Thermoelectric Blaster Flings Ice Projectiles

Nerf blasters are fun and all, but flinging foam can get old. Picking it up again, even moreso. This blaster from [Concept Crafted Creations] gets around that annoying problem by shooting ice instead. 

The concept was to build a better water gun with longer range—and what better way to do that than by shooting ice instead? The blaster relies on a PVC air tank for propulsion—one of the most controversial design choices you can make if you read the comments around here. It’s charged by a small air compressor, and dumping the air is handled by a solenoid valve. So far, so simple.

What makes this blaster special is where it gets its ammunition from. The blaster uses a custom CNC-machined block from PCBway to act as a freeze chamber. Water enters an aluminum block, and is cooled by thermoelectric elements. Once the projectile has frozen inside the chamber, it’s stuck in place, so the chamber is then heated by a small heating element. This melts the projectile just enough to allow it to be fired.

It’s a complicated but ingenious way of building an ice blaster. It does pack some real punch, too. It shoots the ice projectiles hard enough to shatter wine glasses. That’s enough to tell us you don’t want to be aiming this thing at your pals in a friendly match of Capture the Flag. Stick to paintballs, perhaps. Video after the break.

37 thoughts on “Thermoelectric Blaster Flings Ice Projectiles

  1. Leaving aside all the obvious questions about this build, it’s asking an awful lot of those two little peltier coolers. I wonder what its cycle time is, and how many shots you get from a battery charge. Water takes 330 watt-seconds per gram to freeze, and you can’t get more than very few watts out of those peltiers at that temperature, unless ambient is below something like 10 C.

    1. That heat sink is going to be scalding hot even if the outside temperature is down to zero. There’s 40-60 Watts of heat going into it, plus whatever is getting pulled out of the aluminium block. With the heatsink hot, the peltier is going to be even hotter, and the cooling side won’t reach below zero to actually freeze anything.

      You can compare what a big cooling fan you need to keep a 60 Watt CPU down to room temperature, or less than 10 C above the ambient temperature.

      Another oddity is the heating element in the block, because once you switch a peltier element off, heat immediately starts leaking back from the hot heatsink through the TEC and into the object you were just cooling, so you don’t actually need it at all. You can also switch it backwards for a moment. To me this just sounds like extra detail to make the story – like liars often do to add false credibility.

  2. I’m calling it a fake, because the TEC is not cooled well enough to freeze water.

    I’ve done the calculations and a bit of hands-on experiment as to what it would require to make a single ice cube using peltier elements in a couple minutes, and it’s hundreds of Watts. There is NO WAY this guy is making ice with that dinky little heat sink and fan – this is just an ordinary air powered pellet gun.

      1. I tried to search Youtube for someone who who actually tried freezing water with Peltiers, but ALL of the videos were clickbait clips from India. People gluing computer heat sinks to peltier elements and whoops, suddenly you have ice cream! Catchy music though. The most credible videos showed some ice forming after 10-20 minutes with a 12 Volt 15 Amp unit with the hot side cooled by water off the tap. That’s 180 Watts with a dT of 20-30 C.

        I wanted to build an ice cube maker that could make two small ice cubes in two minutes for a glass of whiskey, and calculated it would need over 500 Watts of power.

        It’s one thing to have thin wisps of frost forming on a Peltier plate and measuring it go down to negative temperatures, and actually freezing grams of water and all of the aluminum etc. with heat leaking in from everywhere, condensation… etc.

        1. Peltier coolers are one of my top-10 sources of hacker tears!

          They sound so good, but when you actually get around to shedding all the heat, it’s a lot harder.

          I would say “crunch the numbers first”, but come to think of it, I don’t know of a good source of such numbers. Anyone?

          1. Even if you do crunch the numbers and read the datasheets, the additional problem is that most Peltier elements you can find from the usual sellers are fake and don’t perform anywhere like the specs would suggest.

          2. I built a chiller for my K40 laser with peltiers four years ago. A prototype to measure performance used two units. I bought another eight from the same vendor for the final unit, and found the performance was half as good. Same vendor, a couple of months apart. They switched suppliers on me. Pandemic supply chain issue maybe.

            3rd time’s the charm. Eight 40 mm peltiers running at about half of max power (200 watts input) keeps my coolant loop at 18 C, up to a room ambient of 25 C. The hot side runs up to about 40 C and uses four 5″ fans and a couple of square feet of heatsink area to cool it.

            It’s kind of ridiculous how poor TECs really are, but they are effective when engineered appropriately.

          3. This is a good example datasheet which even shows COP curves. You just have to eyeball them a bit. https://cdn.awsli.com.br/945/945993/arquivos/TEHC1-12710.pdf

            Vortex tubes are a somewhat similar device – interesting, but tricky to actually make good use of because turboexpansion of half the flow of some compressed air isn’t nearly as efficient as phase change especially when you only need a small temperature drop. The wearable 2500btu/h vests are a great luxury in hot working conditions though, and I think there’s some edge cases for vortex tubes – such as using one instead of joule throttling in some kinds of cryocooler, where they’re actually more efficient again.

  3. This is an obvious fake. The physics numbers don’t add up.

    We built something similar-ish years ago though.
    The difference is, we didn’t freeze the water in-place.
    We shaved a block of ice and rammed the shavings into the chamber.
    Ramming them in and compressing them was good enough to keep them from exploding when they hit the air after firing.
    It was also marginally safer since the packed slush was more like a frangible round that disintegrated on impact.

    We also found we didn’t need to heat the mold/chamber actively.

    So I guess the only similarities were that it was air driven and it shot frozen stuff.

    Oh, and it wasn’t made because we needed to make a dumb video to justify a sponsorship deal…

  4. It does remind me though of something the Robert Redford character said in “3 Days of the Condor”.
    About a man being shot, the entrance wound had the characteristics of a .38 caliber bullet, no exit wound, and no bullet found in the body.

  5. Makes more sense to have pre-loaded copper pipes in the freezer, and you can carry the ammo to the backyard in a cooler. Heat the copper round a little and it should fire, I’d modify a bead blaster though. 15m is too long to wait between rounds, sound like less fun than nerf if you have to wait that long. On a cold day you could have a magazine full……

    1. Why not go with something more stable like sugar?
      Hard candies weight loaded with tungsten powder would probably be more effective. Melts in your wound tract not in you magazine.

    2. Please, that’s weaker than a pellet gun and likely weaker than a paintball gun. Three wine glasses in a row and only the very first one breaks???

      This is how absurd and frankly stupid laws come to be (looking at you 3d printed regular gun laws/outrage)

    3. There is no way this thing has enough muzzle energy to leave more than a welt on a human, let alone kill them. “Professionals” wouldn’t go near it, and the vast majority of gun murders are committed with stolen guns… not “ice bullets” from a smoothbore airgun. The barrel in this thing isn’t rifled, so there’s no spin stabilization, and it simply does not have enough “oomph” to travel far enough with enough energy to do what you’re describing.

      Sure, it’s a neat build, but bad CSI dramas != reality.

  6. I really like the novelty of the idea.

    From my Peltier builds, this is barely plausible at best. The cooling block is way too large and not insulated. Will you really get ice this way?

  7. You can get water ice with a Peltier, but you’ll need multiple stage peltier (each stage must dissipate the heat of the previous stage). We’ve made -40°C with a 3 stage unit (dissipating around 60W for a 0.8W heat source). If it wasn’t only for the “no moving part” rule, we’d never had to use a Peltier for this. Peltier are really bad at lowering temperature (if made perfectly, you can get a delta of -20°K per stage, but with an efficiency of ~10%), so every watt of heat you’re extracting means you have to extract 10 watt on the hot side. A 2 stage design means extracting 100 watt per watt (and this is purely technical, in reality it’s a lot worse since heat creeps in everywhere). Then you have to account to the poor insulation of your cold cavity (every watt that’s entering through conduction), the poor flat surface area of the Peltier’s face that means that as soon as ice is forming on it, the ice layer will insulate the remaining water and prevent if from freezing. In the end, you’ll get much better results with a “cheap” stirling cooler that can show efficiency and good thermal chamber shape…

    1. From what I read about TEC’s, they can have a COP of 6 or greater, but only at a small temperature difference. I never used them myself, but I think it makes sense to make a 6 stage design.

      1). Stack two on top of each other.
      2). Heat spreader, and then use 2s2p.
      3). Head spreader, and then 2s4p.

      That would get you a 6 stage design with 14 TEC’s
      If the COP is high enough, it even makes sense to put 3 or maybe even 4 in series before you start putting them in parallel.

      But to be clear, I have never done this myself. I don’t really have a use for this technology, and the previous posts here of TEC’s that barely work are not encouraging either.

  8. Thanks to a guy that loves talking about various temperature-and-other-property-changing-boxes, I know that freezing a bullet with 2 peltier elements is straight up bullcrap

    1. It’s a very convincing hoax otherwise though. I wonder how he’s really doing it, two guns that look the same where one is the “ice gun” and the other is an ordinary pellet gun? Or maybe it works like he says except for the freezing part, and he’s putting the whole gun in the freezer for a few hours to freeze each shot?

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.