So, I've been talking about this for a while now, but it's finally happening. I've been thinking about this project off and on for over a year now. I'd do a little bit of research for components, research other builds, and just ponder all sorts of variations on how to build the thing so it works like I want.
I've been playing around with a bit of arduino stuff lately, and I've discovered that to be super easy to throw stuff together and get it to work like one would hope. So I finished iterating my physical design, so I thought I would get this thing built, then get to work putting the electronics and controls in it.
Background on the science of cooking
In the smoking foods world, there are essentially three different types of preparation: smoking, barbecuing, and grilling. The definition of each is specific to what's being prepared, but the general idea is that smoking happens at temperatures below the point at which the fat in the food renders down. Smoky flavors bind to fats very well, so to get the smoky flavor to stick to something, there has to be plenty of fat in the meat, and not melting and running out on to the floor. For fishes and cheeses, this temperature can be less than 90°F. For beef and pork, it can reach up to almost 120°F. This process has some history in food preservation. It tends to dehydrate the food, which makes mold and other bacteria less likely to grow. It concentrates the salts and makes foods tastier by adding that wonderful smoky taste. The final product, however, is usually not one that can be eaten without further preparation.
Barbecuing is the slow cooking of meats in above the temperature of rendering fats and connective tissues, but below the temperature of denaturing the proteins, at which point they shed their water and the meat becomes a dry tasteless piece of shoe leather. Beef brisket and pork shoulder are two meats that are prime candidates for this treatment because they are full of connective tissue and fat. If you slow cook them for many hours, the fats rend down to keep the meat juicy, and the connective tissues break down into sugars and other carbohydrates, which makes the meat fall apart tender and slightly sweeter. Pulled pork and other such dishes are made this way. They traditionally have a smoky flavor because they were cooked over wood fires, but the smoke is just a flavor enhancement, not a requirement for dehydration or food preservation. The final product is fully cooked and ready to eat.
Grilling is something we all know about. It's cooking at very high temperatures to sear the meat and seal the goodness in, and allowing it to cook up to temp from there.
So, grilling is usually a pretty short endeavor, and requires direct attention and action (flipping the meat, etc). However, both smoking and barbecuing require very little direct interaction and take a relatively long amount of time to finish. So the plan became to build a smokehouse that would be capable of both smoking and barbecuing with minimal interaction. The easiest way to accomplish this is to separate heat generation from smoke generation. Most commercial smokers use either gas or electric heat and some sort of smoke generator using sawdust or chips. They also tend to be made out of stainless steel, are heavily insulated, and cost a few thousand dollars, so that wasn't what I wanted to do. My idea was to still use wood as the heat source like traditional smoke houses, but be able to automate temperature regulation without having to baby sit the fire. The plan is to use a separate smoke generator as well, to make sure that I continue to provide sufficient smoke even when modulating my temperature.
In a traditional smoker, the chimney effect was used to create a draft from the fire to carry air (and smoke) through the smoker, but it required lots of tending. If the fire burns too hot, you can melt the fats in your smoked products and ruin them. If it gets too cold, the food can spoil. The traditional way to control this was to constantly add small amounts of wood to the fire.
My design ended up using two blower fans to move air about. One fan constantly recirculates the air inside the smoker because without the fire in the bottom to create a draft, it's hard to ensure that the smoke will fill the chamber effectively without the recirculation. It also has the added effect of basically making my smoker a convection oven. That should mean that when I'm barbecuing, I can do it at a lower temperature for shorter periods of time.
The second fan is used as an induced draft fan for exhaust. The idea is that I'll seal the smoker up air tight, but create one exit, where i'll put the fan, and one inlet, that leads to the stove pipe of an external fire box (a wood stove). If the box is sufficiently sealed, all the air being exhausted should cause the smoker to draw hot air from the stack of the firebox. If I modulate my exhaust fan, then I can very easily control the temperature inside the smoker, and all I'll need to do is just keep the fire burning well in the stove without having to very closely tend it to keep it at just the right fire level. When I'm not drawing air into the smoker, it will just exhaust out the chimney of the stove without overheating my food.
I ended up designing this thing in sketchup, which is a surprisingly good tool for designing stuff for woodworking. See some screenshots below.
The full design:
Here is the same view without the front:
Here is smoker with the right wall missing. I've sketched in the blowers to size. No electronics or fire box equipment is currently shown.
And here's another view with the same pieces missing:
Building the Box
To get to the cool part of making this crazy automated smoker, I first need to get myself a box that I can start playing with. THere are some pictures from construction of the box based on the above design. This took place mostly over a weekend.
The plan is to make a block foundation and use ground anchors to prevent it from being pushed over by operator error/wind. So we got started by making a base:
A friend was nice enough to donate a dozen 1"x12"x12' boards to use for the skin of this thing. I'm going to make a board and batten exterior for this thing, but I need a frame to mount it all to. So we marked out the frame on the base:
Then we made two identical frames, one each for the top and bottom:
Then we cut and drilled the uprights:
Then we installed the divider wall and assembled the frame:
The small section is to house the fans and all the electronics. The larger section is the smoker itself. The cross beams in the larger section actually show where baffle plates are going to be installed. The baffles are necessary to ensure that I get even distribution of the air in the smoker, so things get evenly smoked/cooked, even if it's loaded up.
Here's what my garage looked like after one day of working:
Sunday was a pretty short day. I had to help my neighbor with a few of his projects, then go to a superbowl party. We did manage to finish up cutting the pieces for the skin of this thing, and get started on the baffle plates. It kinda sucks making them, and I'm not exactly 100% certain that they are going to work right, but we'll see how it goes.
Fans and Ducting
Ok, so time for update the next. I've gotten the baffle plates finished and installed, so it was time to work on installing all the equipment that mounts to the dividing wall. The easiest way to do that was to tip it over.
Once it was all set up, I started with the recirculation blower and its associated ductwork. The way these fans are designed, they sit flush with no lower flange. So I got the recirc blower positioned.
Once I was happy with the position, I marked the location, cut the inlet hole, then drilled the holes for the mounting bolts.
Once I had the fan mount positioned, it was time to work on the duct work. My method for putting it together was to make a bottom piece to the duct work that would bolt directly to the outlet flange of the blower, then the verticals could sit directly on that piece.
Once the fan was mounted, I was able to attach the vertical pieces. I used pure silicone for sealing the duct work. Airflow through this thing is going to be tightly controlled, so I needed to make sure that no gaps would cause problems down the road. Pure silicone has a service temperature of up to 400°F and is FDA approved for incidental food contact, so it's the sealant to use.
The sealing was actually pretty important. I'm making this stuff out of the free rough-cut lumber that I was given. Some of it is pretty warped. I actually got something of a poor fit-up at the bottom. You can see here, along with the gobs of silicone.
Here's an interior view of the ductwork, looking down before the top was attached. You can see that the silicone made a nice bead on the inside as well.
I did get myself a pneumatic caulking gun to help out with all the caulking that will be needed to seal this thing up tight. All I can say so far is that it is totally worth the 23 bucks I paid to get it from amazon.
(Link in case anyone is interested)
Similar to the recirc fan, I mounted the exhaust fan. This fan is design to take air directly from the smoke chamber, so it's mounted a bit higher.
However, unlike the recirc fan, this one is not going to run continuously, nor will I need only one air flow from it. I need a way to modulate the exhaust flow, as that will be the way I'll control temperature. Exhaust flow should be (roughly) equal to flow that I can pull through the firebox. First I researched variable speed options. For single phase motors, there are a number of limitations on low speeds, mostly related to the starter circuitry that are required because single phase power does not provide constant torque. After researching this enough, I'm pretty convinced that I want to take a stab at making a phase converting variable speed drive for a motor in the future, but for now, I decided to just got with an output damper to modulate the airflow. I cut out some pieces and set it together to see what it would look like.
It looked like it was going to work out ok, so the plan was to take 1/4" steel rod, and drill a couple of holes to be able to bolt the axle to the plate. I also wanted to grind down the axle rod to have a reasonably flat surface so the plate would not rock around the curved surface.
Once I got that set up, I checked fit again:
Now, the trouble is I need some method to actuate that damper. I remembered I had a bunch of parts left over from mechatronics class way way back. I found a couple of servos and associated parts. I did some test fits and here are the results:
I was pretty happy with the way that all turned out, though I had a slight linkage problem, so I ended up needing to get some longer servo arms to avoid getting so close to a dead position. After getting those corrected, here is what I ended up with:
All assembled, I have a damper mechanism that was ready for attaching to the exhaust fan:
Smoke Generation System
With that completed, the next item for me was to create the box for my smoke generator. The idea for the smoke generation system would be to make a sealed box to house the smoke generator. I would supply it air via a small air pump, then exhaust directly into the lower baffle chamber, which would then get sucked up by the recirc fan and well mixed, then distributed by the baffle plates, so everything inside gets a nice even smoking.
Followup: That was the plan, but it didn't work well. I'm still including the construction information, but I don't really use this thing anymore. I did a calculation to estimate the amount of air I would need for combustion, then what I thought was a reasonable factor on top of that to make sure everything stayed lit. The reality was that the design did not give good mixing of the air in the box for combustion. It would burn ok in some places, but in others the fire would die out. What would end up happening is that I would not get good mixing in the box, so some of the air would just go by, and not take part in combustion. A mixing fan inside a slightly larger box max have helped.
So I started by building a box out of some leftover plywood:
I have it all attached via pocket hole screws. I'm a big fan of the pocket hole jig that I got:
Here it is set into rough position:
I cut a top to match, then routed a channel for gasketing material. Here the lid is open next to the box:
Out of an abundance of caution, I decided to line the box with cement board to present a non-combustible surface. I was on such a roll that I didn't stop to take any pictures of that process. I went ahead and took some leftover copper tube from my compressor project to distribute air inside the box:
For reference, the smoke generator itself is a stainless steel "maze" from A-maze-n products. You fill it with pellets or sawdust, then light it to start some coals. When you blow out the flames, the coals should smolder down the length of the maze, generating a nice, consistent amount of smoke. Here's a picture:
Followup Note: The A-maze-n products smoke generator works like a charm. I would easily recommend that to anyone.
Once the smoke generator box was mounted, I need to mount my air pump and connect everything up. I originally thought of putting it on a shelf, but last minute decided to just strap it directly to ducting:
At this point, all the large equipment is connected to the divider wall, so put the smoker upright again so I could test the smoke generator. Here's a shot of it all upright:
And here's a closeup of the equipment mounted:
And a closeup of the smoke generator system:
And at that point, I couldn't help myself, I had to connect everything and test the smoke generation system. Although the results are a little difficult to show on camera, here's what it looks like:
Followup Note: It looks good to start, but it doesn't keep up at that rate long term. Oh well.
On to skinning and smoking
The next part was to finish up the structure with a few tweaks, then put the thing on a foundation and skin it up. One of the first things I did was to put some safety grills over the fan inlets, so I didn't loose anything important in there.
After that, the structure was complete, so it was time to move the thing outside to skin it up before it got too heavy to move. In this case, I decided to go with a floating foundation. I removed the top few inches of soil and removed all the roots, then back-filled with a couple inches of gravel, then used blocks on the gravel to keep the thing up off the ground. The blocks were leveled out using the gravel below them. Once the level was set, the blocks were attached using some generic construction adhesive.
The foundation means that it has a nice place to rest, but I intended for it to stay there. To keep it in place, I bought a shed mounting kit that uses augur style earth anchors. Here is an picture that shows one anchor fully inserted, another partially inserted, and two others laying on the ground so you can see the full length of these things. They are about 24" long, so they dig in well.
To hold the smoker in place, I put threaded eyes into the base, and connected the eyes on the smoker to the ones in the earth anchors using turnbuckles. I tightened them down, and the smoker is so stable, I can't rock it from swinging from the frame.
Here is the overall view of the anchored frame at this point.
At this point, I was ready to begin attaching the skin. Because I wanted the smoke chamber to be air tight, I needed to board up the frame and seal all the gaps. To do this, I made each wall of the smokehouse by cutting the boards to length, then using the pocket hole jig to screw the boards together, with a bit of food-safe silicone between them. I didn't take any pictures of the process, but here is the image of the first one attached. You'll notice that it has a hole cut in it for the exhaust air.
I repeated this for three sides. The front was a little different. Because of the door, the boards along part of the front are cut into three sections. I used a router to make a 1/4" wide lap joint where the door would fit so that I would have room to put in a gasket. You can see the fit up steps here. A bit of a learning lesson for me. As the wood absorbed moisture and swelled, the gap that I left around the door eventually closed up. I had to take the hand plane out to the door and cut down on it a few times to get the door to close.
The door that covers the fans and electronics isn't screwed down, but it has handles and gate clasps to hold it in place. Otherwise, it's made the same way. Because I used some of the leftover boards for that piece, it's curved out a little. I didn't take many pictures as I was finishing up here. I added a roof and shingled it to keep the water off the flat top. At this point, it was a dumb box with no controls, racks, or gaskets, but I wasn't going to let that stop me from trying it out. I get some cheeses out, set them on cooling racks and got to smoking. This picture shows my batch of cheese going into the smoker. You can see the lap joint around the door better in this image as well.
And finally, here is an image from a later cold-smoke. I've added an EPDM gasket around the door, and toggle clamps to hold the door shut. You can see the roof at the top edge of the picture, and I've painted it to match the shed next to it. Finally, I've added shelf brackets inside. The brackets are made to accept wooden dowel rods. the idea is to either set metal racks down (as shown in this picture) or to be able to remove them to hang things while smoking. In this image, I'm using a MAPP gas torch to start the smoke generator. You use it to get some hot coals going, then blow out the flames to let it smoke.
At this point, it's a reasonably complete cold smoker. There is no heat source. There are no controls. I can put the smoke generator inside, open the exhaust damper all the way up, and go back a few hours later and reclaim my cold-smoked goodies. I've made many, many pounds of smoked cheese so far. It even makes an excellent Christmas present. I've also cold-smoked bacon that I've cured myself. I got busy with a few other things, and I stopped there with the smoker for a bit. The next hurdle was to work out the electronics and controls, as well as get a heat source together. I'll put the development of that in another wiki article. This one has enough pictures already.