Coffee Cold Brewer




Introduction: Coffee Cold Brewer

About: I'm an inventor / maker / designer based in the Bay Area. My background is in residential architecture, film set design, animatronics, media arts, exhibit design, and electronics. I use digital design and fa...

With this coffee cold brewer, you can make the perfect cup of iced coffee in style. Using 3D printed fittings, some off-the-shelf laboratory glassware, and maple dowels, this piece turns the cold brew process into a performance.

There are a many methods of making iced coffee, here are a few I'm familiar with:

Japanese Iced Method: Coffee is brewed hot (pour-over, Aeropress, etc.), then dripped directly over ice, cooling it instantly. This basically tastes the same as hot coffee- same acidity, similar flavor profile and aromatics.

Crash Cooling: An easy way to do this one is to brew a cup of coffee, pour it in a steel cocktail shaker, then put the shaker in a bucket of ice until the temperature drops. This will take the coffee down to temperature without diluting it in melting ice, the way the Japanese Iced method will.

Nitro (pressurized): "treated with nitrogen and/or carbon dioxide under high pressure, then chilled in a keg and served on draught with a foamy head like a Guinness." - Bon Apetit

I love the taste of this kind of coffee. It really is like a coffee Guinness. Alas, I don't have the space for a setup like this, so I resort to the next best thing...

Cold Brewing: With this method, a small amount of coffee grounds are steeped in cold water, and drip the resulting brew at about one drop / 3 seconds. My impression is that boiling water releases caffeine and acidity much faster than cold water does. Steeping the grounds in cold water and releasing it slowly cuts the acidity significantly (you can taste that), but I've heard heard the argument that since it's steeping for so long (2-6 hours), the caffeine content is about the same as it would be in a hot brewed cup.

In any case, this method makes incredibly smooth, nutty, floral coffee without the acidic bite. I love the bitterness of hot coffee, but for some reason it's not as pleasant to me when it's cold.

Step 1: Design

This design went through a few iterations. At first, I was thinking of it as something that would hang on a wall, but I decided to go with a tripod design using dowels because I thought it would be more versatile- no need to dedicate wall space to it.

In the first design, there were 3D printed clips that would grab the necks of the flasks, but this came across less elegant to me than a simple hook with a recessed ledge. The hooks holt the flasks just fine, and would work if they were made out of ceramic, which I might explore later.

The STL files in this step are ready to prep for 3D printing (they'll need support structures on an FDM print), and the .f3d file is the final design file available for tweaking for personal use.

Step 2: Tools & Materials

McMaster Carr:


Hardware Store:

Tools Used:

  • FDM 3D Printer (I used a Dremel Idea Builder)
  • Drill Press (for boring out stopper)
  • Belt Sander (for making boring tool)
  • Pressure Clamps

Step 3: 3D Printing

3D printing with the latest version of Fusion 360 is a breeze. You can send your file directly to the new Print Studio app by going to File > 3D Print. Make sure it's set to High Refinement. Fusion 360 is free for students and hobbyists, and there's a ton of educational support on it. If you want to learn to 3D model the kind of work I do, I think this is the best choice on the market. Click the links below to sign up:



The Print Studio App let's you position the model and add supports, then sends it to a 3D printer. I used a Dremel Idea Builder.

  1. PRINT: The first version of the tripod part didn't work because there was too much overhang and not enough support structures, so I flipped it and it came out perfectly.
  2. SANDBLASTING: This step is by no means necessary, but I wanted to try it out because I saw Andreas Bastian's sandblasting finish pieces, and they looked amazing. However, he warned me that sandblasting white PLA probably wouldn't look as good as the black parts did, and he was right. The pieces had a smooth matte finish but the coloring was blotchy.
  3. PAINTING: To try to make the parts presentable again, I spray painted them with some cream colored acrylic spray paint, and they looked okay again.

Step 4: Prepping Laboratory Parts


In order for the water to drip, you have to have a siphon in the upper flask, otherwise you end up with a vacuum and no water will come out. To do this I inserted a brass tube into the stopper next to the valve.

  1. Cut the brass tube to about 4" long.
  2. Make a boring tool: sand, file, or grind the end of the off-cut tube to make it sharp around the edge.
  3. Screw the stopper down to a piece of scrap wood so it's snug, but no compressing the stopper (you don't want distortion.
  4. Chuck the tube into a drill press, and bore into the stopper.
  5. Press the 4" piece of tubing into the new hole for an airtight seal!


  1. Screw the two brass barbed fittings into either end of the needle valve. Make sure these are tight- some teflon tape or silicone caulk wouldn't be a bad idea here.
  2. fit the brass barb on the inlet end of the valve (the valve has an arrow indicating flow direction) into the large end of the stopper with the siphon.
  3. If the hole in the second rubber stopper doesn't fit the filter funnel, use the same trick described above to bore the hole to size.

Step 5: Assembly

Cutting the Dowels:

There are 5 dowels altogether:

  1. Three 6" long dowels for the three legs
  2. One 4" dowel for the lower segment of the vertical stand
  3. One 8" dowel for the upper segment of the vertical stand

Gluing and Clamping:

When the 3D printed parts' paint dried, I mixed some 60 second epoxy and glued all the parts together. Gluing end grain on wood is usually a no-no, but There's so little weight involved i this project that I wasn't concerned about it.

As you can see in the first photo, there's a hole recessed in the faces of the 3d printed parts- these are intended for pegs that would joint the ends of the dowels to the ends of the 3d printed parts. Unfortunately, the joining dowels I was planning on using had run out in the shop, and I didn't want to put off building this project any longer, so I just glued the ends together. It's been nocking around the shop for a week now, and it's still holding up for what it's worth.

It's very important that the two hook parts are perfectly aligned, so I clamped them down to some scrap plywood while the epoxy cured. In about an hour, the whole piece was glued together.

With everything in place, I went over the piece with two coats of spray-on matte acrylic for easy cleaning.

Step 6: Brewing

  1. Fill the upper flask with ice and push in the stopper with the spigot. This one goes on the upper hook pointing down (duh).
  2. Place a piece of filter paper in the funnel. Get it wet to even out the contact with the coffee grounds.
  3. Please the stopper with the funnel on the lower flask. Don't make it air tight! You have to allow air to escape, otherwise the coffee won't drip. Fill the funnel with medium grind coffee grounds. Don't pack it in, because it'll overflow when it's saturated- just fill to the top and wipe off any excess so the grounds are level with the top of the funnel. Prima Coffee recommends 500ml of ice water to 80g of coffee grounds, but I just eyeballed it and the result was good.
  4. Saturate the grounds slowly with cold water. Slow is the key because you don't want it to overflow. In hindsight, a larger filter funnel would probably be a good idea to give you a little room for error.
  5. Place another piece of filter paper on top of the grounds. This piece keeps the water evenly distributed through all the grounds; without it, the dripping would slowly carve a tunnel through the middle and the coffee would be very weak.
  6. Open the valve ever-so-slightly. You'll want one drop every 5 seconds or so, but the key is to sync up the dripping of the steeped coffee with the dripping of the valve. If the valve is dripping faster than the funnel, it's going to overflow (again, duh).

Step 7: Wait...

500ml of ice water took about 4 hours to brew. When it's done, you'll have about 2 cups of coffee by american standards. It's mesmerizing and kind of relaxing watching it drip.

Again, a larger filter funnel would be a great idea. As you can see in the gif, it overflows ever so slightly- thankfully not enough to lose an real volume of coffee grounds or spill into the flask.

Step 8: Savor

If you're a coffee lover, you've got to try this. The taste is familiar, but it's the smoothest, creamiest version I've ever had.

A few things I would change:

  1. A bigger filter funnel would probably keep the coffee from overflowing as it's steeping. If you plan to build this project, I recommend the 7cm filter funnel listed in the Tools & Materials step.
  2. For some reason, I spaced out and made the second post (the one between the two hooks) 13" long instead of 8" long. Making it 8" long (the way I designed it) would bring the drip nozzle closer to the coffee and mitigate splashing.
  3. As I mentioned in the assembly step, a more robust attachment between the 3D printed parts and the dowels would give me more peace of mind, but like I said, it's still standing up without any issues.
  4. I might change the design a bit to accommodate bigger flasks, in case I wanted to make more the 2 cups at a time.

Want to buy one? Go to my online store, Roundhead Design and put in an order!



    • Tiny Home Contest

      Tiny Home Contest
    • Metalworking Contest

      Metalworking Contest
    • Creative Misuse Contest

      Creative Misuse Contest

    35 Discussions

    All of your McMaster Carr links redirect me to the homepage. Can you tell supply me with some updated links? Thanks!

    1 more answer

    Looks like those links were out-dated. I just updated them, please try again.

    How did you go about in getting the ice in the top beaker? Im assuming that you crushed it up? I plan on taking this to work and my ice crushing abilities are fairly limited. Using a hammer is no problem, but having something toput the ice in to crush it would be a little more challenging.

    3 replies

    We have one of those bead ice machines at work, but you could make smaller cubes with one of these:

    Thanks, I'll look into those.

    Another question. I have it completed and am doing some test runs and I'm noticing that the system is having a hard time getting the water to drop. It is almost as if the top bulb is not getting enough air to release the vacuum. It is also leaking around the rubber stopper. I have tried opening up the needle valve more, but then it starts to come out in a small stream until it slows to a small drip. I have the air tube about 4in as the was suggested. Should I maybe add a second air tube?


    What you've got is a vacuum because there's nowhere for return air to get back into the flask. Check step 4: I solved this problem with adding a copper tube as a siphon.

    I have experimented using a large tea diffuser type jug. I place the required coffee grounds into the diffuser, add a quantity if ice cubes and tap water.

    Let it sit overnight in the fridge (or on the bench, if the nights are cool).

    Remove the coffee grounds and serve as you desire.

    This works well and is simple.

    1 reply

    Yeah, I think they call that the toddy method. It tastes completely different than the cold drip method even with the same beans. There are so many ways to make coffee.

    I'll stick with my half gallon mason jar sitting on my counter. This is an adorable outfit but it seems to take a really simple project and turn into something ...... just a bit over-engineered.

    1 reply

    That's kind of the point. I'm turning something mundane into an exhibition...

    What an awesome product! I love the look of artistic cold coffee brewers. Would you be willing to give a rough estimate of how much this cost to build?

    2 replies

    Glad you like it! Not counting tools, if you buy everything new it costs about $40-$50 in materials.

    No way, that little?? Amazing! Especially considering the cost of some "high end" cold brewers. May take awhile, but I will definitely be making this at some point!

    autolycos, the stopcock isn't integrated, I bored one out using the same process described in the siphon step so it would fit the filter. The coffee and paper filters slow down the drip significantly. It drips about once every five seconds, where at least three seconds is recommended.

    2 replies

    I think my mention of the sep funnel was confusing. I mentioned that because the sep funnel does have an integrated stopcock, where the boiling flasks don't. Also, I've never needed to put a siphon in a sep funnel, which also removes a step.

    Those reasons are why I asked about your choice of a boiling flask on top.

    Aha! I was hoping someone with some chemistry equipment knowledge would weigh in. Thanks for the tip!

    Great idea! I love the look of the two boiling flasks. But, with the price of the needle valve setup, I'm curious about using a separatory funnel with an integrated stopcock. Do those drip too fast?

    This is a great idea, perfect for summer. I notice you said it took about four hours to brew, how warm was the environment, was it in direct sun?

    I feel like you could set it up for an overnight drip so you can have awesome chilled coffee for summer mornings (currently in Australia, so Summers get pretty warm)

    1 reply

    It wasn't in direct sunlight and it's a pretty steady 72°F indoors at Pier 9. My understanding is that the colder the water steeping the grounds, the smoother the flavor will be. So yeah, if it's warm during the day it's probably best to do it at night.

    Hi, great project. What is the design tool you used? Thx