你的空气净化器有自动模式吗?数据警告别用。

我们生活里的很多东西都越来越智能化,从手机到汽车,连空气净化器也加入了智能化大军。不过实验数据却显示智能空气净化器错误频出,经常把用户置于危险的空气中。

 

 

净化器自动模式的理想

 

空气净化器自动模式想法很简单:只有在空气不好的时候运转。这么做的好处显而易见,可以节约能源,减少噪音。

 

 

自动模式实地测试

Smart Air的共同创始人Anna测了两款流行空气净化器的自动模式:小米2和飞利浦AC4072 。

 

 

测试方法就像大多数人怎么用。在北京一个普通的15平米的卧室,Anna用13天时间每天睡觉或者白天把小米或者飞利浦净化器调到自动模式开了8-13个小时。

 

 

她用了Dylos Pro镭射颗粒物测试仪来追踪测试0.5微米和2.5微米的颗粒物。(Dylos跟官方PM2.5数据的相关性很高123)。为了算效果,我比较了开机前和醒来以前的四个小时里的平均颗粒物数据,比如12月7日的测试:

结果不健康

下面的照片是小米测试刚开始和结束的空气质量。虽然小米在密闭的房间内运转了整个晚上,空气还是在不安全的范围内。

 

 

平均来说,小米减少62%的0.5微米颗粒,飞利浦59%的0.5微米颗粒。

 

 

对比来看,这组数据还是比只单纯用个带子把过滤网绑在风扇上效果差了20%。

 

 

那机器有问题吧?

不过,会不会只是飞利浦或者小米的净化器不够好?也许他们用了不合格的过滤网,也许他们的风扇不够强力?

 

我们其他的测试表明情况并非如此。在之前的高转速测试中,飞利浦净化器的得分一直名列前茅。

 

 

小米2在高转速测试中也表现很好,但不能维持在高档的状态。真的不能维持在高档的状态吗?3个小时以后,无论用户选择与否,机器都会自己转到自动模式。

 

 

自动模式真的不安全吗?

存疑的读者可能会说:“颗粒物减少60%也并不坏啊。这还不够好吗?”

 

60%肯定比0%好,可是如果看空气污染对身体健康的影响的研究,这些水平明确对人有害。室内空气平均达到世卫组织公布的24小时PM2.5上限两倍 (这个上限还是世卫组织的宽松上限!)。

 

 

为什么自动模式这么烂?

高速运转测试表明这不是风扇或者滤网的问题。那怎么解释自动模式的问题呢?

 

原因1:内置空气质量测试仪不准的离谱

一个原因是自动模式是由内置空气质量测试仪控制的,而且测试发现小米测的跟实际PM2.5差异达到218毫克!绿色AQI和黄色AQI之间的差异才12微克

 

 

所以这些机器有时候会在空气“有毒”的时候判定空气是安全的。

 

原因2:小米对“安全空气”的定义太宽松

也有可能这些净化器只是对“安全空气”的定义过于宽松。中国设定的PM2.5上限是35微克,世卫组织的年平均上限是10微克。然而研究表明即便是20微克的差距也会对健康造成严重的影响

 

自动档须知

自动模式经常会让室内空气停留在一个远超过健康空气底线的状态。我不推荐使用自动模式 (而且因为Mi2过三个小时就强迫用户开自动档,除非改比较复杂的地理围栏设置,我不建议用Mi2。以前的Mi1没有这个问题。)

 

公开数据

跟每次一样,我把原始数据和更多测试方法细节贴出来和书呆子们分享。

芝加哥大学行为科学系的助理教授; Smart Air创始书呆子。

在 “你的空气净化器有自动模式吗?数据警告别用。” 上有 7 条评论

  1. I have commented on your other posts as well.

    1) dont rely on the auto mode. just set it to “favorite” mode at a reasonable level which isnt too loud.

    2) to be clear, favorite mode allows you to set the speed exactly. the max speed in favorite mode is 34-37m2. However, if you go into the settings menu, you can enable turbo mode within favorite which allows you to go all the way up to 42-45m2.

    3) for me, even the normal max speed (without turbo) is too loud. This is just a function of any purifier which is monstrously huge. the sweet spot is 17-20m2.

    4) smartAIR would do its readers a lot of service to analyze xiaomi’s 2’s performance at 17-20m2 setting looking at both pm2.5, below .5 and also the decibel level at this speed.

    5) i repeat (from the comment on your other post) in my experience, the xiaomi does NOT revert back to auto after 3 hours. it has maintained the speed set in favorite for as long as it receives power. dont know why smart air isn’t acknowledging this.

    6) lastly, the DIY and Cannon options that Smart Air sells are extremely ugly and not smart at all. i can’t even stress how amazing it is to control the xiaomi’s via iOS app. the Blast Air’s, which aren’t as hideos as the Cannon DONT SHIP OUTSIDE OF CHINA! So what option is SmartAir leaving us? Dont understand how SmartAir PHDs are being so biased, as their posts otherwise come accross as objective.

    1. Hi Amin, I’m happy to post any data you’ve got on whether the Xiaomi can stay on high past three hours. One way to do it is to use a decibel meter for your phone. The decibels will correspond to the fan setting, as long as the room is otherwise quiet.

      We’ve repeatedly tested the Xiaomi using every setting we can find to see if it can remain on high past three hours and continue to find that it reverts back to auto mode. We’ve also asked Xiaomi customer service on multiple occasions, and they say the same thing. We’ve saved the chat records and posted them here: https://smartairfilters.com/en/blog/xiaomi-auto-mode-leaves-air-unsafe-for-86-of-hours-supplemental-data/

      Objectivity is very important to me. That’s why (1) I’ve published all the original data and chat records and (2) remain open to any new data documenting whether the Xiaomi can remain on high.

      And you’re absolutely right that the DIYs are not pretty! They’re simply filters strapped to fans. When I “designed” them, my goal was to get clean air, not design something that looks beautiful. If the looks of the machine are important to you, there are plenty of more finished looking machines out there.

      Sorry we can’t ship the Cannon or the Blasts to Pakistan! It’s not our choice. Because they’re heavy, the shipping is prohibitively expensive. Beyond that, China Post has regulations against shipping motors internationally. Go figure!

      Regardless, I hope you can get some clean air where you are! One option is that we can ship HEPAs to you, and you can source a fan locally. I recommend that to people who ask about shipping to the US.

  2. The new Xiaomi 2S (alongside the Pro and Max) use laser sensors instead of infrared to measure air quality. Would you mind testing the Auto Mode of one of those models?

    I expect their measurements to be almost identical to those of your Dylos Pro.
    So you could rule out Cause 1 and see of Cause 2 stands.

    Best Regards

    1. Hi Ben, thanks for the suggestion! I’ll have to run some more tests. However, my particle counter tests did include the 2 Pro. I’ve heard some people claim that Xiaomi says that has a different particle counter in it (although I can’t find it on their website). Regardless, it was just as inaccurate as the regular Mi2.

  3. Contrary to what Amin says, I’m told that even the Xiaomi 2s goes to auto mode after 3 hours. However, there is a fix! You have to use the smartphone app to turn it on to the high setting every 2.5 hours. For example, you set it to high at 9 a.m. and again at 11:30 a.m. and so on. Then it never goes to auto mode.

    1. Good question! The truth is that no room we live in is truly sealed. Dirty air from outside is constantly leaking back into your room through the door, windows etc. So as soon as you slow down the speed of your purifier, if it’s not strong enough to filter the dirty air leaking into your room, then the particle count will increase. Take a look at the data in this post which shows how quickly air pollution increases after switching off the Cannon in a Beijing flat: https://smartairfilters.com/en/blog/do-you-have-to-run-your-purifier-all-day/

发表评论

电子邮件地址不会被公开。 必填项已用*标注