Tempeh incubator

How to make a cheap and easy tempeh incubator.

how to make a tempeh incubator
Supplies needed: 20qt storage bin, thermometer, weatherproof snap in socket, 40 watt light bulb, plastic junction box, electrical cord, non-breakable plate cover, and a dimmer switch.

Hardware needed: Wire clamp connector, 2 small bolts and corresponding nuts, electrical tape, wire nuts (often supplied with the switch), and 2-sided tape

Tools needed: Stanley knife, drill, wrenches/drivers corresponding to the bolts used.

Remove the entire knockout on the side of the junction box that will be secured against the bin. Remove the appropriate knockout on an end, size corresponding to the clamp connector. Secure the junction box with the bolts/nuts, such that the side opening is approximately centered. Use the Stanley knife to cut away plastic around the opening, slightly larger than the hole, itself. A drilled, 1.75 inch hole works as well. Note: During the trial run, it was obvious the bulb was too close to the beans, and so a shield was created with layers of aluminum foil and paper towel, folded into a flattened V around the bulb. While putting the bulb through the lid would be a more ideal position, I don’t think the heat of the bulb that close to the plastic is a great idea, nor do i like the idea of all the heat rising into the junction box. For this setup, to overcome this problem, creating a rack with an installed shield is an easy resolution. Caulk the opening around the base of the socket.

Snap the socket through the cutaway on the bin, and into the junction box. Use a screwdriver to pull the snap-springs tight within the junction box to ensure the socket is firmly secured (the two, metal prongs that snapped into place – stretch them slightly further).

Secure the wire clamp, and then insert the power cord.
Connect the power cord to the dimmer switch, and socket. Follow the instructions that came with the cord and switch, but typically, the power source will either have a black stripe, or be smooth. Connect the white socket wire to the white, or ribbed wire, the smooth, or black-striped wire to the black wire on the switch, and then, the black wire on the socket to the other wire on the switch.

Note that this is a three-pole switch, because it was the cheapest one in stock. There is an additional wire that is not used, and that is simply capped. Two pole switches are all that are needed and will have only two wires.

Use electrical tape to secure the caps.

Secure the power cord. In this case, Romex sheathing was used to hold the cord securely. Electrical tape can be wound around the wire to give it the girth to be safely secured.

Secure the switch in place. Note that any exposed metal should be covered with electrical tape. That is, the back-side of the switch plate, in this example, should be completely covered with electrical tape.
Secure the thermometer to the bin side using two-sided tape. Screw a bulb into the socket. Note: During the test run, our two-sided tape failed, and we were no longer able to monitor the moisture. Make sure the tape you use can withstand time, gravity and humidity, or devise a better way of securing it (i.e., put it on that rack that should be built to keep the beans off the bottom and hold the light bulb shield.) . Ours was rated for 10lbs, but obviously not.

The switch-plate cover may need to be modified (cut) to fit. Non-breakable covers are fairly simple to cut, by just scoring with a Stanley knife, cutting the curved ends, and then bending free. Make sure the back-side of the switch plate, beneath the cover, is taped off with electrical tape.

The incubator is now fully functional. However…

The test run found the cover became fairly warm to the touch, and so an insulating blanket was added to the lid, above the light bulb: Four folded layers of aluminum foil backed by, likewise, four folded layers of paper towel. The towel was secured to the aluminum with two-sided tape, and the entirety was secured to the lid with the same. The lid now remains cool to the touch even with hours of operation, at full power.
Initial observations are that at full power, the light bulb increases temperature within by seventeen degrees, at house ambient temperature between 64 and 72 degrees Fahrenheit. That is, at 64 degrees, it raised the temperature to 81 degrees, and at an ambient house temperature of 72 degrees, it maxed and held a temperature of 89 degrees. With insulation, results could be significantly improved.

One issue that arose during the test run, was runaway humidity, as i wanted to use water to help moderate the temperature. I still think that’s a good idea, but some or all of the water would be better sealed in a hot water bottle or shallow container that could be placed in the bottom.

Despite some poor planning, total cost was under $30, far lower than incubators available, dehydrators that cooked at appropriate temperatures, and yogurt makers, the same. Some instapots have yogurt settings that are cool enough and, likewise, if you have an oven with an incandescent bulb, you can use that, presuming no one thoughtfully shuts off the light, overnight…

Travails, fails, but nevertheless – success

There’s a lot to improve on with this incubator, and the process, but despite concerns about the humidity, the proximity of the bulb, and a really heinous, last-minute, last-ditch construct of a rack just before running out the door, the first attempt at making tempeh in this incubator did, in fact, produce two perfectly edible, if misshaped loaves of tempeh.

The setup kept the temperature very stable, starting out at 85F, quickly rising to 90F with full bulb output, then held steady at 89F at about three-quarters power. The first day it fluctuated between87F and 90F as we adjusted the dimmer, then, by the end of the day it had been honed in right where we wanted.

The laughable rack, made out of an old wire hanger, scrap wood, and painter’s tape:

As the boy would say: Oof. I’d hoped something around the kitchen would be suitable, but none of the cooling racks fit in the tub… So, this was the two-minute solution. Ugly and never to be used again, but it served it’s purpose – just to see if this contraption would actually work.

I don’t know if the proximity of the bulb would actually affect the end result, but after cooking beans in past attempts, this high-tech baffle was installed to protect the nearest corners. Initially, it was a flattened ‘V’, but unfolded and apparently levitated.

This was the low point of temperature and humidity. To combat the moisture, condensation was wiped from the sides, and one corner was propped open with a knife – straddling the corner to provide ventilation. Even with the lid partially propped open, temperature remained steady at 89F.

Improvements to be made are obviously creating a better rack, adding a ventilation valve to the lid, and more securely attaching the thermometer. Otherwise, this is a perfectly viable set-up for making tempeh, and much cheaper than moving to Indonesia.


All tidy and secure, now. With the shiny tray the cost neared $50, in total, but it cuts down on waste and even came with a little rack to hold it above the water. The thermostat is now securely hooked over the lip of the container.
And lastly, a fabricated heat shield that snaps over the bolt ends.

And, one more modification that was always the weak-point of this design: Heat loss. As temperatures plummeted, it became more difficult to maintain the temperature needed. That was resolved by simply tossing a towel on the top. A more permanent solution was added using the bubble-wrap envelope that Amazon helpfully provided with the starter, shop towels and duct tape:

All in all it’s worked very well. The first two trials with just the tray (Credit for that idea, here), seemed to dry out on top, and the top was not fully ensconced. There were many variables changed for the latest, so it’s impossible to say with certainty, but laying a layer of perforated parchment paper over the tray seemed to protect the beans from over-drying.

Additionally, i stopped worrying about humidity altogether, and it doesn’t seem to have mattered at all. I changed the starter i was using – and the beans: The latest using garbanzos. Whichever or many of the reasons why, this last one had the best cake of any. (Outside of the plastic bags.)

There are many sites out there with instructions and suggestions, and i’ve found none exactly meet my experience. The basics are there: Soak the beans, hull the beans, cook the beans, and dry them well.

Then, it’s 2 tablespoons of vinegar and a teaspoon (Or, in the current case one packet) of the starter mixed with the beans.

Temperature in the incubator should stay between 85f and 90f, and i usually aim from right in the middle of that.

Where the greatest variation occurs, is with the time it takes for the cake to fully form. Recommended times range from 22 hours, to four days! At 22 hours, my beans are still loose, and if i continue cooking them much past 36, I find the cakes start to spore.

Consistently, I’ve been concerned as i check the beans, that nothing’s happened. Then, by twenty-four hours they’ve solidified into a block. However, i don’t get a well covered block until closer to 36 hours (a little less, perhaps.) By 48, i start to get dark spots.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

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