How to make Soy Sauce

This article sets out to explain how to make your own soy sauce, also called Shoyu in Japan.

First off, make sure that you have read the article which explains what Koji needs to grow. In it, you’ll find 80% of the information you’ll need to make your own Shoyu.

Pressed, finished shoyu

What you’ll need

Soy sauce is made of soft wheat and soy beans (and water and salt of course). There are two kinds of wheat berries readily available, durum and soft wheat. Make sure to get soft wheat (it’s the kind that bread flour is made of).
As for the soy beans, I have been getting them from a local farmer who toasts them. I think that had a great effect on the taste of the final sauce.

You’ll also need some equipment, an oven to roast the wheat in, and something to mill the wheat berries with. The wheat needs to be milled as coarsely as possible. Ideally it just breaks into three to four parts or so. I’ve never managed to achieve that and I ended up with much finer stuff, and I think that’s ok. I‘ve always used my coffee mill set to the coarsest setting, which worked well enough. I started out with a small food processor, but working in such small batches got boring quite quickly :)
A temperature probe for keeping track of the koji’s temperature is very helpful! If it has an alarm it’s even better.

The ratio of soy : wheat

You’ll probably wonder on how much of each ingredient to use.

Usually I go with one part soy (soaked in water) and one part wheat (dry), by weight. This is more wheat than is traditionally used. More traditional would be one part soy (dry) and one part wheat (dry).

Generally you can say that, the less wheat you use, the longer the sauce will have to ferment in order to get a satisfactory taste. Shorter term shoyus, like shiro shoyu (meaning white soy sauce) are made with more wheat than soy, and they get a balanced flavor within a shorter time.

In all likelihood you have heard of tamari shoyu, which is made only with soy beans. It’s a matter of taste, but in my opinion miso or shoyu only made from soy are a bit lacking. This is because wheat brings in a lot of carbohydrates, which are converted into manifold other compounds.
By introducing wheat, you’ll get a good amount of lactic acid produced by salt-tolerant lactic acid bacteria, which is important to balance out the saltiness of the sauce. The bacteria themselves also produce other compounds that improve the aromatic profile of the shoyu.
You will also get some alcohol, which is produced by salt-tolerant yeasts. This alcohol is then converted further into esters and probably some acetic acid too. The yeasts themselves also produce a good amount of esters, just like in beer. Possibly at some point you’ll find that your soy sauce smell has inclinations of wheat beer. These are the esters you are smelling!

Wheat also has a different composition of amino acids compared to soybeans. This is probably also a reason why soy sauce made with wheat and soy has a more balanced taste than a soy sauce made only with soy.

The Practical Steps

There are quite a lot of steps, for this reason I broke them up into five sections: “Preparation”, “Cooking”, “Inoculation”, “Incubation”, “Fermentation” and the „final steps“.


Do these things the day before:

  1. Soak your soybeans (toasted or not) in plenty of water. Don‘t worry, you can‘t „oversoak“ them.
  2. Roast the wheat berries in your oven until they are golden brown. Most likely they are going to pop a bit. This is when you know that they‘re soon toasted enough.
    • This step creates a lot of Maillard compounds, the same kind of compounds that people love about grilled meat, fries, even coffee, and generally any browned food (except caramellized food, that is a different set of reactions).
    • I once roasted the wheat until it was dark brown and tasted almost like coffee. It was great, but unfortunately in the process you are likely to get high amounts of acrylamide, which is a carcinogen, so don’t overdo it!


The day you are going to start your Shoyu Koji:

  1. Start cooking your soybeans.
    • In a pressure cooker I’d say it takes about 15 minutes of full pressure until your beans are done. Make sure not to depressurize the pot too fast, otherwise your beans will split open and you end up with a lot of mush. I have started to cook them in a normal pot, so I can check on them more often. They are done when you can squeeze them easily between your fingers. In a normal pot that’s easily going to take 40 minutes, possibly more.
    • With all things Koji, usually it is best to steam your substrate. Here it‘s different! It‘s best to cook your beans, because you are going to mix them with the crushed toasted wheat. The wheat is going to soak up any excess moisture from the beans, and you will then have a substrate with a good moisture content for growing Koji!
  2. While the soybeans are cooking, coarsely crack the wheat in the meantime. Use a mill in the biggest setting or a food processor.
  3. When your beans are done, strain them in a big sieve.
  4. Now you can mix the cooked soybeans with the coarsely cracked wheat.
    • If your mixture seems too wet you will need to add some more of the wheat. Probably you will not feel like roasting wheat again if you don’t have any roasted wheat anymore. You can use some grits instead.
    • You can crush some of the soybeans with your hands, but it’s not absolutely necessary.
    • If your mixture is too dry, spray some water onto it or some more beans.
    • Judging the correct amount of water in a substrate for growing Koji is a tricky thing, which is why I recommend starting out with rice Koji to learn the craft. If you steam your rice, you are bound to get a perfect substrate for Koji, and you will learn about the correct amount of water a substrate needs.


  1. As with any other type of Koji, dilute your spores so that they are easier to handle. Typically I sieve out some of the flour that’ll inevitably occur when you crack toasted wheat. That flour can then be used to dilute the spores. Use either the soy sauce spores, the mild soy sauce spores, or A. Sojae spores.
  2. Check if your soybean/wheat mixture has cooled down to 40°C. At this point you can start to spread the spores over your mixture. A tea strainer is very helpful for this process. Mix well.
  3. You can now put your mixture into your muro (incubation box).
    • I have found that a great way to grow this Koji is to use a deep dish (like a pyrex or a gastronorm dish) and:
      • Put a moistened (not wet!) piece of cloth inside your dish, then put your mixture on top of the cloth. Then put another moistened piece of cloth on top.
      • Put cling film over the dish. Put some holes into the cling film so that the Koji doesn’t suffocate (I once didn’t make any holes, you could really smell the CO2, the Koji wasn’t happy!).
      • I like this technique, because you will achieve a really high relative humidity inside the dish and the koji will have an easy time to grow. If you already have a different setup for growing Koji, don‘t worry and use your setup.
  4. Let the Koji grow.
    • The strains for soy sauce are quite different to the miso strains. The soy sauce strains grow like crazy and they overheat very easily, so keep a close eye on the temperature!
  5. After 24h, stir the mixture. Put the cling film back on top, maybe moisten the cloth again.
    • If you find that your Koji is overheating you will have to stir more often.
  6. After 36h you should see a lot of growth. If your Koji didn’t overheat (i.e. it didn’t rise over 37°C for a long duration) it shouldn’t have sporulated.
    • Sporulation is a tricky topic with soy sauce Koji. The growth is so vigorous that everything happens much faster than usual. You should check on your Koji regularly. If you notice any dustiness, yellow or green spots, stop immediately and put your Koji into the brine (more on that later).
    • If you find that your Koji is heavily sporulated, don’t panic. You can still use it, it’s just not as good as it could be. If a Koji that is meant for miso sporulates, I don’t use it anymore, the resulting miso is too bad. But if the Koji is meant for soy sauce, I found that it’s acceptable to use it. Just make sure not to breathe in the spores when handling the stuff. It is going to be very dusty. Either hold your breath or get a mask.
Spreading the spores over the mixture of soy and wheat
Spreading the diluted spores over the mixture of soy and wheat
Diagram of the temperature of the Koji and the Muro. The temperature of the Koji rose to 39°C while the temperature of the Muro stayed at around 30°C.
Here you can see just how vigorous soy sauce Koji can be. Within just 14 hours of inoculation, the temperature of the Koji (green) rose to 39°C! The yellow line is the temperature in the muro, which also was heated up by the warmth the Koji produced.


  1. Prepare a 15% brine.
    • At this point you will have to guess how much brine you will need.
    • Use 176 g of salt per 1000 g of water. This will result in a 15% brine.
  2. Put your Koji into a well-cleaned vessel in which you want to ferment the shoyu.
    • The vessel may be glass, ceramic or stainless steel. I guess you could use food-grade plastic, but alternatives without softeners are easily available, so I wouldn’t use plastic.
  3. Now pour the brine over the Koji until the Koji is covered. Let it soak for one or two hours. Probably the brine level is going to recede as the Koji is soaking up some of the water, so pour over some more brine until the Koji is covered again.
    • Honestly I don’t know if the big producers are using more brine than I am suggesting here, maybe they do. But I figured that if I am going into the trouble of making my own soy sauce, I want to get the tastiest stuff I can make, so I use as much Koji as the brine will be able to accommodate.
  4. Your shoyu is now ready for fermenting.
    • In the first 2-3 weeks it needs to be stirred daily. Otherwise mold or kahm yeast will start to grow on top, the first produces toxins and the latter tastes awful.
    • Your shoyu is going to be very active in this time. The bacteria and yeasts are going to produce a lot of gas, and because of that the solids of your shoyu are going to float on top. You need to push the solids down again, otherwise they will get moldy.
    • Once your shoyu has settled down and is less active, you can start to stir only every second day. Basically, the more aged it is, the less important it is that you stir the sauce.
What it looks like in the beginning…
… and after 6 months.
Notice how the single grains have totally dissolved. They have been “digested” over the course of several months.


After your moromi has fermented for 2-12 months it’s time for harvest. You can use the fermented mash (called moromi in Japanese) just like that, or you can filter and press it. I like to use it unfiltered, I have found that it tastes much fruitier.

For pressing you will have to improvise a bit. I used a big sieve and a piece of cloth. Put the cloth into the sieve, pour the moromi into the cloth, fold it and put something heavy on top. Let it press overnight. Oh and don’t forget to put the sieve over a pot to catch the shoyu ;)

In the morning you can go ahead and press the cloth a bit more, sort of like when you wring out a towel.

If you have a cider press or something like that, count yourself lucky :) Just pour your moromi into layers of cloth (maybe even bags of cloth) and then go ahead and press them.


Make sure your bottles are well cleaned. It doesn’t hurt to sterilize them in the oven – just put some aluminum foil over the opening and keep them at 160°C for half an hour, let them cool down overnight. Also boil the caps in some water.

Heat your Shoyu to 90°C. The heating is important to kill off any yeasts or molds. If you don‘t boil your shoyu, you will get a really yeasty taste after some time. By yeasty taste I don‘t mean the esters they produce, but the taste that you get when the yeast cells die and autolyse. It tastes sort of like biting in a cube of fresh yeast.

Pour your Shoyu into the bottles and then close them with the caps.


Of course you do not have to use wheat and soybeans. You could also use, say, oats and lupines. Or Einkorn and chickpeas. Go crazy :)

However, I always recommend to start with the tried and true before starting experiments. If something goes wrong, you can be sure it is not because of the ingredients you chose.

Thanks for reading! I hope this article is of help to you! If you have any questions, please ask in the comments section :)

This Post Has 7 Comments

  1. Great timing on the soy sauce article! Here’s my feedback:
    * Mention explicitly whether the 50:50 ratio of soy/wheat is by weight or volume (I assume weight).
    * “a manifold” -> “manifold”
    * Is the lactic acid you mention wheat providing due to the activity of lactobacillus or is it some other reason?
    * Consider pulling the example math for 15% brine into a footnote.
    * Same for the note on sporulation.
    * Consider breaking “The practical steps” section into several subsections: Perhaps “Preparation”, “Cooking”, “Inoculation”, “Incubation”, “Fermentation”. It’s really long, and a bit tricky to navigate in present form.
    * Is aerobic fermentation recommended for the last stage, or, since I have airlocks, would it make sense to use them?
    * Did you inoculate with any yeasts or lactobacillus for the last fermentation, or just rely on atmospheric?
    * Why did you boil the soy sauce at the end?

    Higher level note: What I love about your other articles is that the describe not just what to do, but help build an understanding of the why, which helps the cook recognize and correct any issues, makes it easier to adapt the recipe to constraints, and provides a more fulfilling home cooking experience. This article goes into a lot of detail on what to do, which is great, but more insight into the “why” would make it much more informative.

    1. Thanks a lot for your input, it is highly appreciated! :) I will work through the list to bring this article up to par.

      * Mention explicitly whether the 50:50 ratio of soy/wheat is by weight or volume (I assume weight).

      yes, by weight

      * “a manifold” -> “manifold”


      * Is the lactic acid you mention wheat providing due to the activity of lactobacillus or is it some other reason?

      yes, it’s due to the action of various salt-tolerant lactobacillus.

      * Consider pulling the example math for 15% brine into a footnote.

      good idea

      * Consider breaking “The practical steps” section into several subsections: Perhaps “Preparation”, “Cooking”, “Inoculation”, “Incubation”, “Fermentation”. It’s really long, and a bit tricky to navigate in present form.

      That was one of my concerns, will do that for sure :)

      * Is aerobic fermentation recommended for the last stage, or, since I have airlocks, would it make sense to use them?

      I think it is important to submerge the stuff that floats on top every now and then, you can’t do that without introducing air (at least not without getting quite technical,..). Probably the introduction of oxygen does bring the fermentation forward, or if not, I am sure there’s no harm.

      * Did you inoculate with any yeasts or lactobacillus for the last fermentation, or just rely on atmospheric?

      I’ve done both. Generally, with shoyu you can easily rely on atmospheric, but I have found that it’s nice to add a bit of miso to ensure a balanced “ecosystem” of microbes.

      * Why did you boil the soy sauce at the end?

      To stop any fermentation. The concern is mostly with yeasts and molds. The former keep on working and that can lead to an unappealing yeasty taste. Mold is very much unwanted, too, since you can’t be sure if it’s Koji or a toxin-producing cousin.

      Thanks again :)

      1. Thanks very much! I soaked my soybeans last night and am going to start steaming today! I’ll be using the soy sauce koji I got from you awhile back. Now that I have a precisely controlled incubator, growing koji is extremely easy.

  2. I tried this, and it has been incubating for 24 hours. One thing that would be helpful is more discussion of getting the moisture level right — I tried to follow this as is, with 50:50 wet soy to dry grain, but for some reason it ended up too dry, and I had to add some moisture at the end. I *think* I got it reasonable, but I have barely seen much growth in that 24h. I know it’s an art not a science. One thing is that I steamed my soybeans, I did not boil them. I think that maybe if I had boiled them, and they’d soaked up more water, the dry toasted flour would then have put the moisture into a good range. The recipe does mention steaming as a way to cook them, but I think that it may require moisture corrections.

    Anyway I think it will probably be ok; I added a (very small) bit of water and covered with a moist towel, and the soybeans did have a bit of white fuzz on them. But it’s not like the last batch of mixed grain I did (for black vinegar; soy sauce koji) which was fully covered in 24h with thick white fuzz.

    I’ll update how it ends up going if I remember.

    1. Maybe 16 hours after I added the additional moisture, and the flour is covered in mycellium. A few of the soybeans do have green spots of sporulation, but I think it should be ok. Not bad for a first attempt. Probably shouldn’t have started with 1kg, but it’s easier in bulk.

      For anyone else reading this: Let me just reiterate how fast the soy sauce koji grows…you really, really need to keep an eye on this stuff!

  3. Bread flour is made from hard wheat. Cake and pastry flours are made from soft wheat.

    1. Probably there are regional differences, around here in Austria soft wheat is used for bread.

Leave a Reply