Plate Chiller for Robobrew (Brewzilla)

Robobrew (Brewzilla) comes with a solid immersion chiller. If you’re happy with it, then read no further. But if you’d like to make your life a little easier and enjoy your brew day a bit more, then I highly recommend a plate chiller. I won’t get into the details here of what makes a plate chiller awesome because I’ll assume that if you’re reading this, you’d like a plate chiller for your Robobrew too.

There are a nearly countless number of options for equipment that can be used for this process. I won’t bother getting into the various options and instead I’ll just list the exact materials I used. They work excellent and you should have no reason to choose anything else. And before you ask, yes, I’m very cheap and chose the cheapest components I could find, while also making sure they were of high enough quality.

First, let’s start with the bill of materials

Grand Total: $186 (less if you already have some common items)

Part Selection

For 5 gallon batches, the small 20 plate chiller will get the job done. Here in New England we get very cold water in the winter time. When I run my faucet at full speed I can pump my wort directly into my fermenter at 70 degrees. In the summer it will take a bit longer so I’ll just recirculate back into the kettle until the ThruMometer tells me that the wort is coming out an appropriate temperature for pitching yeast.

Using the camlock quick disconnects is a no brainer. This allows you to bypass the recirculation arm and connect your chiller directly to your pump. Besides, the recirculation arm is designed for 3/8″ ID tubing so you wouldn’t want to use it anyway.

I chose to use male fittings on the chiller itself for a couple of reasons. Having both of them be male fittings means that you can connect your pump to either side of the chiller. This allows you to recirculate your cleaning fluid in both directions when you’re finished with it. Just be careful, because this also means you can run your wort in the wrong direction and not get very efficient cooling.

Having the ThruMometer is optional obviously, but it makes life so much easier. You can immediately see what temperature your wort is coming out at which saves you time and energy because once it reaches the appropriate temperature, you can just put the out tube directly into your fermenter. It only takes a couple of minutes for the pump to empty the entire Robobrew.

You will want to make sure to get the 1/2″ ID and 3/4″ OD silicone tubing. The thinner walled tubing is garbage. It’s so thin that your tubes will immediately kink and stop the flow of wort. I don’t even know why they make that stuff.

You’ll want to make sure you use a proper garden hose for your faucet. Using anything less strong could potentially burst. Trust me… (Did I learn this from experience? …. no…. why would you ask that?)

Having quick disconnects on the cold side of the plate chiller is awesome. The pack I recommend comes with enough quick disconnects to put it on your faucet, your garden hose, and your plate chiller. The catch is that you’ll need both ends of your garden hose to have female connectors, hence the garden hose repair kit.

Finally, I use zip ties as “hose clamps” for the silicon hoses.


Plate Chiller

You’ll want to attached your male camlocks to the chiller before the garden hose quick disconnects because you’ll need a wrench to tighten these, whereas the garden hose quick disconnects just take hand tightening.

First, apply a generous amount of teflon tape to the threads of the 3/4 NPT studs on the plate chiller. I did about 8 wraps. Then put the male camlocks and tighten with a large wrench. You do NOT want these to leak.

Once the male camlocks are on then you’ll want to add the female garden hose quick disconnects. I found that I needed to use two of the garden hose washers per disconnect in order to get a tight seal. I kept having leaks when I tried to use just 1 each. I added a little keg lube to each washer, just for a little extra protection against leaks and to help them to last longer. Once you have both washers inside the disconnect, just tighten each of these by hand, as tight as you can get it. Don’t worry, you won’t be strong enough to over tighten these.

Hot Side Tubing

Start by determining the length of tubing you’ll need from each side. Keep in mind that for the output you’ll have an extra 10″ or so provided by the ThruMometer. I chose to put a rolling table near my Robobrew while I use my system, so I only needed 6 feet of silicone tubing. I used about 3.5 feet on my input tube and about 2.5 feet on my output tube.

Input Tube

I used about a 3.5 foot tube for my input tube. Insert the male camlock with 1/2″ barb on one end of the tube and then insert the female camlock with 1/2″ barb on the opposite end of the tube. Use zip ties just past each of the barbs to prevent leaks. Warming the tubing can help these zip ties get nice and snug.

Output Tube

You’ll need two pieces of tubing for the output. I used about 2.5 feet of tubing and cut it in half.

One tube you’ll insert the female camlock with 1/2″ barb and the opposite end you’ll insert the ThruMometer. Technically there is no right or wrong way to insert the ThruMometer so just choose a direction you’re happy with.

Insert the opposite end of the ThruMometer into your other piece of tubing. Add zip ties to all of the connections, just past the barbs. Again, warming the silicone tubing will help you to get these zip ties nice and tight.

Use a nice pair of dykes or lead trimmers to cut the excess length of the zip ties. Make sure these are nice and smooth. You don’t want to cut your fingers on poorly cut zip ties. You can also use a small file and clean up the sharp edge.

Cold Side Garden Hose

Start by measuring how long you need your garden hose to be in order to reach your faucet and to reach your drain. Cut the garden hose in half (or whichever dimension you determined) and grab your repair kits.

Input Hose

Use the hose with the female fitting still attached as your input hose. On the cut end, drop a hose clamp around the hose, and then insert the female fitting from the repair kit. I found that using a little keg lube helped to get this fitting installed. If you don’t have any keg lube, you could probably just use some petroleum jelly. This means that you should now have a female fitting on both ends of your input hose.

Output Hose

Your output hose should be the hose with the male fitting still attached. On the cut end, drop a hose clamp around the hose, and then insert the female fitting from the repair kit. As with the input hose, using a little keg lube helps this go on easier.

Now tighten up all of your hose clamps (I highly recommend trimming it to length too, so that it doesn’t cut your fingers) and you should be ready to install the male quick disconnects. Thread a male quick disconnect into each female hose fitting (unless you don’t want a quick disconnect on your faucet). Don’t forget your garden hose washers!

I plumbed a faucet and a drain hose right where I brew


Before brew day, make sure to fill your Robobrew with some water and turn on your pump. Check for leaks. Do the same with your garden hose. If there are no leaks, then add some PBW to your Robobrew and run the pump for a while to make sure the tubing and the plate chiller are nice and clean.

After this you may want to run some fresh water through the plate chiller too to make sure there’s no residual soapy water in there.

Brew Day!

Once you’re ready to use your system, connect all of your tubing and hoses while you’re still boiling. You’ll want the output tube to be flowing back into your kettle at first. Make sure you have your cold side hoses connected to flow the opposite direction as your hot side tubing. If you run these both in the same direction, you’ll get very little cooling power.

I start by running boiling wort through the system for at least 10 minutes, in order to sanitize everything. I don’t like to take any chances.

This was taken during my testing to make sure there were no leaks. Hence why the excess length of the zip ties isn’t cut yet.

(Side Note: I *highly* recommend some kind of a hop filter or hop spider. You don’t want to clog your pump or your plate chiller on your big double NEIPA)

Once you’re ready to start chilling simply turn off your elements and keep your pump running. Now open your garden hose faucet all the way (you checked for leaks, right?) and watch the temperatures drop. Once you get a reading on your ThruMometer that you like (yeast pitch temp) just close the valve on the recirculation arm, pull the output tube out of the RoboBrew, and insert it into your fermenter. Now open the valve and wait for the system to finish pumping all of your cool wort into your fermenter.

Once you’re finished, I recommend keeping a small bucket or pitcher nearby to catch any drips as you disconnect each tube/hose.


Cleaning is so simple. I start by disconnecting the hot side tubing from the recirculation pump and use a garden hose nozzle to run fresh water through the system. I do this on both sides of the hot side tubing, back and forth, to get as much junk out as possible.

Once it’s running nice and clear and there’s no more hop particles or grains coming out of the plate chiller, then I connect the hot side tubing to the recirculation pump again. Then I put all of my various brewing pieces into the Robobrew (malt pipe, extra tubing, mash paddle, false bottom, hop spider, etc.) and fill it almost to the top with cold water. I turn on the heating elements and set the temperature to about 140° F and then add an appropriate amount of PBW (I’ve tried everything else, PBW is the best).

Now I turn on the pump and let the soapy water recirculate through the plate chiller for at least an hour. After an hour I turn off the pump, lift the plate chiller into the air to get most of the water out (prevents spills), then I swap the two female camlocks on the plate chiller so that I can reverse the direction of flow through the plate chiller, and run the pump for another 20 minutes or so. This ensures that the plate chiller gets thoroughly cleaned and there’s no hidden surprises when I go to brew next time.

Finally, I drain the Robobrew and rinse everything off and then put another couple gallons of water in the Robobrew and turn on the pump to rinse the chiller out really well. You could also just spray the end of a garden hose nozzle into it to rinse it, but you’ve got to rinse out the pump of your Robobrew anyway, so this is just one less thing to do manually.


It occurred to me while writing this that I spent a lot more money on this project than I expected to. I already owned the garden hose and the teflon tape and zip ties, but that still means I spent around $150 to do this.

Had I known going into this that I would have spent $150, I’m not entirely sure I would have done it. Not because it’s not worth every penny, it definitely is, but because I’m super cheap.

Am I glad I spent the money to do this? ABSOLUTELY! Brewing with this is so much easier than an immersion chiller and cuts out at least 30 minutes from my typical brew day. I value my personal time and saving 30 minutes of a rather boring task of chilling wort, is totally worth it to me, considering I brew 1 to 3 times per month.

I can’t recommend this enough. If you’ve even considered doing this yourself, don’t hesitate. Get the materials, put it together, and enjoy the most leisure wort chilling experience you’ll ever have.


Silkscreen has long been used on circuit boards to guide the assembler to know what components go where, as well as to show users details about the board such as where power and IO pins can be found. It’s a small detail that is all too often considered less important than other details of a circuit board design, but it really shouldn’t be.

When a manufacturer is getting ready to assemble your circuit board there will be a lot of details they need to pay attention to. They’ll need to make sure all of the locations that should be populated actually are populated, as well as making sure that any locations that have been marked as DNP (Do Not Populate) are indeed NOT populated. They’ll need to confirm that polarized components such as IC’s and LED’s are being populated the correct way so that they function properly. If there is a problem, they’ll need to quickly identify the location and work out the details of the problem. Silkscreen will help with all of this.

Cozy Silkscreen Parameters

When setting up the parameters of your EDA tool’s silkscreen text, I recommend the following “cozy parameters”. These are not minimums by any means, but if you have the room for it, these are good middle-of-the-road parameters that a PCB fab can easily print and an assembly shop can easily read while also leaving enough room on the rest of your board for your layout.

  • Text Height: 1mm
  • Text Width: 1mm
  • Line Thickness: 0.18mm
Cozy Text Height = 1mm
Cozy Text Width = 1mm
Cozy Line Width = 0.18mm


Whenever possible, try to keep the relative reference designators for components visible, even after the components get populated. Make sure that the silkscreen doesn’t get drawn across any open vias or thru-holes. And since silkscreen is usually only printed on top of soldermask, it would be a good idea to keep the silkscreen away from any exposed copper pads. If the silkscreen is too close to the soldermask this will be removed, often called “clipping”. In order to avoid silkscreen clipping, make sure no silkscreen is within about 0.1mm of the edge of the soldermask opening.

Keep reference designators and polarity indicators visible even after parts are populated

Polarity Indicators

Polarity is something that every assembler is concerned about. Silkscreen is usually the first thing an assembler looks for to define the polarity of a component. Having that silkscreen remain visible even after the components are populated is a great way of confirming that everything was done correctly.

Some examples of how to make sure the polarity of components is obvious would be to put a dot of silkscreen near the pin 1 of an IC, put a numeral 1 near the pin 1 of a connector, and put a C near the cathode pin of an LED. There are plenty of other methods, but these are some of the more popular ones.

Examples of polarity indicators

Taking the time to make sure your silkscreen is printable, visible, and legible will save you countless headaches and ultimately result in higher yields, quicker lead times, and less phone calls!

This article is part of an ongoing series of articles that will ultimately end up becoming a book I’m calling “Your Manufacturer Is Stupid”. Click this link to see more articles just like this one.

KiCon 2019

When I’m not brewing beer, I’m manufacturing circuit boards for a living. A lot of what I do every day revolves around how to handle projects that have design flaws. Most everything we build is very well designed, but because we introduce so many new designs to our process every single day, I spent a lot of time working with our team to figure out how to work around these design flaws.

Good thing is, I really enjoy it actually. It’s a fun challenge and I really enjoy working with our customers to come up with solutions to make their designs easier to assemble. With that in mind, I thought it would be a good opportunity to share some of the most common problems we run into while assembling circuit boards and share that with the attendees of KiCon in Chicago on April 26th and 27th.

The title of my talk is “Your Manufacturer Is Stupid – Help Them”. I know, it’s a real obnoxious title but figured it would drive traffic. The main point of the presentation will be to show the audience how many decisions are made during the manufacturing process and how KiCAD users can help their manufacturers answer questions before they’re ever even raised. I’m really looking forward to it and hope the turnout is strong so that we can have a 2nd even next year, and hopefully many more after that.

Earl Dreams of IPA – Post 2

So about that Experiment #4….

Something went dreadfully wrong. Imagine taking a band-aide off your sweaty arm, putting it in a microwave until it gets nice and crispy on the edges, and then chewing on this while drinking a nice glass of water with a few too many pellets of hops dropped into it. Yeah, that basically sums up how well Experiment #4 came out.

Obviously, I got an infection in the beer, which is a real bummer. Had a friend of mine over to taste it and he agreed, something went horribly wrong. It couldn’t have just been the recipe itself. All of the ingredients were solid ingredients. No weird curveballs there. I suspect my brewing partner (my 13 year old pug) may have gotten himself a bit too close to the fermentor while filling and added a fur or two to the mix.

Stay tuned for Experiment #5. This time my brewing partner has agreed to wait upstairs while dad fills the fermentor.

iSpindel Project – Post 2

Whenever it comes time that you want to build a DIY product, you really must understand the bill of materials first. For the iSpindel, it just was not very straight forward. There’s nobody really explaining what everything is and how it’s all used. The documentation for this just doesn’t make it all very clear. So I’ll do my best explain what each electronic part is, what it does, and where you can find it.

The brains of the whole operation is the ESP8266 chip. This tiny little chip packs of a lot of cool stuff in it. But unless you’re an electrical engineer, you don’t want to go out and buy this chip and build a circuit around it. You want all of that stuff done for you. The German folks who first put together the iSpindel were fans of the Wemos D1 Mini. The Wemos D1 Mini is a cool little single-board micro-controller that has all of the various things you need to interact with the ESP8266 chip.

But the D1 Mini doesn’t have the necessary sensors to give you the information you need to take all of your measurements. For that, you need to add a temperature sensor and a gyroscope.

For the temperature sensor, that’s easy. There’s a very popular temperature sensor called the DS18B20. This little guy is super cheap and readily available anywhere.

For the gyroscope our friends in Germany are recommending the MPU-6050 which is a very popular chip. But much like the ESP8266, you don’t want to buy just the chip. You want to buy a daughter board (often called a shield) that has all of the various control circuitry built into it. The GY-521 is a great daughter board for this purpose.

Once you’ve got those 3 main pieces of silicon (micro-controller, temp sensor, gyro), it’s time to power these bad boys. That’s where your Panasonic 18650 battery comes in. This is a hefty, lithium-ion rechargeable battery, packing 3400 mAh into one tiny package.

In order to control this battery, and be able to recharge it, you will want to use some kind of lithium-ion battery controller. Again, our German friends to the rescue. They recommend the TP4056 controller. This is a very popular DIY battery controller module that plays nicely with the Panasonic 18650 battery and the Wemos D1 Mini.

So now we have our silicon, we have our power, it’s time to wire all of these things together and give them an on/off switch. In order to do that properly, you’re going to want a few other electronic components. A solderable breadboard, an on/off switch, a 470 ohm resistor, a 4.7k ohm resistor, and a 230k ohm resistor. If you’re not familiar with resistors, ohm is the unit of measurement and “k” means “thousand”. So when you say 4.7k what you really mean is 4,700. When you say 230k what you really mean is 230,000 and so on.

There are all kinds of switches out there. They’re all dirt cheap and they all usually work really well. That being said, because they’re so dirt cheap, it’s nearly impossible to find just 1 of these at a reasonable price outside of a large electronics distributor’s website. So if you order these from the link I recommend, you’re going to end up with 50 of these. Better get creative with the other 49 of them.

There are a bunch of various solderable breadboards out there, but you’ll probably want one that is designed for the Wemos D1 Mini. It’ll just make your life a little easier when you solder all of these things together.

For the resistors, you could buy each of these resistors individually, but usually you’ll spend just as much, if not more, for shipping these things than if you just bought a kit with all kinds of resistors, so that’s what I’m recommending in my bill of materials.

Finally, you’ll also need a few various sized jumper wires. If you’re really desperate, you can just cut the leads off of some of the resistors you’re not using to make your own jumper wires, but in case you can afford another $10 kit, I’m recommending a popular kit from Amazon.

Once you’ve got all of your electronic parts, you can begin your assembly and write the code to the ESP8266 chip. More details on how to do all of that in an upcoming post, but for now, reference these couples of pages from the community.

I hope you found this information helpful. And finally without any further ado, here is your bill of materials, with all items available on Amazon.

Bill of Materials

  1. Micro-Controller (Wemos D1 Mini)
  2. Temperature Sensor (DS18B20)
  3. Gyroscope (MPU-6050)
  4. Battery (Panasonic 18650)
  5. Battery Controller (TP4056)
  6. Switch
  7. Wemos Compatible Breadboard
  8. Resistors
  9. Jumper Wires

Earl Dreams of IPA

When it comes to tasting beer, entirely too much emphasis is placed on the variety of hops used in the brew and in what volume those hops were used. About 5 years ago I had my very first Tree House Brewing Company beers. It was a Haze and I had never tasted anything like it. Although I had a very simple pallet back then (still do) I recognized that something was very special about the yeast character of this beer.

As the years went by and I continued to brew ever all kinds of varieties of beer, I quickly recognized that London Ale III was not the only yeast that THBC uses in their beers. There’s definitely something more going on in there that LA3 is not giving to this beer. LA3 is far too clean for the flavors you get from Haze.

One morning while eating a bowl of Cheerios with bananas, I was smacked in the face with a reminder of Haze. Immediately I realized I had to start brewing experiments to target this yeast profile.

Not alone in the pursuit, I borrowed greatly from those who’ve blazed this trail before me, especially the group on Homebrew Talk and Marshal Bishop over at Trinity Brewers.

So here are links to my first 4 experiments. Note that the grist and hops changed every time as I was just messing around with those ingredients, mostly based on what I had lying around. My real goal has been and continued to be, to isolate this yeast profile. I really feel like with experiment number 4, I’m starting to get close.