3DPrinting @lemmy.world Krauerking @lemy.lol 1w ago

Sigh, why do people use these hygrometers?

These have to be the least accurate things I have ever seen.

The rectangular one is accurate or accurate enough and has been what I used but I noticed files all had cutouts for these round hygrometers...

Well from my 6 pack 1 is within a margin of error to even be useful.

I get they aren't expensive but seems like a waste of money for this bad.

49 comments

As an old toolmaker who might have dabbled in accuracy. I just shake my head when people complain about things like this because they think absolute accuracy --an impossibility-- is the important thing, (you can't afford anything close absolute accuracy), when it's repeatability that matters the most. Choose the one that repeats the best, toss the rest. Then learn what the "magic number" is that makes you happy to read and need to get the results you are looking for. Learn to apply some bloody "windage".

Remember: Rick Sanchez is a dumb-ass. And you only need to be 5% smarter than the tool you are using to be successful. So be smarter than the tool and understand the process.
Yeah I looked at a pack of these on amazon recently and had a good laugh at the tolerance they provided for one of the measurements.
Do you have a recommendation for a more accurate unit?
- Working on it. Waiting for some packages.
  
  The rectangular ones are slightly more reliable from personal experience but likely just as possible to be low quality. From working with them the design layout leaves less space to get obvious flaws such as the RH sensor being installed facing the board which seems to be true for most of the round ones, they also have a separate and dedicated temp sensor unlike the round ones and use a double battery setup since voltage can also affect the accuracy of the reading.
  
  But the RH sensor between the units seem to be identical as is the inability to calibrate.
  
  Personally the paper strips are very reliable if just looking for a binary answer of if it's gotten past a threshold and are incredibly cheap.
  
  Abe recommended an electronic daughter board sensor that was cheap but requires skills I don't have.
  
  And I have found medical grade ones that can be purchased cheap "used" as they are often single use for shipping purposes to make sure the item was kept in specific conditions. But that brings me back to waiting for the. To arrive and test.
  
  TL;DR
  
  These cheap electric square or circle ones are both luck based with some preference to the square. But cheap paper humidity sensors can be gotten if you just want to toss them in a bag to make sure they haven't gotten water damaged as a binary identifier.
  
  For reliable accuracy look elsewhere.
I used 10 rectangular one for my dry boxes because they are cheap and its a rectangular one that was supplied with the Prusa enclosure.

It's good to know that they seem to be more accurate than the round one.
Mine work somewhat okayish, which is within the margin of error I need them for. I think there was one that was terrible though.

Mostly I use them for the temperature aspect, mostly for reminding me if it's too hot or cold in my room (because due to my autism, I often don't notice whether I'm at a comfortable temperature). I have a few scattered about my room and basically they act as a visual prompt to consciously ask myself if I'm at a comfortable temperature, and to act as a rough backup to whatever I'm feeling (because even when I'm consciously aware that my temperature is feeling Bad, I can't reliably tell whether I'm too hot or cold, so these terrible thermometers at least help me answer "should I get a blanket, or open a window?"
I would not be at all surprised to learn that both the commodity round and rectangular units use the exact same components inside. These surely must be made from jellybean off the shelf parts, and at similar price points to each other neither could possibly be assembled with much care or attention to detail.

I have a six pack of the rectangular ones from Jeff Bezos' Knockoff Whitebox Emporium. Even all sealed in the same container with each other, they all disagree by a spread of about 15%. I have no idea which of the six, if any, are actually producing an accurate number.
- So, just because my rectangular one has been with me for a few years I'm gonna stick up for it a little. It does have 2 sensors in it separated for temp and humidity compared to the circular ones that seem to have the temp sensor baked into the board if it exists at all and the humidity sensor was poorly soldered upside down. But agreed on likely carelessness of assembly from identical parts.
  
  Same batteries and sensor modules it seems and even the press fit nature of them. And the only reason I know that specific one I'm reading against is accurate is because I only kept it because it was the accurate one from a pack of them I grabbed from a lab in college.
  
  I just don't get why everyone uses them is my point if we are in agreement they kinda suck.
because if it says anything other than 10% (the lowest thing it reads), then the dessicant needs to be refreshed.

it's more of a binary output rather than trying to look at 53 vs 55%
- You get your humidity down to less than 10%? Or still literally treating it like a binary thing and it's just reading way under real?
  
  I am at about 25% relative humidity and it's showing as a 1 in the ams sensor so 10% seems impressive even though I'm not using much desiccant.
  
  I've pretty much been running on the binary theory with the AMS sensor: anything but a 1 is too high. I guess I'm glad I didn't waste any money on those digital jobbies. I wonder if the old school analog style are better?
  
  yeah, it's usually 60% ambient humidity here, and then in the dry box it'll read <15%. so that's a pretty decent indicator to me that it's working fine. I don't really care if it's 10, 15, or even your 25%, those are all way less than the ambient baseline and let me know that the dessicant is working.
  
  If you've got some need to the humidity accuracy then that's another thing, but for me that's why I use those cheapos.
  
  I use modified cereal containers with dessicant on the bottom and have a mount modelled up for those sensors.
@Krauerking@lemy.lol Hygrometers are only as good as their components. Buying a DHT11/22 or SHT31 from AliExpress ($1-2) alongside an ESP8266/32 and you'd have much better results than buying these "are my cigars dry" pucks.
- I've had massive differences in readouts in DHT22 from aliexpress. They are really not good sometimes
  
  @callcc@lemmy.world SHT31/41 are better than DHT22 tbf. DHT would have a variance of about 2-5%. It also takes a while for it to stabilize.
- I'd really love to know how, once having purchased and aquired these two parts, how to join them, and get them powered, and to display. this is like secret engineering knowledge i'd love to be walked through
  
  @shoulderoforion@fedia.io I have about $150 worth of Ali parts and components coming just this month for whole house monitoring for this kinda thing - temp, humidity, CO2, VOCs, pressure, light sensors etc. Would be glad to ping y'all once that writeup is done :)
  
  LOL same. And this is why I bought bad tech. It's a whole wild world if you can actually take electronic components and just wire them together and program them yourself.
- I don't even think these would helpfully let you know if your cigars are dry. 2 of them barely changed humidity over time and I noticed the sensors were flush with the board instead of exposed.
  
  I ended up buying inkbird ITH10s because I generally don't go completely self made since I'm not overly tech crafty and more work with my hands crafty.
Is the one built into the AMS not good enough? I get that it only gives a level between 1 and 4 but it's been fine for me with PLA and PETG 🤷‍♂️
- "It's on 3 now, I should do something."
  
  Does something
  
  Waits two hours
  
  AMS says 2
  
  "Is it at 2.0 and dropping to 1 any moment now, or is it stabilized at 2.9? Or..."
  
  I guess it's not important, but I really would like to have some decimals. It's not that I care about accuracy, but I would like to be able to see which direction it's going.
- I am doing some testing with a new desiccant and wanted to have some extra data for humidity. So it is honestly fine for function but not what I was going for.
  
  Interesting. Well, when you get some numbers together, I'd be interested to see what you find out
@Krauerking you could look at calibrating the hygrometers or at least the offset.
https://www.stevejenkins.com/blog/2014/06/how-to-calibrate-a-hygrometer-humidity-sensor-using-the-salt-test/
Two point calibration with calcium chloride and sodium chloride might be even more accurate
- Nah these do not have a calibration method.
  
  Cheap and seemingly what the average consumer uses based on the 3d files all using them but not calibratable and wayyyyy off.
  
  @Krauerking ah but you can measure the offset. A two point calibration should cover your measurement range. Not ideal but you could get a reading.
Mine are pretty consistent, max. 5 percent points difference between them. Cheap brand from a big internet warehouse.
- Lucky. Pretty sure these are the same from the same internet source but the swing on them is insane. They all showed as 75% in the bags but spending some time with them I realized the complete lack of accuracy when I realized my humidity wasn't what it was reading.
  
  I have since taken some apart and cleaned them out but now those are just under the humidity by the same percent they were over before.
  
  The analogue one I own (round body) can be tuned/adjusted by a small screw on the back. Maybe that's an option?
Figures, I just bought two for the AMS lite.
- Best of luck. Seems they are kinda luck of the draw.
  
  They can I guess let you know trend of humidity go up or down but also do note that I noticed the sensor was installed upside down so readings also were pretty delayed from real world adjustments.
How has your experience been with the inkbird ITH10?

49 comments