 3 centurions - Aristarchos, Democritos, and Gorgias. Granted 1400BC is a thousand years yet prior to that already ancient time, but life peaks seem to be relatively unmoving for humans, and so I don't see any major reason to think there would have been a major difference between 400BC and 1400BC.
reply
ErroneousBosh
14 hours ago
[-]
So up until about two or three years ago when everyone suddenly became terrified of "particulates", people died at 40?

The Victorians called, they want their Night Air Panic back.

reply
wpm
19 hours ago
[-]
Quite poorly in fact
reply
HappyJoy
22 hours ago
[-]
A lot of us didn't
reply
SauntSolaire
22 hours ago
[-]
Glad to know that's solved now
reply
dangus
22 hours ago
[-]
This is pretty crappy one-size-fits-all advice in itself.

If you’re willing to use distilled water, ultrasonic humidifiers have their own advantages over evaporative.

I’m personally willing to buy distilled water. It’s a dollar per gallon, and we only need the humidifier during a short few months. You can even buy a small countertop water distiller for under $60.

reply
butvacuum
21 hours ago
[-]
I'm thoroughly unconvinced.

Doing some basic research... hard water is overwhelmingly various carbonate and bicarbonates of magnesium, calcium, sulfur, iron, maganese, and aluminum. All of which are essential nutrients and readily soluable in water.

The other proposed problem was pathogen aerosols- however I was unable to access anything but an abstract. So, I don't know if they survived being aerosolized, produced more and/or worse pathogen count than evaporative humidifiers, Nor the size of the pathogens.

It seems to me the known risk is mostly mechanical (Asthma, exacerbated COPD, etc) and nonpersistent (particles dissolve and are used or excreted via the same pathways as when consumed). With an unknown risk on the pathogen side.

reply
bsder
21 hours ago
[-]
> If you’re willing to use distilled water, ultrasonic humidifiers have their own advantages over evaporative.

Unless you are anally retentive about cleaning it, ultrasonic humidifiers vaporize microbes into the air. There have been loads of studies about this.

The only real way to avoid this is to use the humidifiers that are boiling the water.

reply
dangus
3 hours ago
[-]
The real pro tip is to get a whole home humidifier. They aren’t even that expensive. But I haven’t taken the plunge yet.

They avoid all these issues and require way less maintenance than any other solution.

reply
skeledrew
3 hours ago
[-]
This got me thinking about how my AC is somewhat ticking me off. A couple years ago I bought a smart AC, and when I got around to wanting to use the smart feature (via the app), I learned that I need to create a Tuya account and connect via that. Today I'm still manually pushing the buttons, as I wasn't having any of that, and the community Tuya tools I found out there are dependent on an account.

A couple weeks ago I took a preliminary look jailbreaking it. Main thing holding me back is a fear of bricking it and being left with an expensive, oversized paperweight, as the electricity here tends to chip at random times and could do so just at a critical point of the process. It also bugs me that I can find 0 information about the device; it's like the "Bluesonik" brand doesn't have an internet presence. But perhaps one day I'll just throw caution to the wind and attempt a Tasmota flash (without even knowing if the board is supported) and hope for the best, similar to when I rooted and flashed my first Android phone for the first time 15 years ago.

reply
hs586
15 hours ago
[-]
Tangential rant: It’s becoming hard to buy dumb appliances.

I was looking at robot vacuums, and most need internet connection at least for setup - by which point it’s already uploaded your floor plan and who knows what to the cloud.

reply
gloxkiqcza
14 hours ago
[-]
https://valetudo.cloud/

The project was recently discussed on HN as well. It has its issues but it works.

reply
einsteinx2
13 hours ago
[-]
Not sure that’s a great example when you can easily buy a regular vacuum. Robot vacuums are sort of by definition already the “smart appliance” version of the “dumb appliance regular vacuum”.
reply
hs586
8 hours ago
[-]
I understand what you mean, but I disagree. The technology allows for robot vacuums to exist and I don’t see the cloud connection as a mandatory need for it. Similarly, I want my car to have say lane assist but we don’t expect to have cloud connection with it.
reply
roger_
11 hours ago
[-]
Pretty shocked that Xiaomi publishes the protocol: https://iot.mi.com/new/doc/accesses/direct-access/embedded-d...
reply
rokoss21
12 hours ago
[-]
Great project! This resonates with me - been using ESPHome for a year now and it's solid. One tip: if you're concerned about reliability, pair it with a PoE switch for your ESP devices. Makes recovery much easier if something goes wrong.

Also curious about your power consumption - did you measure watts before/after switching from Xiaomi's cloud solution?

reply
N_Lens
21 hours ago
[-]
A humidifier needs network capability incase someone discovers a new version of water, or for the manufacturer to be able to patch remote exploits.

https://xkcd.com/3109/

reply
alephnerd
21 hours ago
[-]
It's becuase Xiaomi integrated it just like all of it's other smart home products with Mijia - Xiaomi's smart home mesh [0].

In Asia (but arguably the same in the West given the proliferation of Ring and smart home hubs), consumers have less of an aversion to smart home and connected products in general.

Keeping IoT devices on a separate segmented network with strict DMZing, turning off unused features, and not sharing passwords would provide enough security for most home users. I recommend reading James Micken's essay "The World is Ours" [1] on the diminishing returns of certain security features at the expense of user experience. I also agree with it as someone who used to do edgy stuff with SHODAN as a teenager.

HNers tend to be the minority amongst consumers, which is assuming the opposite of the HN herd mentality tends to be a fairly successful strategy.

[0] - https://www.mi.com/global/smart-home/

[1] - https://www.usenix.org/system/files/1401_08-12_mickens.pdf

reply
techsystems
19 hours ago
[-]
Hah! How exact
reply
nrhrjrjrjtntbt
19 hours ago
[-]
Xkcd for everything as always
reply
DocTomoe
18 hours ago
[-]
I'm all for KISS.

But in a rare instance, xkcd is missing the point here. People do not live in their rooms 24/7, but they do want to be able to, e.g., turn stuff on or off remotely, or based on environmental conditions (turn on/off based on outside sensors or the current electricity price...) or to get status alerts ("tank empty, refill").

Now, I do that via Home Assistant and keep anything "smart" on a highly-restricted vnet ... but not everyone is a geek. While the standard implementation (some cloud service) comes with a bouquet of problems, it basically acts as a simplified Home Assistant, and ultimately as a necessary crutch. Preferably we'd be in IPv6-land, where ISPs would not NAT everything to death and we could talk to our devices remotely without an intermediary ... but well ... it cannot be helped.

"You're not going to need it" and "In my time, we just flipped a dumb switch" is paternalistic hogwash, not clever social commentary. Back in my days, we also didn't need satnav (just read a paper map), or cell phones (write them a note, leave it on the fridge, nothing is so important to demand imminence), or dishwashers (just do your dishes by hand)

reply
mrweasel
14 hours ago
[-]
Obviously all these smart appliance are about remote management, but I have to question how much usage it's getting in real life. My parents got a few smart devices for their holiday home, as my dad didn't want to drive 45 minutes both ways to check up on things during the winter. I think he probably ended up spending more time managing the IoT stuff that he ever did driving.

It create to have the option to manage something remote, but when remote become the only option, the usability takes a dive. When I have to go find my phone, unlock it, find the app, possible update the app, find the right setting or menu, stare at "Failure to connect to device", and whatever else might go wrong, it's quicker and easier to just manage the device directly. We got rid of our robot vacuum clear, because it's literally quicker and better to go get our 20 year old regular vacuum, and the floor is done in 3 minutes, not the 20+ the Roomba needs (and I needed to clear the room for it). When we used the Roomba, 99% of the time I pushed the "Start" button on the device, because it's way quicker than using the app.

There's a place for smart devices, but they need to be much better and have local controls.

reply
somehnguy
9 hours ago
[-]
I still think the value prop is dubious for a device like this.

> turn stuff on or off remotely

Why? Nearly all modern humidifiers have a sensor to measure humidity and will cycle on and off based on the setpoint. Getting to the setpoint also takes time so I don't see any reason someone would want to turn it on and off based on presence.

> (turn on/off based on outside sensors or the current electricity price...)

Not sure why the outside sensors would matter, it's concerned with the inside humidity which again it has a sensor to read. The amount of electricity these take to run isn't worth even mentioning.

> get status alerts ("tank empty, refill")

So you can refill it remotely? You have to be present to fill it anyway - just look at the thing and you'll know its water level

I say all this as someone who also run Home Assistant and automates various things.

reply
airstrike
20 hours ago
[-]
Can you do HP printers next
reply
CoastalCoder
19 hours ago
[-]
Yup! Step 1: fill your printer with two liters of distilled water.
reply
kjkjadksj
3 hours ago
[-]
I have an hp envy 5030. It has done nothing dystopian so far. Just acts as a dumb wlan printer it seems.
reply
aeve890
11 hours ago
[-]
A whole ass esp32 module in the board? Never seen something like that. I mean I've seen esp32 iot devices but with chips directly in the board, not as a separated module. It looks like hobbyist job.
reply