Temperature/Humidity Data Loggers

Knowing the temperature certainly rates among the nice to have, but not a necessity. When I first really took off with the cat and the trailer (winter solstice 2018), I didn't have any means to measure it. Wondering what the cat was subjecting himself to while I tried to make him understand that under the blankets may be restrictive, but it's warm, got me wanting a way. Knowing the real temperature, not just if it feels a bit chill, helps a little for planning. It would be really nice to know the outside temperature without having to experience it first. Somewhere around Lake Havasu (mid January 2019), I stopped into an Ace and picked up a Taylor indoor/outdoor thermometer. One month. That didn't take long.

Much later, I picked up a cheap hygrometer (which is technically a humidity reader, but it needs to know the temperature to get relative humidity, or at least they did historically, so they do both) to tuck away in my backpack too. I didn't have the same feeling of need for that. Maybe because you've just go to be out in it whatever it is? Maybe because a backpacking trip is generally a week or less and a weather window can be selected? I'm pretty good at forgetting it when I backpack.

Temperature monitors. Indoor/outdoor thermometer set on the left, cheap hygrometer for cigar boxes or whatever on the right.

The indoor/outdoor temperature monitor also keeps track of the minimum and maximum temperature it has seen since it was last reset while I've just got to catch the hygrometer saying whatever it will. Once you're into knowing the temperature, the extremes are quickly of interest too. Maybe that's another reason the hygrometer only gets packed half the time. It won't tell me the important thing (how cold was it last night?) anyway. It's certainly not the 18g weight I'm worried about.

You may notice the indoor/outdoor thermometer isn't actually reading an outdoor temperature at the moment. That would be because on my way down from Blue Lakes (early July 2022), while struggling down a road that had been, on the way up, one of the most beautiful dirt county roads it has been my pleasure to drive. That was before a particularly powerful storm. I must have managed to splash the outdoor monitor particularly badly. It lived tucked under a front corner on the trailer, velcroed to the top of a brace. It's actually a very protected spot from splashing and had gone through driving in some heavy rain perfectly well before. This was after the rain. Admittedly, the velcro was so I could put it inside for driving, but after a few times forgetting and finding it didn't matter, I stopped bothering. And then it finally mattered. It worked in the morning when I started. It didn't work in the evening when I found a new camp. New batteries didn't help.

So I tried to get a new outdoor monitor. I don't want to buy it all again, just the bit that's broken. Taylor don't sell just that bit. Sometimes I looked at full replacements. Not the same one. This indoor thermometer is out of range at freezing temperatures. It's very annoying to know the temperature outside is 9°F and the cat's water bowl has gone solid (here, have some of mine, there seems to be a little liquid in it still) but get told the indoor temperature is "LL". I'm pretty certain they're using the same thermocouple in both. This is a programming decision. Unfortunately, it's one that has been made on all the indoor/outdoor monitors I could find. When I most want to see how cold it is outside the covers, it won't tell me. I'm not buying another like that.

So after occasional rounds of shopping and not buying for over a year, a user on mastodon mentions his new toy: a bluetooth data logger hygrometer/thermometer combo. Now, there's a thing. The one he's got, tempi.fi mini, uses coin batteries (light, but so annoying to find more when needed), you have to turn on the bluetooth to read it, and it costs $40. (And claims 28g, in case you are wanting to backpack with it.) To me, it looked a lot like a replacement for that outdoor thermometer part. Not that I wanted to spend $40 replacing part of a thing that originally cost $16, nor have to deal with coin batteries.

Fortunately, it's not the only one out there. After a search, I settled on a SwitchBot Indoor/Outdoor Thermo-Hygrometer to be my outdoor sensor, replacing the dead piece from Taylor. I also got a SwitchBot Meter, which has a display, but is a little less waterproof, to replace the still working but unsatisfactory piece from Taylor. Both of these add humidity and logging (for 43 days on the device) down to -20°C (-4°F). I got them as a bundle for $20 while they are normally $15 each. The only down side is that they are no longer an integrated package so the display only shows the indoor reading. Both of these run on two AAA batteries. A couple weeks later, I gave in to the desire to also buy a ThermoPro TP357, which is a somewhat lighter package and only requires 1 AAA battery, for backpacking. It's only $10 and won't tell me how long it stores data onboard, but indications are 30 days. It claims to bottom out at -30°C! The lightest (including the alkaline AAA batteries each needs) is actually the more waterproof outdoor SwitchBot with no display at 44g, followed by the ThermoPro at 46g, then the indoor SwitchBot at 74g. The ThermoPro could be lightened by removing the magnet and stand with no compromise to waterproofness. More lightening may be possible.

The data loggers. On the left, the SwitchBot indoor (top) and outdoor (bottom), nicely unbranded. On the right, the ThermoPro advertising itself and giving its opinion on how I should feel about the temperature/humidity.

Shortly after getting sensors from these two brands, I had a look at a comparison video of these sorts of things. I do so hate watching YouTube videos for such things. Here's what I got while grumping along trying to watch it: 1) SwitchBot and ThermoPro actually make some pretty solid sensors and might be the best quality for dollars. No others look interesting from a budget buyer standpoint. 2) Sensors using coin batteries have a tendency to cut out in freezing temperatures. This is odd since the coin batteries in my hygrometer have gone to the low teens (F, -10°C range) without even dimming the display, but it doesn't have to write to memory too. The temp.fi that inspired the purchase has been used for hours below freezing bottoming out at 19.5°F at least once. So I did that in the wrong order, but it makes me happy I picked these. Also, he should have left his samples in the freezer a little longer.
And what do I think after using them some? They seem at least as accurate and responsive as the old device and...

The SwitchBot sensors are part of suite of automation gizmos and the app for it is huge because it can interface with all of the possible things you might buy from them. It still loads up pretty quickly and I've got it set so that what I see when it starts is temperature and humidity readings from the two hygrometers. The current temperature and relative humidity is broadcast on the bluetooth, so that is quickly seen. To get the logged data, one must connect to the paired device, which takes a bit longer. Once connected, it downloads data at an excruciatingly slow pace. It's about 30 seconds per day of data. I have connected up instruments with 9600 baud as their fastest data transfer and it isn't this slow! Even if they were asking for each point individually and the bad latency of bluetooth was slowing it down, surely it wouldn't be this slow? My best explanation is they do it to encourage people to buy the hub. The hub won't get me my two temperatures on a display together, so I have no desire to own it. Once data is downloaded, exporting it is pretty easy. One selects the range and "shares" it to wherever one wants it. It's non-intuitive, but similar to exporting tracks on Peakbagger. I share it to my file manager (Total Commander) and it prompts for a name and saves. No internet required. All of those humidity measurements seen in the screenshot below are included in different columns. The device saves 43 days of data and the app will keep 2 years of data.

Screenshots from the SwitchBot app in use. First is as it opens, second is after tapping on the "outside" sensor. It calculates and displays the humidity in a number of different ways and shows plots over time for temperature and any of these humidity measurements. Third is scrolled down for more of the plot, in this case over the last week. It is currently inside the shut up trailer.

The ThermoPro has a small app specifically for thermometers, because that's what they do, so no big footprint in the primary memory on my handheld (phone) which is getting quite tight. It also reads a broadcast for the current measurements and has to take a moment to connect to download the data. Once connected, it'll download a week of data in a second, making the SwitchBot download even more maddening. The battery it came with only lasted a month and it seems to have lost all the data since I had last downloaded, not just from when the battery ran out. The battery I provided is still reading full after another month. For data export, it asks for an email to send the file to, thus requiring internet access where it shouldn't be needed. The export only includes date and time, temperature, and relative humidity. The email has an ad tacked on because of course it does. It is not consistent about saying how long data is kept. It might be a month onboard. The manual says a year in the app but he app say 2 years.

Screenshots from the ThermoPro app in use. First is as it opens showing broadcast data. Second is after tapping on the sensor to open the historic data and start downloading the most recent data. It only displays humidity as relative.

I took the ThermoPro with me on my recent backpacking trip to Humboldt Redwoods State Park. Initially, I dropped it into the mesh back pocket, so it had plenty of exposure to the ambient air, but also could end up with a little direct sunlight. At least it's a light colored package so it doesn't heat up quite so quickly when that happens! During the day and in the evening, I did look at it a couple times. When I went to bed, it got tucked away in my lunch pocket, which is quite a bit more sheltered from the whims of ambient breezes. It ended up staying there for the rest of the trip and I turned on bluetooth if I was curious. Each night was distinctly colder than the last leading up to a very wet weekend starting the next day. The end result is I can look at the temperature (and humidity) in excruciating detail (down to a point per minute).

The temperature (red) and relative humidity (blue) plotted over the three days backpacking, plus the extra from and to midnight.

So now I can see that it really was a little warm that first day. The second day was very comfortable, although slow to get there with the fog. It really did come in cooler for the second night. The third day took a while to warm to barely comfortable. There's a spike for when it was close to a foot vent during the hour long car ride back. I think that spike over 80°F on the first day was probably some direct sun. It looks like the morning climb on the second day cooled off the air faster than the morning warmed. Or it's just the changing face and dampness. There certainly is a climb in dampness with the fall in temperature on that early morning portion of the second day. It's interesting to see how one night was quite steady and the other had a general drop in temperature.

So there's some fun data to have. We'll see if I can remember this gadget in the future, which will probably be a good indication about if I liked having the data or not. Feeling good about it right now, but I did forget to put it into the write up. I'm already including too much in the write ups.

I forgot a bit of bird song, too. Merlin says it's Pacific wren and there's a little bit of chestnut backed chickadee scattered in there. They were singing on my way down Grasshopper Trail on the third day. (Other times too.)



A note about the lower temperature limits of these: I found inconsistent claims from both ranging from -40°C (-40°F) to -20°C (-4°F). The maker of the sensors for the SwitchBot actually states the lower. The problem here may be solved by the Taylor manual. In it, it stated that the lower temperature for the outdoor sensor was actually dependent on the batteries. Alkaline batteries will freeze out around -28°C (if memory serves) but lithium batteries will take you lower, more like -40°C. They would also last a disappointingly short time in those temperatures. I don't expect to run the Iditarod, so alkaline batteries should serve me well. No idea how the rechargeables do.




©2024 Valerie Norton
Written 13 May 2024