Super flat rice polishing – A Daishichi special

The level of polishing or milling our rice has a major impact on the characteristics of the sake we produce. The characteristic difference is a function of the amounts of proteins, fats and oils that remain in the rice used for brewing. The more of these that can be removed the more refined the sake. This is why daiginjo has a higher milling rate or lower seimaibuai than ginjo. Sense the proteins, fats and oils have the highest concentrations on the outer portions of rice with steadily dropping concentrations the further from the outer edge you get, going from 90% seimaibuai to 85% has a greater impact than going from 60% to 55% seimaibuai.

However, the methods used to mill our rice determines how effective the milling process is at removing the proteins, fats, and oils without wasting the starch we need. Imagine that the level of concentration declines linearly1 with distance from the surface. Then removing a square unit of the outer 1% will remove more proteins, fats and oils than removing a square unit of the second 1% (from 1% down to 2%). Based on this we would want to mill our rice, such that, we uniformly remove the outer most layer down to the desired seimaibuai. The method to do this is called the Flat Rice Polishing method.

While the Flat Rice Polishing method seems ideal, Daishichi has developed a method that is a step better. It is called the Super-Flat Rice Polishing method and works better because it focuses on the amount of high concentration material it removes. To better understand this let’s back up and consider four different ways of, or goals for, milling rice. These are:

  • The conventional method
  • The grain shaped method
  • The flat rice method
  • The super-flat rice method

The first of these is the conventional method and is used for almost all rice milling today.  It is optimized to mill the rice as quickly as possible while keeping rice cracking to a minimum.

Modern sake rice polishing proceeds by circulating rice being milled passed a milling stone. The milling stone spins around a vertical axis with the rice dropped through a chamber around the outer edge of the milling stone where it may contact the stone and have some small part removed. The rice is then circulated back up above the stone where it will again be dropped. This pattern of circulation past the milling stone can continue for days depending on the amount of material to be removed.

The conventional method proceeds with the milling stone turning at a very fast rate and controlling the amount of rice in the milling chamber to allow for considerable free motion for each grain of rice. This results in the rice having a tendency to fall through the chamber with its long axis being horizontal and hence the most likely parts to come in contact with the milling stone are the edges and the ends which results in a rounder milled grain. The following illustration shows how the milled rice changes in shape as milling proceeds. The outer purple represents the original shape of the un-milled rice while the inner blue shape represents the evolving shape of the milled rice as more and more of the rice is milled away. In each of the three milling stages shown, both a front and side view of the grain is given to help you imagine the full three dimensional shape.

Conventionally Milled Rice

As can be seen in this illustration, the distance from the top of the original grain to the top of the milled grain is changing much more than the distance from the side of the original grain to the side of the milled grain. Another way of saying this is that the percentage of milling along the long axis of the grain is higher than the percentage of the milling along the width or depth axes.

Second is the grain shaped milling method. With this method the goal is to reduce the grain by the same percentage on each axis. While considered to be the ideal method for some time, it still has a tendency to remove too much material from the longer axes. Say the long axis was twice as long as the depth axis, then removing the same percentage from the long and depth axes would mean that we are removing twice as much material from the long axis than from the depth axis. While this method is more balanced than the standard method it does not do as well as the flat rice method we mentioned above.

Third is the flat rice method that we discussed above. In this method a smaller amount, percentage wise, is milled from the longer axes than from the shortest axis. However the milling distance from the original rice surface is equal for every axis. The following illustration shows how the material that is milled away remains the same thickness all the way around the milled rice that remains; that is the purple ring around the milled rice is the same width no matter where you measure it.

Flat Rice Milling
Flat Rice Milling

Finally, the fourth method is the super-flat rice milling method developed by Daishichi Sake Brewery. In this method the goal is to remove more material from the flat part of the rice than the edges or the ends. Doing this results in more of the unwanted material remaining and leaves the milled rice flatter and longer proportionally than the flat rice method.

For the most part the way we get from the conventional method to the super-flat rice method is by slowing down the speed at which the milling stone spins around the vertical axis and increasing the amount of rice falling through the milling chamber at any point in time. The added rice in the chamber causes more of the rice to fall with the long axis in a vertical position. This prevents the ends from striking the mill stone more frequently than the other surfaces.

Daishichi has found that using their super-flat rice polishing method to polish rice to a seimaibuai of 70% removes the same amount of unwanted materials (proteins, fats, oils, ash and minerals) as milling rice using the conventional method to a seimaibuai of 58%. So using the super-flat method saves 12% of the rice material while removing an equal amount of undesirable material. One drawback of using this super-flat rice polishing method is that it takes longer. It can take three times as long to get down to the same seimaibuai or twice as long to get down to the equivalent level of unwanted material being removed.

How is that for scratching at the surface? 🙂

  1. I don’t know whether the rate of diminishing concentration is linear or some other non-linear function.
Click to Enter a Comment

Leave a Reply to dWiGhT Cancel reply

Your email address will not be published. Required fields are marked *

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

5 thoughts on “Super flat rice polishing – A Daishichi special”

  1. So where would you get larger quantities of milled rice? Interwebing shows this as really the only provider of milled rice but in small quantities. “: (

  2. Thanks for setting me straight on the rice situation. In fact I was using pure-white dinner rice, when I said 100% I thought I was referring more towards the grain size rather than colour. I basically just took the rice out of the bag and washed it, and went from there. The sake I made and am making now are more towards a doburoku than actual sake. Also, thanks for the aricle links and the dryer tip. I think I’ll give it a try!

  3. What a great article! However, for me it is very technical. I do not own a rice mill yet so my sake is 100%, or whole-grain if you will. In lamans terms, how do polish 100% rice? Also, do you of any methods to crush/mill rice by hand?
    Thanks!

    1. SakeNewbie,

      100% would be brown rice. White rice that you get most places is somewhere around 90% to 93%. So if you used white rice you were not using 100% rice. That aside, I have two articles on personal rice milling:
      https://homebrewsake.com/2010/08/01/milling-your-own-rice-with-the-twinbird-mill/, and
      https://homebrewsake.com/2010/10/03/a-pictorial-look-at-the-twinbird-rice-mill/

      I have heard one person say that you can put the rice in a bag and let it go in the dryer with the heat off. It seems like this would take too long. However, I haven’t tried it so I don’t really know.

Leave a Reply to dWiGhT Cancel reply

Your email address will not be published. Required fields are marked *

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