WEBVTT Kind: captions Language: en 00:00:00.160 --> 00:00:03.236 This slowly rotating disc of... stuff 00:00:03.236 --> 00:00:07.055 may not look like anything but a slowly rotating disc of stuff. 00:00:07.055 --> 00:00:11.246 But would you believe it's actually part of a dehumidifier? 00:00:11.673 --> 00:00:14.010 Well, probably. You saw the title. 00:00:14.010 --> 00:00:19.346 Not long ago, I made a video about what I'd call ordinary dehumidifiers. 00:00:19.346 --> 00:00:22.671 You can check it out through that card if you like, but long story short: 00:00:22.671 --> 00:00:28.714 most dehumidifiers work by making something cold and blowing air past the cold thing 00:00:28.714 --> 00:00:35.214 so the air drops below the dew point temperature, which forces the water in the air to condense on the cold surface. 00:00:35.214 --> 00:00:39.615 And the most efficient way to make something cold is to use the refrigeration  cycle, 00:00:39.615 --> 00:00:47.861 So dehumidifiers, at least good ones, are really just air conditioners with the parts rearranged and with different controls. 00:00:47.861 --> 00:00:49.584 Don't be confused by that, though- 00:00:49.584 --> 00:00:54.261 they do not cool the air. In fact, they make it hotter as they extract moisture 00:00:54.261 --> 00:00:56.802 because of how latent heat works. 00:00:56.802 --> 00:01:00.149 The other video explains all that in much too much detail. 00:01:00.149 --> 00:01:04.282 But it turns out you don't have to use the refrigeration cycle. 00:01:04.282 --> 00:01:08.148 This is a totally different kind of dehumidifier. 00:01:08.148 --> 00:01:10.403 It's nowhere near as bulky and heavy. 00:01:10.403 --> 00:01:15.112 It's a lot quieter, and it doesn't have anything inside of it that gets cold. 00:01:15.112 --> 00:01:18.079 In fact, part of it gets very hot. 00:01:18.079 --> 00:01:26.803 And somewhat strangely, while it does rely on the condensation of water on a cold-ish surface to collect it in this bucket, 00:01:26.880 --> 00:01:30.942 it doesn't have to do anything at all to get water out of the air. 00:01:30.942 --> 00:01:33.403 That happens naturally. 00:01:33.403 --> 00:01:38.444 This is called a rotary desiccant dehumidifier, and they are quite fascinating. 00:01:38.444 --> 00:01:43.789 But they're also really, really uncommon here in the US. 00:01:43.789 --> 00:01:48.855 I had to resort to this "Aeocky" thing from Amazon just to tinker with one. 00:01:48.855 --> 00:01:50.980 And having run some tests with it, 00:01:50.980 --> 00:01:54.067 I think I'm pretty sure I know why they're not common here. 00:01:54.067 --> 00:01:57.034 Uh, first though, let me explain how it works. 00:01:57.034 --> 00:02:02.501 If the word desiccant rings a bell, that's a substance which is hygroscopic. 00:02:02.501 --> 00:02:07.322 It likes moisture so much that it'll just absorb it right out of the air. 00:02:07.322 --> 00:02:11.829 Though it might adsorb it, but I am not getting into that distinction. 00:02:11.829 --> 00:02:13.194 Ask a scientist. 00:02:13.194 --> 00:02:18.417 Silica gel is a common desiccant and  you've probably run across that in those little pouches which, 00:02:18.417 --> 00:02:21.969 through the misuse of quotation  marks, appear to be daring you to eat it 00:02:21.969 --> 00:02:24.951 despite the fact that you genuinely should not. 00:02:24.951 --> 00:02:29.931 Uh anyway, there are also desiccant products sold specifically for dehumidification purposes. 00:02:29.931 --> 00:02:35.594 Damprid is a product which contains various salts which are very hygroscopic. 00:02:35.594 --> 00:02:38.682 Those work well for small spaces  like closets, 00:02:38.682 --> 00:02:43.933 but the trouble with that approach is that most desiccants are single-use items. 00:02:43.933 --> 00:02:48.096 Once they've absorbed moisture, they're used up and stop working. 00:02:48.096 --> 00:02:51.083 Except, not always. 00:02:51.083 --> 00:02:53.608 Silica gel has a trick up its sleeve. 00:02:53.608 --> 00:02:57.873 It'll adsorb moisture from the air just fine, but if you get it hot enough 00:02:57.873 --> 00:03:01.493 it'll actually release the moisture it's adsorbed. 00:03:01.493 --> 00:03:03.732 That means it's a reversible process, 00:03:03.732 --> 00:03:10.857 and some clever person figured out a way to exploit that and make a functional dehumidifier using silica gel. 00:03:10.857 --> 00:03:12.912 That's what's going on inside of here. 00:03:12.912 --> 00:03:18.711 And through the magic of buying two of them, I have a pre-made one right here so I can show you. 00:03:19.728 --> 00:03:20.726 Now, 00:03:20.800 --> 00:03:24.738 you may notice there's a bit of a wiring situation going on. 00:03:24.738 --> 00:03:27.890 This was to run an experiment which turned out to be a dud. 00:03:27.890 --> 00:03:31.577 I'm not going to get into it here, but I will tell the story on Connextras. 00:03:31.577 --> 00:03:38.080 The Cliffsnotes version is that if you spot what might be a design flaw in this thing's operational logic... 00:03:38.080 --> 00:03:39.166 no you didn't. 00:03:39.166 --> 00:03:42.212 This is more or less an optimal design. 00:03:42.212 --> 00:03:48.939 What we find inside is a slowly rotating wheel made of silica gel with a tight corrugated structure. 00:03:48.939 --> 00:03:54.226 Quick side note, my brain doesn't like that this is called gel. 00:03:54.226 --> 00:03:59.265 To me, that implies soft and squishy and this stuff is very hard. 00:03:59.265 --> 00:04:03.191 But the technical meaning of gel  has to do with chemical structures and polymer science, 00:04:03.191 --> 00:04:04.626 and I don't make the rules. 00:04:04.626 --> 00:04:07.979 Anyway, the corrugated structure gives it a large surface area, 00:04:08.209 --> 00:04:14.878 and a fan not unlike one you'd find in a PC case, gently pulls air through the desiccant wheel. 00:04:14.878 --> 00:04:16.898 But not all of it. 00:04:16.898 --> 00:04:22.004 One section, roughly a fifth of the wheel's area, is covered both front and back 00:04:22.004 --> 00:04:25.288 so it's not in the airstream of the large  fan. 00:04:25.288 --> 00:04:30.653 Instead, that section of the wheel has a completely separate airflow situation going on. 00:04:30.653 --> 00:04:33.119 Notice this small blower. 00:04:33.119 --> 00:04:40.639 It takes air in through this hole, then sends it via a duct to a metal box sitting right behind the desicant wheel. 00:04:40.639 --> 00:04:42.792 That box contains a heating element. 00:04:42.792 --> 00:04:45.605 You can actually see it glowing when operating. 00:04:45.605 --> 00:04:51.665 And that heats the air coming from the blower before it's sent through that small section of the desiccant wheel. 00:04:51.760 --> 00:04:57.898 The now hot air is able to regenerate the desiccantand release the moisture it's holding on to. 00:04:57.898 --> 00:05:04.266 I can actually feel when I put my hand in front of the hole that the air coming out is very damp. 00:05:04.266 --> 00:05:08.837 It's like the air in a bathroom after taking a long hot shower. 00:05:08.837 --> 00:05:10.525 But that's not really helpful. 00:05:10.525 --> 00:05:15.717 We're trying to remove that water from the air, not just heat it up and move it around. 00:05:15.717 --> 00:05:20.068 And that's why this machine also has this heat exchanger. 00:05:20.068 --> 00:05:25.332 This simple plastic... thing is constructed much like a radiator. 00:05:25.332 --> 00:05:28.812 The bottom side of it has a wall separating it into two halves. 00:05:28.812 --> 00:05:32.964 So these two holes become an intake and an exhaust. 00:05:32.964 --> 00:05:39.261 Air pushed through this hole will travel up all these tubes which are definitely not just plastic drinking straws. 00:05:39.261 --> 00:05:46.081 And then when it reaches the top, it ends up coming back down these tubes before exiting the second hole. 00:05:46.081 --> 00:05:50.088 This bundle of straws may be simple, but when it's installed in the unit 00:05:50.088 --> 00:05:56.580 we find that its intake and exhaust holes line up with the intake and exhaust of the small blower. 00:05:56.580 --> 00:06:04.687 That means the blower is actually just repeatedly moving air through this thing and that small section of the  desiccant wheel in a loop. 00:06:04.687 --> 00:06:10.025 Before long, because of the heating element, that loop of air gets very  hot and very damp 00:06:10.025 --> 00:06:13.033 from all the moisture it's releasing from the desiccant wheel. 00:06:13.033 --> 00:06:21.635 In fact, it's so damp that the air inside of it is practically fully saturated with water and at nearly 100% humidity. 00:06:21.635 --> 00:06:27.001 That means if the air inside these tubes is cooled even just a little bit, 00:06:27.001 --> 00:06:31.565 it's gonna fall below the due point and moisture will condense out of it. 00:06:31.565 --> 00:06:38.293 And wouldn't you know it, when assembled, the straws are in the same air path as the rest of the desiccant wheel. 00:06:38.293 --> 00:06:42.627 So, the large PC fan ends up cooling them with ambient air. 00:06:42.627 --> 00:06:46.331 You can actually see them right here in this assembled machine. 00:06:46.331 --> 00:06:52.295 The slightly chilly walls of these straws will quickly build up condensation on their insides. 00:06:52.295 --> 00:06:58.308 And once enough builds up, it falls down into the bottom cavity and comes out these two holes. 00:06:58.308 --> 00:07:02.288 Then that water is directed into the little collection bucket. 00:07:02.288 --> 00:07:04.510 This is a really clever idea. 00:07:04.510 --> 00:07:09.299 The desiccant naturally grabs moisture from the air regardless of temperature, 00:07:09.299 --> 00:07:14.263 which is a huge advantage compared to a conventional dehumidifier. 00:07:14.263 --> 00:07:21.067 Since those work by cooling, in colder ambient conditions they can freeze up, which limits their effectiveness — 00:07:21.067 --> 00:07:24.999 a potentially huge issue if you live where it's cold and damp, 00:07:24.999 --> 00:07:29.953 as this freezing up can start to happen at fairly normal room temperatures. 00:07:29.953 --> 00:07:34.261 In contrast, this thing doesn't care what the air temperature is. 00:07:34.261 --> 00:07:39.965 It does need to be above freezing, of course, as otherwise the water it collects will turn to ice. 00:07:39.965 --> 00:07:41.433 But if the air's got water in it, 00:07:41.520 --> 00:07:47.385 the silica gel will simply yoink it right out as it passes through the honeycomb structure. 00:07:47.385 --> 00:07:51.217 And since you can reverse that process simply by heating it up, 00:07:51.217 --> 00:07:57.805 a slowly rotating wheel of the substance can create an infinitely repeatable moisture charge discharge cycle 00:07:57.805 --> 00:08:00.000 with just a few moving parts. 00:08:00.560 --> 00:08:05.507 You could say this machine has more moving parts than a conventional dehumidifier 00:08:05.507 --> 00:08:11.350 since it has two fans plus the synchronous motor to rotate the  wheel and the gears inside of there. 00:08:11.350 --> 00:08:16.225 But all of those parts are a lot simpler than a refrigeration compressor. 00:08:16.225 --> 00:08:21.684 And this approach doesn't require any environmentally harmful refrigerant, which is a nice bonus. 00:08:22.504 --> 00:08:23.365 Although. 00:08:23.365 --> 00:08:26.584 We've made a lot of progress making better refrigerants 00:08:26.584 --> 00:08:31.304 and I kind of wish we'd stopped paying so much attention to that as a boogeyman 00:08:31.304 --> 00:08:37.207 because it's not only a solvable problem  but actually has been solved in many applications. 00:08:37.207 --> 00:08:42.277 Every article I read about some new alternative to refrigeration generally ignores that progress 00:08:42.277 --> 00:08:44.384 and it annoys me very muchly. 00:08:44.384 --> 00:08:52.144 On that note, what a perfect segue to the, uh not exactly small caveat of this approach. 00:08:52.144 --> 00:08:55.768 That heating element is pretty power hungry. 00:08:55.768 --> 00:09:00.903 While the motors in here are small and only consume about a dozen watts of power, 00:09:00.903 --> 00:09:05.248 the heating element needs roughly 300 watts. 00:09:05.248 --> 00:09:12.224 That puts it at an extremely similar power consumption to this conventional vapor compression machine. 00:09:12.224 --> 00:09:16.259 Which is somewhat surprising considering how much smaller and lighter this is. 00:09:16.259 --> 00:09:23.172 But it's also convenient because it means we can simply run these things in the same conditions for the same amount of time 00:09:23.172 --> 00:09:28.127 to judge how much more or  less energy efficient they might be compared to each other. 00:09:28.127 --> 00:09:33.854 And if this footage looks familiar,  that's because I did all this testing last year. 00:09:33.854 --> 00:09:41.278 I set up two very large humidifiers in this small bathroom to keep the humidity as high as reasonably possible. 00:09:41.278 --> 00:09:45.281 Together, they got the room to around 80% relative humidity. 00:09:45.281 --> 00:09:47.696 Mm mm moist. 00:09:47.696 --> 00:09:54.643 The first test I ran last year was with the rotary desiccant machine because this was the one I was most curious about. 00:09:54.643 --> 00:09:58.074 I've used plenty of these  conventional dehumidifiers in my life, 00:09:58.074 --> 00:09:59.924 but never one of these. 00:09:59.924 --> 00:10:06.461 After two hours of running in that miserable room, it had consumed 600 watt-hours of energy. 00:10:06.461 --> 00:10:12.577 And at the end of two hours, it was clear that the dehumidifier was losing the battle with the humidifiers 00:10:12.577 --> 00:10:17.725 as the room remained at about 80%  humidity the whole time. 00:10:17.725 --> 00:10:20.723 But how much moisture did this thing get out? 00:10:22.003 --> 00:10:24.080 N- not, not a whole lot. 00:10:24.080 --> 00:10:31.476 I poured the water from its collection bucket into this container on a tared scale and it read 188 grams. 00:10:31.476 --> 00:10:33.868 After I spilled a little of it... 00:10:33.868 --> 00:10:39.062 I tried to get as much as I could back into the container and got it up to 191 g. 00:10:39.062 --> 00:10:42.179 But let's just go ahead and call that 195 just to be fair. 00:10:43.360 --> 00:10:47.346 That doesn't seem like very much, but to calculate liters per kilowatt-hour, 00:10:47.440 --> 00:10:51.348 first I needed to convert the 195 grams of water into ounces. 00:10:51.348 --> 00:10:53.814 And I got 6.87 oz. 00:10:53.814 --> 00:10:58.513 And since this is water, that's the same as 6.87 fluid ounces. 00:10:58.513 --> 00:11:02.763 Now, as I'm sure you know, there are 128 fluid ounces in a US gallon. 00:11:02.763 --> 00:11:06.510 So, that's 0.05367 gallons. 00:11:06.510 --> 00:11:09.516 And of course, there are 3.8 L in a US gallon. 00:11:09.516 --> 00:11:13.513 So, I multiplied that by 3.8 and got 0.2L, 00:11:13.513 --> 00:11:16.138 though I think some rounding may have happened. 00:11:16.138 --> 00:11:25.169 Anyway, with 0.2L produced using 600 watt-hours of energy, that works out to 1/3 of a liter per kilowatt-hour. 00:11:25.169 --> 00:11:29.531 Remember that number, 0.33L per  kilowatt-hour. 00:11:29.531 --> 00:11:32.858 Then I moved on to the conventional dehumidifier. 00:11:32.858 --> 00:11:35.622 With this guy set up in the same testing environment, 00:11:35.622 --> 00:11:40.115 at first it used a little less electricity than the rotary desiccant model. 00:11:40.115 --> 00:11:46.601 When first powered on, it was only consuming 265 watts, and this was on its highest fan setting. 00:11:46.601 --> 00:11:49.262 Power draw does vary though as refrigerant moves 00:11:49.360 --> 00:11:51.434 throughout the system and pressures change, 00:11:51.434 --> 00:11:53.235 so that did climb up a bit. 00:11:53.235 --> 00:11:57.573 Shortly into the test, it had increased to 345 watts. 00:11:57.573 --> 00:12:04.909 Also noteworthy, this fella was actually able to overcome the humidifiers and drop the ambient humidity. 00:12:04.909 --> 00:12:10.649 Running non-stop, the room stabilized at about 60% relative humidity. 00:12:10.649 --> 00:12:14.358 If that sounds like this machine was removing a lot more moisture, 00:12:14.358 --> 00:12:16.510 that's because of how it was. 00:12:16.510 --> 00:12:22.119 At the end, it had consumed 680 watt-hours, just a hair more than the desiccant unit. 00:12:22.119 --> 00:12:25.337 Yet, it nearly overtopped the water container. 00:12:25.337 --> 00:12:30.193 It produced over a liter of water, in fact, 1.2 L. 00:12:30.193 --> 00:12:33.956 After correcting for the difference  in weight between these two glass containers, 00:12:33.956 --> 00:12:38.560 the precise total was 1,207 grams of water. 00:12:38.560 --> 00:12:46.389 Which means this machine under these conditions will remove 1.775 L of water per kilowatt-hour, 00:12:46.389 --> 00:12:51.429 which is over five times as efficient as the rotary desiccant unit. 00:12:51.429 --> 00:12:53.568 But you know what's worse? 00:12:53.568 --> 00:12:57.063 Well, I also tested this piece of garbage. 00:12:57.063 --> 00:13:04.288 This is a "dehumidifier" which uses a Peltier element to create a cold surface via thermoelectric electric cooling 00:13:04.288 --> 00:13:10.030 and has a piddly little fan to blow a bit of air across that cold surface. 00:13:10.030 --> 00:13:14.125 Now thermoelectric cooling is just not an efficient technology 00:13:14.125 --> 00:13:22.236 and I'm pretty confident that the only reason these exist is to have the cheapest search result on [insert website] for "dehumidifier" 00:13:22.236 --> 00:13:26.146 and they're mostly just exploiting people who don't know any better. 00:13:26.146 --> 00:13:32.053 Want to know how much water this thing produced in that same incredibly damp testing environment after 2 hours? 00:13:32.053 --> 00:13:34.047 22g. 00:13:34.047 --> 00:13:36.517 Twenty-two grams. 00:13:36.517 --> 00:13:39.710 And how much energy did it spend producing that? 00:13:39.710 --> 00:13:41.625 70 watt-hours. 00:13:41.625 --> 00:13:45.124 Its only saving grace is that it uses just 35 watts of power. 00:13:45.124 --> 00:13:51.447 But its efficiency still works out to a paltry 0.31 L per kilowatt-hour. 00:13:51.447 --> 00:13:56.264 That is 1/5th as efficient as the vapor compression machine. 00:13:56.264 --> 00:13:57.578 Uh-oh. 00:13:57.578 --> 00:14:04.933 If this piece of junk is 1/5th as efficient as the machine which is five times as efficient as this machine... 00:14:04.933 --> 00:14:08.751 that means these two machines are equally efficient. 00:14:08.751 --> 00:14:17.166 It's not quite exactly the same, but 0.31 versus 0.33 L per kilowatt-hour is awfully darn close. 00:14:17.166 --> 00:14:22.431 And both of them are much much less than 1.775. 00:14:22.431 --> 00:14:24.139 Oh boy. 00:14:24.139 --> 00:14:31.010 So, uh, are rotary desicant dehumidifiers just as inefficient as Peltier dehumidifiers? 00:14:32.618 --> 00:14:34.495 Well... 00:14:34.495 --> 00:14:41.552 First I would be remiss if I didn't mention that this technology isn't unique to standalone dehumidifiers like this. 00:14:41.552 --> 00:14:44.801 It has a lot of applications in commercial spaces, 00:14:44.880 --> 00:14:50.930 especially where humidity must be kept in  control even when the air has to be quite cool. 00:14:50.930 --> 00:14:56.725 You'll also find similar rotating wheels of  silica gel in some energy recovery ventilators. 00:14:56.725 --> 00:15:00.613 Those are devices which allow for the exchange of indoor and outdoor air 00:15:00.613 --> 00:15:05.075 without changing the temperature or humidity of the indoor air very much. 00:15:05.075 --> 00:15:09.995 It's not always a rotating wheel, sometimes it's more like a cube surrounded by careful ducting, 00:15:09.995 --> 00:15:13.989 but they're starting to become common in residential settings, which is great. 00:15:13.989 --> 00:15:18.391 But the rest of this video is only going to address devices like these: 00:15:18.391 --> 00:15:25.471 standalone boxes you put in a room and plug into the wall which extract moisture from the air and collect it. 00:15:25.471 --> 00:15:31.436 Also, I don't want to suggest that this machine is representative of the technology writ large. 00:15:31.436 --> 00:15:34.871 For a start, this model is very cost cut. 00:15:34.871 --> 00:15:37.337 It can only be on or off. 00:15:37.337 --> 00:15:43.394 It doesn't have a humidistat - it only has two fan speeds and a sleep timer. 00:15:43.394 --> 00:15:46.084 That doesn't scream peak performance. 00:15:46.084 --> 00:15:51.293 Plus, my test conditions weren't exactly this technology's bread and butter. 00:15:51.293 --> 00:15:53.723 It was pretty warm in the room when I ran the test, 00:15:53.723 --> 00:16:00.960 which might have limited the ability of the plastic straw radiator  thing to cool the air inside below the dew point. 00:16:01.520 --> 00:16:05.286 So perhaps it would work better in cooler temperatures. 00:16:05.286 --> 00:16:09.675 Well, I've let this thing run for 2 hours in a whole bunch of different conditions 00:16:09.675 --> 00:16:16.089 and the amount of moisture it extracts from the air is weirdly consistent no matter what. 00:16:16.089 --> 00:16:18.687 I first tried my air conditioned office. 00:16:18.687 --> 00:16:26.246 The relative humidity during that test was only about 40%  and the air temperature was 72 Fahrenheit, 22 C. 00:16:26.246 --> 00:16:30.622 It still managed to pull out 160 grams of water in 2 hours, 00:16:30.622 --> 00:16:35.493 which is over 80% of what it did in the damp AF bathroom. 00:16:35.493 --> 00:16:38.866 Later, I did a test at home  with ambient humidity much higher, 00:16:38.866 --> 00:16:46.840 about 60% and roughly the same temperature, and it still extracted 186 g of water - not much more. 00:16:46.840 --> 00:16:54.927 Then I did it again in the morning with the temperature  down to 66° F, that's about 19° C, and the humidity at 50% 00:16:54.927 --> 00:16:58.713 and it extracted 169 grams of water. 00:16:58.713 --> 00:17:04.077 Kind of seems like no matter what it's going to pull out something like 80 to 90 grams of water per hour. 00:17:04.077 --> 00:17:09.152 So its efficiency is apparently not that affected by ambient conditions. 00:17:09.152 --> 00:17:13.035 Now, granted, I never tried this over the winter. 00:17:13.035 --> 00:17:17.194 I should have, but I forgot because of the fact that I live in the Midwest 00:17:17.194 --> 00:17:22.923 and DEhumidifying is the last thing on anybody's mind from October to March. 00:17:22.923 --> 00:17:29.377 But the weirdly consistent results I was getting were backed up by a little internet sleuthing. 00:17:29.377 --> 00:17:31.766 And speaking of internet sleuthing, 00:17:31.766 --> 00:17:38.707 if you watched the first video, you'll remember that I said cold but damp just isn't much of a thing here. 00:17:38.707 --> 00:17:43.121 And to clarify by "here" I mean most of the United States. 00:17:43.121 --> 00:17:47.410 Those are the conditions where a rotary desiccant machine is the most useful. 00:17:47.410 --> 00:17:51.532 So long as the air temperature is above freezing, this will keep working. 00:17:51.532 --> 00:17:54.971 And so they're fairly common in other climates. 00:17:54.971 --> 00:17:57.662 Like for instance, the United Kingdom. 00:17:57.662 --> 00:18:04.420 So I spun up Amazon.co.uk in the hopes I could see what the efficiency of other models is like. 00:18:04.420 --> 00:18:07.098 Luckily, it gave me this nice comparison chart. 00:18:07.098 --> 00:18:12.028 And, well, none of this looks much better. 00:18:12.028 --> 00:18:15.329 Granted, I have not done my own testing on any of these, 00:18:15.329 --> 00:18:21.217 and I don't know whether these numbers are the result of government testing or manufacturer testing. 00:18:21.217 --> 00:18:26.304 But assuming their best quoted moisture extraction occurs on their highest power setting, 00:18:26.304 --> 00:18:30.104 then these six models can manage 0.54, 00:18:30.104 --> 00:18:32.134 0.54 again, 00:18:32.134 --> 00:18:34.002 0.535... 00:18:34.002 --> 00:18:36.574 which rounds to 0.54 again, 00:18:36.574 --> 00:18:38.064 0.66, 00:18:38.064 --> 00:18:39.234 0.61, 00:18:39.234 --> 00:18:42.910 and 0.62 62 L per kilowatt-hour, respectively. 00:18:43.963 --> 00:18:46.518 That's just not that great. 00:18:46.518 --> 00:18:50.474 The best of these is twice as efficient as the one I've tested, 00:18:50.474 --> 00:18:55.406 but that's still only one third as efficient as a vapor compression machine. 00:18:55.406 --> 00:19:00.465 But I wanted to see how big boy commercial versions of this technology might fare. 00:19:00.465 --> 00:19:05.861 So I found this brochure from Technofrigo Tuscany, and it has charts! 00:19:05.861 --> 00:19:10.314 They back up the notion that the moisture removal of these things is oddly consistent, 00:19:10.314 --> 00:19:12.226 at least above room temperature. 00:19:12.226 --> 00:19:15.759 It seems to plateau right at about 20 degrees C. 00:19:15.759 --> 00:19:22.658 And the first model listed,  the AD3000, can remove 23 kg of water per hour. 00:19:22.658 --> 00:19:26.195 That's quite a lot, but so is its power input. 00:19:26.195 --> 00:19:29.403 32.3 kilowatts! 00:19:29.403 --> 00:19:31.818 That's like 10 British kettles! 00:19:31.818 --> 00:19:37.158 And it means its efficiency is only 0.71 L per kilowatt-hour, 00:19:37.158 --> 00:19:46.635 which is still significantly worse than the 1.771 L per kilowatt hour I observed in my test of the vapor compression dehumidifier. 00:19:46.635 --> 00:19:52.344 Frankly, I'm surprised this thing doesn't quote some average moisture removal per energy unit but... 00:19:52.344 --> 00:19:58.426 then again, the moisture it can pull out of the air varies a lot, especially once the temperature starts to fall, 00:19:58.426 --> 00:20:03.206 so I can understand that being  a practically impossible metric to give. 00:20:03.206 --> 00:20:09.772 And I should also note that the conditions for the vapor compression machine were fairly close to ideal during my testing 00:20:09.772 --> 00:20:13.123 and its efficiency is by no means consistent. 00:20:13.123 --> 00:20:19.228 So, the same general [?] factor applies to giving it an efficiency metric. 00:20:19.228 --> 00:20:22.092 So, seems pretty bad, right? 00:20:22.092 --> 00:20:25.758 Well, here's where things get a little complicated. 00:20:25.758 --> 00:20:31.273 First, setting efficiency aside, the rotary desiccant machine does have several advantages. 00:20:31.273 --> 00:20:33.117 For one, noise. 00:20:33.117 --> 00:20:34.682 Here's how loud it gets. 00:20:35.109 --> 00:20:41.198 [slowly building soft whooshing sound] 00:20:42.463 --> 00:20:44.877 And here's the one with the heat pump in it. 00:20:45.271 --> 00:20:48.751 [compressor growls to life and a loud fan starts] 00:20:49.112 --> 00:20:50.929 Yeah, these are pretty noisy. 00:20:50.929 --> 00:20:53.961 There's a compressor in there and a blower, which is quite loud. 00:20:53.961 --> 00:20:58.030 But in fairness, that's mainly because these move a lot of air. 00:20:58.030 --> 00:21:01.949 The little fan in here is hardly doing a thing in comparison. 00:21:01.949 --> 00:21:05.269 But in its lower power mode, it's even quieter, 00:21:05.360 --> 00:21:09.417 though its power level and effectiveness are  cut roughly in half. 00:21:09.417 --> 00:21:13.959 This thing is also much less resource-intensive to manufacture. 00:21:13.959 --> 00:21:21.038 It doesn't have nearly as much copper, steel, and aluminum in there as this thing, and is mostly just a plastic box. 00:21:21.038 --> 00:21:24.263 And while it has more components to put inside the box, 00:21:24.263 --> 00:21:26.457 a lot of them are commodity parts 00:21:26.457 --> 00:21:32.952 and none of them require brazing pipes together or pulling a refrigeration loop into a vacuum or charging it with refrigerant. 00:21:32.952 --> 00:21:36.819 And unless there's some dark secret to silica gel I don't know about, 00:21:36.819 --> 00:21:42.390 this thing doesn't appear to be any more environmentally costly than your average gadget. 00:21:42.390 --> 00:21:45.304 But more importantly to the efficiency conversation, 00:21:45.304 --> 00:21:49.527 because these things regenerate the silica gel using a heating element, 00:21:49.527 --> 00:21:53.608 they add heat to whatever space they're put in. 00:21:53.608 --> 00:21:57.063 Now, you'll recall from the last video that so does this thing, 00:21:57.063 --> 00:22:02.932 but most of the heat it generates is coming from the water condensing on the evaporator. 00:22:02.932 --> 00:22:08.448 In a similar way, the water which is adsorbed into the silica gel will also release latent heat energy. 00:22:08.448 --> 00:22:13.697 But if this thing pulls 100 grams of water out of the air per hour, 00:22:13.697 --> 00:22:16.833 which is more than I've ever observed it to do, 00:22:16.833 --> 00:22:22.651 that would only result in 73 watts of heat output, which is small potatoes. 00:22:22.651 --> 00:22:27.927 But the wrinkle here is that in the conditions where this machine is most helpful, 00:22:27.927 --> 00:22:29.893 cold and damp, 00:22:29.893 --> 00:22:32.846 those 300 watts of heat from the heating element 00:22:32.846 --> 00:22:38.320 are also going to lower ambient humidity  simply through heating the air in the room. 00:22:38.901 --> 00:22:42.494 For the sake of argument, let's say we've got this running in a bedroom 00:22:42.494 --> 00:22:46.655 with a floor area of 200 square feet and 8ft ceilings. 00:22:46.655 --> 00:22:52.140 That's 1,600 cubic feet of air volume or about 45 cubic meters. 00:22:52.140 --> 00:23:02.309 If that volume of air is at room temperature and 60% relative humidity, then there's about 460 g of water in that room's air. 00:23:02.309 --> 00:23:06.949 Now, if this is able to remove 80 grams of water from the air in an hour, 00:23:06.949 --> 00:23:11.786 then through moisture removal, it will drop relative humidity to about 50%. 00:23:12.738 --> 00:23:20.239 But if the air temperature in that room goes up by just 2° C thanks to the 300 watts of heat it's putting out, 00:23:20.239 --> 00:23:27.690 then even without removing any moisture, relative humidity would still fall to 53%. 00:23:27.690 --> 00:23:33.590 So the effect of the desiccant wheel is only marginally greater than simply heating the air. 00:23:33.590 --> 00:23:40.937 But wait, we're not just heating the air or just using the desiccant - we're doing both at the same time so... 00:23:40.937 --> 00:23:43.359 that's still good, right? 00:23:43.359 --> 00:23:48.564 Yes. So long as you actually need the  heat. 00:23:48.564 --> 00:23:51.555 That's the big caveat of these machines. 00:23:51.555 --> 00:23:56.217 The regeneration of the desiccant requires a substantial amount of heat energy 00:23:56.217 --> 00:24:03.056 which makes them very inefficient from a liters per kilowatt-hour of input power perspective. 00:24:03.056 --> 00:24:10.191 But if the extra heat they put out is actually useful, as it often is in the cooler climates these are usually sold in, 00:24:10.191 --> 00:24:12.527 then that's not much of a concern. 00:24:12.527 --> 00:24:16.796 It is resistive heat, which is typically the most expensive kind of heating, 00:24:16.796 --> 00:24:23.144 and that makes a holistic cost evaluation practically impossible when your primary source of heating is less expensive, 00:24:23.144 --> 00:24:24.588 which I hope it is. 00:24:24.588 --> 00:24:26.938 But it's not wasted energy. 00:24:26.938 --> 00:24:32.596 And it also contributes to lowering ambient humidity - which is its job. 00:24:32.596 --> 00:24:39.516 And honestly, that's why I find this application of the technology somewhat frustrating. 00:24:39.516 --> 00:24:43.328 The desiccant wheel is definitely effective at removing moisture, 00:24:43.328 --> 00:24:51.718 but since the heating it does to make that process happen also lowers ambient humidity by virtue of increasing the air temperature, 00:24:51.718 --> 00:24:56.253 it's hard to determine how much work the desiccant is actually doing. 00:24:56.253 --> 00:25:00.592 And this is especially hard to tease out because each room you might put this in 00:25:00.592 --> 00:25:05.356 will have a different rate of moisture ingress and a different amount of insulation - 00:25:05.356 --> 00:25:11.566 both of which affect how much each  of the two mechanisms will actually lower the ambient humidity. 00:25:11.566 --> 00:25:15.741 But there is clarity in when these make sense to use. 00:25:15.741 --> 00:25:19.528 If you're in a climate where it's chilly enough that you need to heat  your home, 00:25:19.528 --> 00:25:25.356 but it's so humid that you still have excess moisture issues, this is perfect. 00:25:25.356 --> 00:25:32.251 The extra heat it produces is still useful both because it's heat and because it will also lower humidity, 00:25:32.251 --> 00:25:37.953 but because it also extracts moisture, it lowers humidity even further. 00:25:37.953 --> 00:25:46.233 And if you have a space which is too humid, but also too cold for one of these vapor compression machines to work well due to icing, 00:25:46.233 --> 00:25:48.393 these are the perfect option. 00:25:48.480 --> 00:25:50.715 They're kind of the only option really. 00:25:50.715 --> 00:25:55.958 But if you're not looking for extra heat output when you need a dehumidifier, 00:25:55.958 --> 00:25:58.345 these are simply a terrible option. 00:25:58.345 --> 00:26:03.283 The input power of these two machines is roughly the same, just 300 watts. 00:26:03.283 --> 00:26:07.182 But this can only do about a fifth as much work. 00:26:07.182 --> 00:26:13.570 So if I wanted this rotary desicant machine to actually be as  effective as this small vapor compression machine, 00:26:13.570 --> 00:26:18.390 I would need to quintuple its input power to 1,500 watts. 00:26:18.390 --> 00:26:25.646 And then it literally would become a space heater which happens to be able to extract some moisture from the air. 00:26:25.646 --> 00:26:32.127 And all of that heat will be on top of the heat that's generated in  the desiccant as water adsorption occurs, 00:26:32.127 --> 00:26:37.854 which will be similar to the amounts that this thing generates when only using 300 watts. 00:26:37.854 --> 00:26:43.048 So yeah, if it's warm when you need to dehumidify, these are terrible. 00:26:43.048 --> 00:26:50.753 In conclusion, as nifty and clever as this idea is, it is not fit for purpose in many climates. 00:26:50.753 --> 00:26:54.594 Here in the US, it's not fit for purpose practically anywhere, 00:26:54.594 --> 00:26:59.083 which might explain why this unit has such poor reviews on Amazon. 00:26:59.083 --> 00:27:05.170 For example, here in the Midwest, the outdoor air temperature is so much colder than the indoor temperature throughout the winter 00:27:05.170 --> 00:27:08.286 that we simply do not need dehumidifiers. 00:27:08.286 --> 00:27:12.441 In fact, we often get out the humidifiers during the heating season. 00:27:12.441 --> 00:27:17.943 And by the time indoor humidity starts to climb in the spring to the point of being a problem, 00:27:17.943 --> 00:27:21.346 well, we're only a few weeks away from switching on the air conditioning. 00:27:21.346 --> 00:27:25.918 And since once that happens the  last thing we want is more heat, 00:27:25.918 --> 00:27:32.714 if a dehumidifier is necessary, we always want it to be as efficient and effective as possible. 00:27:32.714 --> 00:27:36.861 And the weather's warm enough that the icing probably doesn't matter. 00:27:36.861 --> 00:27:40.850 But if you're on one of those islands in the middle of the ocean which is always kind of humid, 00:27:40.850 --> 00:27:45.475 and you're far enough north or south of the equator to where it's just a bit nippy outside in the winter 00:27:45.475 --> 00:27:51.212 but not so cold that the indoor air is definitely going to be dry from all the heating you need to do... 00:27:51.212 --> 00:27:53.509 then these are a great solution! 00:27:53.509 --> 00:27:57.857 Although, I will say this one smells a little funny when it's running. 00:27:57.857 --> 00:27:59.814 That might just be this one, but I don't know 00:27:59.814 --> 00:28:03.710 it does have some very hot parts in there very close to the silica gel wheel 00:28:03.710 --> 00:28:08.946 so, I'm inclined to believe it's just the  nature of these things to smell slightly funky. 00:28:08.946 --> 00:28:12.243 It's not horrible or anything, but it is noticeable. 00:28:12.243 --> 00:28:13.519 Anyway, this is the end. 00:28:14.504 --> 00:28:17.036 ♫ adsorbently smooth jazz ♫ 00:28:18.000 --> 00:28:21.935 ...at least good ones are really just air  conditioners with the parts rearranged... 00:28:21.935 --> 00:28:27.317 and yeah, well, we've already screwed  it up because I completely forgot that 00:28:30.400 --> 00:28:32.682 I don't think this should be on the desk yet. 00:28:32.682 --> 00:28:34.809 Uh, first though, let me explain how it works. 00:28:35.400 --> 00:28:37.457 That was bad. That was just bad. 00:28:37.457 --> 00:28:43.210 Once again, I'm doing disassembly work on camera just because that's what I'm doing. 00:28:43.210 --> 00:28:46.255 There's something wrong with this paragraph which I didn't notice. 00:28:46.255 --> 00:28:50.182 I've used plenty of these conventional dehumidifiers over the years of my life. 00:28:51.330 --> 00:28:52.425 Why did I do that? 00:28:52.425 --> 00:28:54.573 Those work well for smalls spa sf... 00:28:55.196 --> 00:28:56.095 Dammit. 00:28:56.095 --> 00:28:58.611 I have a pre-unmade one. 00:28:58.611 --> 00:29:01.406 Oh, yeah. I grabbed... I put my finger right on the heat sink. 00:29:01.406 --> 00:29:02.468 That hurt. 00:29:04.583 --> 00:29:07.679 end of video captions gag 00:29:08.434 --> 00:29:11.361 you thought I forgot about this, didn't you? 00:29:11.361 --> 00:29:13.797 I mean I did forget to do any testing over the winter. 00:29:13.797 --> 00:29:16.202 So I suppose that wouldn't be an outlandish thought. 00:29:16.202 --> 00:29:17.534 This isn't much of a gag, though, is it? 00:29:17.534 --> 00:29:19.557 Well it was another dry subject.