Latest / Specialty Matcha Podcast / How is Matcha Milled?

Transcript

How is Matcha Milled? · Specialty Matcha Podcast

0:01
Hello and welcome to the Specialty Matcha Podcast.
0:03
My name is Ryan, This is my Coho Zhangjun.
0:06
Hello. Hello.
0:06
Who also goes by Sam and we're the cofounders of Sanko Matcha
0:10
Products. Yeah, we launched this podcast
0:12
to discuss our learning journey, Matcha, share startup stories
0:15
and interview experts. And today we want to talk about
0:19
how exactly do you melt Matcha? We have all these different
0:24
mechanisms of muni matcha, right?
0:26
We have traditionally, you know, ichi usus, and nowadays we're
0:30
seeing more industrial innovations on devices like ball
0:34
mills and jet mills. So there are many, many
0:38
different ways to melt matcha nowadays.
0:40
But you know, by the end of the day, they all share one common
0:44
similarity is that you are trying to make a 4mm by 4mm
0:49
Tantra into small powder. They are.
0:52
We recently got some emails and some messages from people ever
0:55
since we kind of announced the beta program for our mill that
0:59
can mill matcha fresh. That's a, you know, kind of has
1:02
a countertop level footprint. But a lot of people ask us
1:06
still, you know, they're still very interested in each USUS or
1:08
potentially ball milling. And we tend to get a lot of
1:11
questions about like what are the differences?
1:14
So in doing our own research and bringing our machine to life, we
1:18
had to learn a lot about other milling mechanisms and draw
1:24
inspiration from them. So we are absolutely biased in
1:27
this episode, but we also consider ourselves relatively
1:32
knowledgeable on the mechanisms and how it's built.
1:35
And you know, we're not going to necessarily try to convince you
1:37
to buy our mill on this podcast, but and we want to give a very
1:41
honest recounting of what are the mechanisms, advantages and
1:46
disadvantages as we have heard from other people who have
1:49
operated them between a ball mill, a cyclone mill and Ishiusu
1:54
or Ishiusu like. Products, yeah, I mean they, you
1:58
know, by the end of the day they all kind of powderize tensor
2:02
into, you know, small powders Frequently they are, you know,
2:05
in similar size to 5 to 10 microns.
2:09
But the powder morphology, the the shape of the powder often
2:12
times are quite different, you know, between different milling
2:16
devices because of their milling mechanisms.
2:19
And the amount of heat that gets generated at the interface,
2:22
that's true as the the milling is actually going on.
2:26
So it's a pretty interesting diverse landscape that can lead
2:29
to some very different flavor profile outcomes.
2:33
And a lot of judgement I feel like gets passed around here.
2:36
Like people generally say ball mills are worse and that
2:38
Ishiyusus are like the only good way to mail authentic matcha.
2:42
And there's actually quite a bit of nuance when you actually look
2:46
at the mechanics and the the science behind what is going on
2:49
and what does and does not make sense for certain applications
2:52
or for certain consumers or for certain types of tensha.
2:55
And it's actually a very nuanced conversation.
2:58
And they're actually both of their own arts because, like,
3:01
you can have very bad issues with milk matcha and quite good
3:05
for milk matcha like that. It's not a clear boundary
3:09
between these two milling mechanisms.
3:11
Only, you know, there's some sort of a blend between, you
3:15
know, these two, this two outcomes.
3:17
Yeah. We try not to bring too much
3:19
judgement into this and sometimes we do reference, you
3:22
know, some judgement on some of the past recordings that we've
3:26
done, probably because we had like just had a really
3:29
disgustingly ball milled matcha or something or really bad
3:32
Cyclone Mills matcha. So, you know, our our
3:35
experiences are also very much influenced to the tasting
3:38
references that we've recently had for the case of this
3:42
recording. We're going to be as objective
3:43
as possible because it's important that people understand
3:46
what this landscape is and the nuances behind it.
3:48
So absolutely. Why don't we start with sort of
3:51
the goals of matcha milling. You know, as you said, you're
3:54
turning this 4 by 4mm tensha particle, which is the standard
4:00
size for a tensha particle that's been de stemmed and
4:03
deveined and pulverizing it enough times to get between 5 to
4:07
15 microns, 5 to 10 Micron particles and what you really
4:11
want is under 20. Anything over 20 is begin to
4:14
really notice some texture difference and we've actually
4:17
heard the finer is not always better.
4:20
We were talking with some people who said if they have the
4:22
ability to mill below 4 microns and apparently it gets so gummy
4:28
and so sticky. Something that is very important
4:32
to know about matcha and matcha milling is that as you go
4:36
smaller and smaller in particle size, you're changing the
4:40
physics in the way that those particles interact with each
4:43
other. So if you have 50 Micron
4:46
particles, it's basically going to act like dry sand.
4:48
Those particles are heavy enough that gravity is going to have a
4:54
larger impact on the force than the static, than the cohesive
4:59
static forces that those two particles are having with each
5:03
other. So it's going to be like dry
5:05
sand. And the finer and finer you
5:07
mill, the more those particles are going to want to interact
5:10
with each other because the surface area to volume ratio
5:14
changes. There's so much more surface
5:16
area exposed and so much greater ability for those particles to
5:22
interact with one another that those static forces start to
5:25
overpower the gravitational forces.
5:28
And then it starts acting like what we all know matcha to act
5:32
like, this sticky clumpy substance.
5:36
And when you go eat below 5 microns apparently gets
5:39
extremely gummy where it's really difficult.
5:43
It clumps, it will really stop moving like a powder.
5:46
It sticks to itself really well into everything else.
5:50
It's basically turning into a non Newtonian fluid, right?
5:54
It's really turning into a gum. You don't want that to happen in
5:57
your mouth. So there is such a thing as too
6:00
fine. And as Umju mentioned there
6:03
particle morphology or particle shape also plays a really large
6:07
difference in its physical characteristics.
6:10
And what's really important is to understand that much is a
6:13
suspension. So when you mix matcha and water
6:17
together, you're not actually dissolving the the powder into
6:20
the liquid. You are doing some extraction
6:23
and you are dissolving some of the compounds in the matcha
6:26
which enter the water, which then become flavor active.
6:30
And that's, you know, really large part of the sensory or
6:32
taste experience. We have all drinking matcha, but
6:35
you also have these particles floating around.
6:38
And there's something called Brownian motion, perhaps you
6:40
remember from like an intro science class.
6:43
But liquids have this random movement.
6:45
So water is moving around in a very random way, kind of like a
6:50
bunch of smoke in a room. And you have this matcha entered
6:53
the system. And the higher the surface area
6:56
it has, the more encounters, the more attractive forces it's
7:00
going to have with the water. And the water is going to help
7:03
that particle dance, meaning it's not going to sink to the
7:07
bottom right away. So if you have very large matcha
7:10
particles, gravity is going to have a stronger force pulling
7:15
that matcha down. Then the water is in its random
7:18
motions, keeping it floating. Like imagine having a bunch of
7:22
little tiny ants helping you stay, you know, lifted up off
7:26
the ground. Very similar thing is what's
7:28
happening with the water. And the more surface area that
7:32
you have, the more those little ants or the more the water is
7:35
going to help you dance around in that fluid, meaning you have
7:38
a very high quality suspension. So typically they say that
7:42
Isiucis have very high surface area particle morphology,
7:46
meaning they form higher quality suspensions, which put in
7:50
another way, they don't sink as fast as ball mill, which tend to
7:52
be more spherical. And sphere is a geometry or a
7:56
shape that has the lowest surface area to volume ratio
8:00
that can exist in the universe, I believe.
8:03
I don't think it gets any more efficient in terms of particles
8:07
or matter forming a shape that has a very low surface energy.
8:12
So those are the goals of matcha mill.
8:14
And you want high surface area particle morphology and
8:17
particles that are under 20 microns because also the finer
8:21
you meld, the less the gravitational forces pulling
8:24
those particles down South. Zhongjun, let's talk about the
8:26
most historic way of how matcha is milled, the Ishiusu.
8:30
Can you talk a little bit about the mechanics that are going on
8:34
that an Ishiusu can do to convert tensha into matcha?
8:38
Yeah, for sure. So Ishiusu or stone melt is
8:41
basically you have two piece of very large granite, soft granite
8:46
usually that gets stacked on top of each other creating a pretty
8:51
large gravitational force from the top plate onto the bottom
8:55
plate. And in between the grind
8:57
surface, you have all these groups that cuts into a
9:01
different or a very specific pattern that forms a scissoring
9:05
effect to the tenta being grinded in between these two
9:09
very large stones. And depending on different
9:11
design, you either of you know, feed the tenta from the center
9:16
of the Ishiyusu or sometimes slightly off the center on the
9:21
sideway in the grind surface. These grooves are not
9:24
necessarily touching each other. In fact, they kind of form a
9:28
small slope. And as you getting closer and
9:31
closer to the edge, the slope gets narrower and narrower.
9:35
And finally they touch each other along the rim.
9:38
And that's how you get the tensile smaller and smaller.
9:40
So you know from each inch the tensile moves towards the rim,
9:46
they gets cut into finer and finer pieces.
9:49
And at the end you know the shearing force is so large and
9:52
they get powderized into very, very small powder.
9:55
And depending on the shape of the groove patterns, you
9:58
actually can end up having different shape of the matcha
10:02
powder or different size of the matcha powder.
10:04
And also depending on the groove and also depending on the the
10:09
weight of the stone and also the rotational speed of the stone,
10:13
sometimes you need to prime the tension as well.
10:17
Sometimes if feeding too large piece of attention into specific
10:21
issues. So it would really start in
10:22
between the groups. You wanted to be as exact as
10:26
possible that fits specific issues with designs.
10:29
Yeah, the feed rate becomes very important.
10:31
And we've learned from some people that work very closely
10:34
with traditional issues to milling operations in Uji that
10:38
if your tension is too fine and like it broke up too much, it
10:44
feeds too much in the mill too fast because the compact
10:48
efficiency, right? It gets more compact, you have
10:50
more tension and a smaller volume.
10:52
And it can actually form a lot of problems because so not only
10:55
do you have the concavity basically on one of the issues
10:59
who faces, like Zhongjun said, the slope, the distance between
11:02
those scissors, so to speak, gets narrower and narrower and
11:06
narrower until they're finally touching.
11:07
But you also have matcha powder pushing itself where you have
11:12
one particle bumping into another, bumping into another.
11:15
And that's how one of the primary forces of how matcha
11:19
migrates from the center of the mill all the way to the edges.
11:23
And if you don't have a very consistent feed rate, the
11:27
particle didn't have enough time or in a frequency to be cut
11:31
enough to make it all the way to the edges.
11:34
Yeah, it's not only the the the stones are driven by
11:39
gravitational force, the tension too.
11:41
For most of the traditional issues, there isn't necessarily
11:45
a feeding mechanism. It's really just, you know,
11:47
channel for the for the tension to float into the milling.
11:51
Chain. Yeah, absolutely.
11:52
And the rate at which it does it is very important, which largely
11:55
comes down to the size of the tension that you put in and
11:58
something else that's interesting.
12:00
The effect of the concavity or the fact that the scissors are
12:04
not necessarily touching each other in the very beginning
12:07
means that these particles are getting dragged instead of
12:11
perfectly cut. And that dragging really helps
12:15
to change that particle shape. And when it does eventually get
12:19
cut at some of those interfaces, right, the angle of that cut
12:23
also can have a real impact on the particle morphology.
12:26
And we heard that a lot of these curved design needs to used to
12:31
instead of having them be straight parallel lines, they
12:33
have curved ones that shape and the angle that that creates
12:38
generally produces much more spherical particles.
12:41
But it does also mean that it can move through and migrate
12:44
through the mill at a much faster speed where those that
12:48
angle of that those two grooves encounter each other also pushes
12:53
the matcha out laterally from the center all the way to the
12:57
edges, what will then be collected.
13:00
Yeah, for the more curved issues to pattern designed, it's really
13:06
less of a cutting by the, you know by by the end of the day.
13:10
So it's more like kind of like a rolling almost, hence a more
13:15
spherical particle morphology. And the other interesting thing
13:19
about issues too is the heat management.
13:22
So you essentially have two things generating heat.
13:25
You have the grind surface where the plates are touching each
13:28
other with some matcha in between as almost like a
13:32
lubricant but it's generating a lot of heat because of the
13:35
friction. And the other heat source is the
13:38
motor and you have usually 2 motor designed pattern in or
13:42
design types in issue uses. You have a top mount motor which
13:47
the motor is moving further away from the the stones instead of
13:53
touching them and you have the bottom mounted motor which is
13:56
directly mounted to the bottom plate and that's touching the
14:00
mill stones, thus transferring a lot of heat into the grind
14:04
surface. Depending on the two designed
14:06
and you can generate different shearing force, you can overcome
14:10
different torques, but also you have to keep in mind that you
14:13
know the heat generation and heat management is very
14:16
different from these these these different designs.
14:19
And frequently in order to save energy and maintain largest
14:24
efficiency for these macho grinding facilities, the motor
14:27
is designed to be as weak as possible, just enough to
14:32
overcome the the torque they need to return the issues.
14:36
So if you have like a poorly managed attention structure or
14:40
you end up getting some stamp to start in between the grind
14:44
surface, the issues was stopped and you have to, you know, open
14:48
it and clean it up. Yeah.
14:49
So Zhongjun for anyone looking to buy an issue.
14:52
So what are some of the operational complexities of
14:55
owning 1? Well, the first complexity is
14:57
definitely learn how to use it and how to assemble it ambled it
15:02
because I would assume that you are trying to use it outside of
15:05
Japan and you know, the whole thing usually are not really
15:08
shipped in one piece and and also after usage, you know,
15:14
issues you really needs to be. We finished with this process
15:19
process called medate, which means the grooves needs to be re
15:23
sharpened by Japanese stone Mason because these are soft
15:27
granites and you know, and when you have them touching each
15:29
other, mealing in constant speed, the grooves are
15:32
essentially going to wear down or wear out, you know, after
15:35
some usage. So that needs to be.
15:40
Especially if you don't know what you're doing, which would
15:42
be most first line in Ishiusu? So usually you, you would end up
15:47
needing to shift the, the plates, which are hugely 50
15:51
kilograms per piece or per set back to Japan to have them
15:54
refinished or, you know, you go to Japan and learn how to do it.
15:58
That might take, you know, a few years, maybe a one or two
16:01
decades. Or a lifetime.
16:04
I don't know if anyone would ever teach you.
16:06
Yeah. Good luck finding a teacher.
16:09
Or you instead of shipping the plates, you ship the sensei, the
16:13
stone Mason to your country. The more interesting or exotic
16:16
the location, maybe you get better luck.
16:19
Like if you're trying to buy one in Paris, maybe it's a more
16:22
attractive vacation destination somewhere else.
16:26
Yeah, yeah, Yeah. Well that's really the most, I
16:30
would say complicated part and cost costly part of operating a
16:35
issue to outside of Japan. And but you know, like these
16:38
machineries are, are large and frequently you need a rewiring
16:44
of your, of your electricity to, you know, turn it from A2 phase
16:48
to a three, three phase electricity set up in order to
16:53
power this, this drying machine. It's quite energy consuming.
16:57
Indeed, and typically you need to have an air conditioned room
17:00
where you're maintaining the facilities.
17:02
Yeah. To have very specific humidity
17:06
is anything that's too humid is going to Tensha loves humidity.
17:10
By the way, if you leave some tensha out in a hot room or hot
17:16
in humid room, come back in two hours, it will be almost soft.
17:19
It's really loves humidity and it will really mess up the the
17:24
matcha milling operations. The other thing too is that when
17:27
you have matcha being milled in an ishiyusu, one of the ways
17:31
that you know that it is working is basically it snows.
17:34
Matcha comes out of the interface, right?
17:37
It it doesn't fall out. It like staticky snows out of
17:42
the mill and it does form a very temporary suspension in air.
17:46
So if you look at any of the matcha milling facilities in
17:48
Japan, everyone is wearing a mask.
17:51
And because long term exposure to any type of particles in the
17:56
air is really unhealthy long term.
17:58
I I'm an amateur Potter and there's something called
18:01
Potter's lung and that it used to be back in the days for
18:05
people doing pottery, clay actually can be ball milled and
18:09
clay particles are super fine. So when it dries and gets on the
18:12
floor and gets everywhere, you know, as you walk or you're
18:15
moving around the studio, it gets kicked up into the air and
18:19
really poorly cleans. Studios are actually pretty
18:22
large health risk for people. And it used to be after working
18:26
a couple decades as a professional Potter, you get
18:28
something called Potter's lung from all basically the crap, all
18:32
the particles in the air collecting in your lungs over
18:35
time. But it's a very true, a real
18:37
concern for macho milling operations.
18:40
So you need to make sure it's very well contained.
18:42
And we've heard that people installing some air conditioners
18:46
in those rooms, after one month of usage, the filter within
18:50
those air conditioners is like basically trashed.
18:53
And then you can imagine what it's doing to your lungs.
18:55
Yeah, it's pretty bad. And it's pretty obvious.
18:58
Everything is covered in grain. Like all of these powders just
19:01
flied around, you know? And while usually these issues,
19:06
these, you know, in those industrial issues, you stay, you
19:10
end up buying from Japan, they are kind of contained in a large
19:14
case like a like a, like a in casing, a glass in casing.
19:18
So it doesn't necessarily fly out that much.
19:21
But still, you know, when swapping the plates out, you
19:25
need to open the in casing, you need to deal with all of the
19:29
residual powder inside the in casing and those needs to be
19:32
cleaned occasionally. And that's a that's a lot of
19:35
work and it's really not a one person job.
19:38
Like these millstones are very heavy, like it's a very hard for
19:42
one person to lift them up. So you need at least two people
19:46
to operate and clean and maintained the operation of
19:50
issues. Or one very jacked person like
19:52
can lift by AGH with confidence. Dwayne Johnson.
20:01
Yeah, actually there's a really interesting photo if you Google
20:03
I, I think it's the one of the best photos that encompasses
20:06
this. If you Google kettle, like the
20:09
tea company in New York, matcha Mill, there's a wonderful photo
20:13
of the matcha collection process where the basically this is ISI
20:16
yusu covered in matcha everywhere and they're taking a
20:20
dry paintbrush trying to kind of collect it.
20:23
So collection is also really a big headache.
20:26
Typically there is a collection drawer where the matcha will
20:30
fall and where most of it accumulates, but there's
20:33
probably a good 20% that doesn't make it in the drawer that is on
20:38
the side of the stones or, you know, somewhere else where you
20:42
have to do quite a bit of manual collection and a lot of
20:45
cleaning. Yeah, to maintain these Ituso.
20:48
Yeah, it's no joke. Or it's really operationally.
20:52
And not to mention they're really hard to purchase.
20:55
Yeah. So I think 2022, I was in the
20:58
market for one. I was at the time thinking of
21:01
starting a cafe. Thankfully, the stream faded
21:03
away from all the cafe owners. I've heard it's no joke.
21:08
It's very difficult endeavor. Thankfully, we moved into matcha
21:12
hardware. But, you know, one time I was in
21:13
the market for one. So if you happen to be in the
21:15
market for one and these operational complexities don't
21:19
scare you away, there's kind of two ways you can do it.
21:22
You can go to the broker route. There are a couple different
21:24
issues you brokers, but what I did is I hired a translator on
21:28
Upwork who spoke Japanese and I asked them to help me research.
21:33
So basically just Google searched this in Japanese for
21:36
the different issue Yusu vendors and he was very kind.
21:39
He offered to reach out on my behalf for quotes.
21:42
I think he had to fax one of them saying that, you know, on
21:46
my behalf, I was interested in buying an issue.
21:49
Yusu. He collected all the materials
21:52
from me and then all the cost breakdowns.
21:54
So if you are interested in buying 1 and that's probably the
21:57
cheapest way to do it, then avoid the broker route.
22:00
Just find someone who speaks Japanese and that's able and
22:03
maybe is located in Japan who can like literally call them or
22:07
fax them because a lot of them work very old school and you'll
22:10
get quotes and then then you'll have to deal with shippings and
22:13
customs and stuff. I was just talking with someone
22:15
who's importing one to Europe a couple days ago, a couple weeks
22:19
ago, something like that very recently.
22:21
And something else that you should be aware of if you're
22:24
pursuing this is that shipping rates are totally negotiable.
22:28
So he was quoted €4000 shipping cost for one stone mill.
22:35
Oh my gosh. And I think it's a bottom motor
22:39
operated, but he has a very good relationship with a DHL broker,
22:45
or maybe it was UPS or FedEx, I forget which one.
22:47
And he was able to get a better rate by going with his own
22:51
broker because he's importing and exporting tea all the time.
22:54
So because he's built up the business, he was able to
22:56
negotiate it down to €2000 and his broker kind of took care of
23:00
the logistics. So if you do buy one, know that
23:03
the shipping price is not the final price.
23:05
And if you already have a relationship with shipping
23:07
people that they might literally be able to cut it in half and
23:10
and $2000 is still quite a lot. But those options do exist.
23:15
If you're looking to pursue it. You can send us a Commission
23:19
check for this tip. Yeah.
23:23
But no in all seriousness, they're they're very cool
23:26
marketing statement like it really there is nothing like an
23:30
issue. So in terms of coolness.
23:31
Yeah. The the footprint, you know,
23:34
it's very, usually very large system sits in inside your cafe
23:38
shop. And it's really a statement, you
23:40
know, it like people who are aware of the operational
23:44
complexity of operating such device and you are still willing
23:49
to do so. It's like really a bold
23:51
statement of, you know, I'm willing to overcome a lot of
23:54
headaches just to get quality products.
23:57
Yeah. At 20 to 40 grams of matcha mild
24:00
per hour, if you amateurize out the cost, it's a it's a bold
24:07
statement too. It's a big investment.
24:11
The prices range, you know for the really tiny tiny motorized
24:17
units we've heard as low as 3 or 4000 sometimes even less.
24:22
But them could become very skeptical at that point.
24:25
For any like large scale commercial milling, the price
24:29
range, not including shipping is going to range anywhere between
24:33
6012 thousand depending on the unit, the manufacturer.
24:37
And that's if you can buy one. There's actually a huge issue
24:40
shortage at the moment, which is caused by the matcha shortage
24:44
and all of these gigantic blending houses and tea
24:47
companies and tea manufacturers also need them.
24:49
But they have buying relationships with these issue
24:52
shoe makers that go back centuries sometimes.
24:54
So it makes sense that they're not only the largest accounts,
24:57
but the oldest. So they'll probably get
24:59
prioritized first. So we've heard waiting lists
25:02
that are longer than a year. Maybe you get lucky and buy a
25:05
used one. It's another path to buy one as
25:08
well. Yeah, there's a wait list for
25:11
purchase, and you know, there's also a wait list for Medate.
25:15
So after some usage, you want your stone mill to be
25:18
resharpened. And yeah, good luck finding a
25:21
stone Mason to do that. And in Japan and also like you
25:25
know, we have heard that the wait list can be 6 months to a
25:28
year for the refination. Too, And those are for some of
25:31
the more established players, not someone that just bought 1
25:34
randomly and who knows if they'll buy more.
25:37
It's, you know, the repeat customer aspects certainly will
25:41
prioritize you in the queue in terms of that business
25:44
relationship that they have going on because they're at
25:48
capacity. There is so much pressure on
25:51
this system that you know, it's makes it very hard for new
25:56
entrants trying to become matcha Millers using these traditional
26:00
equipments. Even with established players,
26:02
we've heard they can't find stonemasons to repair half of
26:05
their milling capacity by December.
26:07
And there, that's a domestic player.
26:09
Yeah. It's no joke.
26:11
And so it's really, and you know, I, I like to say we, we
26:15
should also mention how sustainable realistically is it
26:18
that you have coffee shops and tea companies all over the world
26:21
sending 50 kilograms per pair back to Japan and then shipped
26:26
back to wherever country they're from?
26:28
Like this is from an environmental standpoint, we
26:31
should think twice about decentralizing matcha milling
26:35
when you're milling in an Ishiusu and the carbon footprint
26:38
that that can have, especially if you don't know what you're
26:41
doing, which would be most people buying one for the first
26:43
time. But they're probably going to
26:45
need to get it refinished much more frequently than someone who
26:49
has decades or centuries of experience being the traditional
26:53
players. And a lot of the traditional
26:54
players have their own, like in house stone Mason to do these
26:59
types of repairs and to make sure that everything is working
27:02
properly. Yeah, very difficult.
27:04
But, you know, with the current match equipment set up, Ishi USU
27:09
is really the only, so far the only viable option for, you
27:14
know, cafe commercial milling. Because for the other things
27:18
we're talking about ball mills and, you know, jet mills, and
27:22
they have a much, much, much larger footprint in terms of,
27:26
you know, how much space. They're very loud.
27:28
Yeah. And they're super loud.
27:30
Ryan, you want to talk about the milling mechanism for a ball
27:33
mill? Yes.
27:33
So ball mills are interesting and they're used in a lot of
27:36
other industries. So the production of them is
27:40
much more commodified. You have a lot of vendors that
27:43
make ball mills for a wide range of industries ranging from
27:47
ceramics manufacturing to all types of food processing and
27:51
mining and all sorts of things. Pharmaceuticals.
27:55
So they're very common. Cosmetics.
27:57
Pharmaceuticals. Yeah.
27:58
Autumn Cosmetics to get these super fine powders that get
28:02
turned into paste. And so the technology is very
28:05
well developed, much more industrialized, much more
28:08
commodified, and therefore they're very readily available.
28:12
The ones that we have seen being used for matcha milling are
28:16
generally quite large and there's two types.
28:18
Well, first I'll talk about the grinding mechanism.
28:21
So they're essentially a tumble Miller.
28:24
So you have literally a bunch of balls which kind of bump into
28:27
each other, and then you have a system where they're rotating.
28:32
So think of a giant drum with some agitation, kind of like a
28:36
clothing washer or dryer, and where it's spinning around.
28:39
You have all of these balls bumping into each other.
28:42
Then when they bump into each other, there's some shearing
28:45
force and it can subdivide particles.
28:48
So it's a rotating system. You put in the T and it rotates,
28:52
rotates, rotates and depending on how long it can rotate for
28:56
then you can get different particle sizes.
28:59
And then what happens is is everything gets dumped out.
29:02
The balls get cleaned and then it gets put back into the system
29:06
and you separate out the powder and then maybe you sift the
29:09
powder later on. There's two types.
29:12
There's batch based and continuously fed ball mills.
29:16
So the batch based one is exactly the name implies you
29:20
have this whole system, it's tumbling around together.
29:23
We've never seen a batch based system that has an active
29:27
cooling system, but every batch baseball mill we have seen has
29:31
been in some kind of walk in refrigerator.
29:34
So it's a more of a passive cooling system.
29:36
It's put in a very cold environment, which helps keep
29:38
the whole system because as you can imagine, a whole lot of
29:41
friction is being generated. The continuous ones, the
29:44
migration time and distance. The distance is about two to
29:48
three meters or you know, between I guess 6 to 10 feet.
29:51
And we've heard the migration time can range anywhere from 10
29:55
minutes to 20 minutes. As that tea is fed in, it gets
29:59
goes through the ball mill and then it comes out, you know, the
30:02
other side as a fine powder. Those that we have seen are
30:06
typically water cooled, so there's an active cooling
30:08
mechanism making sure the whole system doesn't overheat.
30:11
And we've touched the sides of these units and they're like
30:14
actually physically cold to the touch.
30:16
Not quite cold enough that it would condensate water, but it
30:19
is like definite really cool to the touch and matcha comes out
30:23
the other side. Ball mills have a really large
30:27
range of quality. There's some really well
30:30
operated ball mills where they're very skillful and they
30:33
can taste very good and actually can have a very nice texture.
30:36
The particles do tend to sink faster, but in a latte format
30:41
the the suspension qualities of a matcha depend on the viscosity
30:46
of the liquid, how thick that liquid is that it enters, and
30:50
then something like a latte, it's much more viscous and so
30:53
the suspension qualities are higher for all particles that
30:56
enter that system. So ball milled matcha does not
30:59
necessarily mean bad, but there's a huge quality range and
31:04
typically no one takes mediocre tensha and runs it through an
31:09
ishiyusu. So there's also kind of a
31:11
survivorship bias that we need to think through when comparing
31:16
ball milled with ishiyusu milled because default it's always
31:20
going to be better going through an Ishiyusu because it's such an
31:23
inefficient method. Yeah.
31:25
And also, you know, on the other hand, no one really puts very
31:28
high quality tensha through a ball mill.
31:31
So it's very hard to compare or make a very objective comparison
31:36
of the two milling mechanisms because people just don't do
31:40
that. Yeah.
31:41
So typically these units are very loud, they consume a lot of
31:45
electricity. They almost always need passive
31:48
or active cooling systems which need to be taken into account
31:52
for installation. The one great benefit of using a
31:56
ball mill is that you can melt anything just like you can mill
32:00
like clay particles you can put in Hoji cha stems.
32:04
So basically all Hoji cha powder that's like, you know, matcha
32:07
like and consistency are all milled on a ball mill because
32:11
each uses cannot handle those woody stems, takes too much
32:15
force to break those particles apart.
32:17
Yeah, but it's basically impossible to operate one in a
32:24
cafe shop. It's maybe in the future, yeah.
32:28
And yeah, so it really is a pretty big investment.
32:31
And for behind the scenes operations, like for some coffee
32:34
shops or something like inside your roastery could totally be
32:38
viable. It requires its own set of
32:40
unique skills and probably a lot of investment.
32:43
And the craft of milling it, to our knowledge, is not that
32:49
accessible. I've never heard someone that
32:53
casually talk about the differences in the milling
32:57
parameters that they use that lead to a better outcome gets
33:00
really deep infrastructure. So the knowledge especially in
33:03
English to improve and that learning curve that might be
33:06
significantly more difficult than something like an ish USU
33:10
where there are some Western vendors that have some
33:12
experience using them. Or maybe not necessarily
33:15
because, you know, there's definitely a shortage of, you
33:19
know, public knowledge, but for, but anyone that we have seen
33:24
that, you know, uses a Bob mill in Japan, they are like super
33:29
practical people. They're not, you know, these
33:32
very traditional stonemasons in Japan that doesn't necessarily
33:36
want to talk about, you know what they know.
33:38
I I bet you can just hire translator and go visit these
33:41
these factories and they will probably just tell you how to do
33:44
it. Good point.
33:46
There's very little cultural baggage.
33:47
You're not going to offend anyone by ball milling A matcha
33:50
and asking a bunch of questions in the same way that you would
33:53
with an issue. Yeah.
33:55
And also it's easier to maintain versus a issues.
33:59
All you need to do is just to switch the balls depending on
34:02
your usage, you know, anywhere from one month to three months.
34:06
And these are usually ceramic balls.
34:07
So they're not, you know that hard to purchase or very or not
34:12
very expensive either. Yeah.
34:14
Absolutely. You want to talk a quick bit
34:17
about cyclone or jet milling and then we can talk about our mill.
34:21
Yeah, yeah. So jet milling is a less common
34:24
but also very widely used milling mechanism for, you know,
34:28
very large macha productions. So what it's essentially doing
34:33
is that it's blowing tensha or whatever that you end up milling
34:37
with the milling device in a very fast bean of or a jet bean
34:42
of air. And these errors will blow
34:45
against each other. You're basically making all
34:47
these, you know, tensha constantly colliding with each
34:50
other with, you know, managing the airflow inside the mill air.
34:54
So you know, you're just, you know, it's just really a
34:58
accelerator that, you know, accelerates, you know, these
35:01
tensha particles and make them collect against each other.
35:03
And after they get collide enough of times, they got turned
35:08
into smaller and smaller powder. And there's usually a
35:12
centrifugal kind of powder collection mechanism on the
35:16
bottom of the the jet Miller to kind of collect the fine powders
35:21
and then, you know, push them into a large bag.
35:25
So with these, the outcome is similar to a bought mill.
35:29
I haven't actually read anything talking about the mythology of
35:33
these jet mills. Do you know anything about it,
35:36
Ryan? I think I've seen some paper,
35:38
but I forget what the outcome was.
35:40
To my knowledge I think they're also spherical, but I don't
35:43
know. We just haven't had that much
35:44
experience with them. They seem to be quite a bit less
35:47
common in the matcha industry. Yeah, yeah.
35:50
And also with that you can just meld everything.
35:53
So depending on the the the model or the make of the jet
35:58
mills, some of the powder collection mechanism is kind of
36:01
like a vacuum that you suck out of the light enough powder in
36:06
the collection chamber out and transport it into a bag.
36:09
So if you're having some, you know, stems which are heavier in
36:14
the tensha, those won't necessarily be able to, you
36:18
know, the the air won't be able to carry them in the first place
36:21
to make them collide. And then after they fall down to
36:25
the bottom of the milling chamber, they won't be
36:28
necessarily able to be sucked up, so to speak.
36:33
So they would just kind of sit there and they need to be
36:36
cleaned. Once while there's some, there's
36:37
some self sorting that goes on in those machines.
36:40
It's super interesting, Yeah. The particle has to meet some
36:43
very specific specifications before it can even exit the
36:46
mill, which is a pretty clever way to do it.
36:49
And something we should say about these.
36:51
I mean, these are very industrial ways of milling.
36:53
I think they can do upwards of 10 kilos an hour or something
36:56
like that. And similar with the ball mills,
36:58
I've heard everything from like like 2-3 all the way up through
37:03
10. Really depends on the size and
37:05
usage. Yeah, it's a lot faster.
37:08
These jet mills also generates a lot of heat, so they need to be
37:12
active actively cool too, usually with the hydro cooler.
37:16
Yeah. So that's the landscape of
37:18
matcha milling options and so do the complexity.
37:20
And obviously the as Anjun said, buying a ball mill or a cyclone
37:24
mill is like really back end food processing infrastructure.
37:28
So it's the type of investment you should think about if you're
37:31
interested in getting one to mill matcha and doesn't
37:34
necessarily mean bad these industrial ways.
37:37
And actually it's very important that they exist because I've
37:40
heard about 10% of the matcha produced in the world is Ishiusu
37:43
Stun Ground. And I don't know if that's an
37:45
outdated statistic, but if tomorrow these ball mills and
37:49
cyclone mills vanished, then you can really complain about matcha
37:52
shortage. There just would not be enough
37:58
for the world at an appropriate price point too, for any of that
38:02
to make sense for the industry to grow.
38:04
So they're a really, really, really important part of the
38:08
matcha economy and matcha milling infrastructure.
38:11
Because I would also not assume that if you're buying from a
38:13
good Western vendor that your matcha is Ishiyusu milled unless
38:17
it explicitly says that, which really goes to show that you can
38:21
have some very good quality most of the time coming out of a ball
38:24
mill. And so there's a lot more
38:26
complexity, even though that it does in theory produce a lower
38:31
quality outcome. If you care about suspension
38:34
quality, which you know, if it's going into a latte, it literally
38:38
doesn't matter as much because it's entering a more viscous
38:41
fluid. So a lot of complexities here.
38:43
Yep. So, so if we trying to assess
38:48
the situation right here. So if you were ever wanted to do
38:51
fresh milling in your location, EC user is actually really only
38:56
the viable option, you know, unless you have a larger
39:00
warehouse behind your facility that can hold a jet mill or a
39:03
bomb mill. So our position really right
39:06
here is trying to, you know, reduce the operational
39:09
complexity while maintaining, you know, the powder structure
39:13
and the quality of the powder and it can be fit in a modern
39:17
cafe. Yeah, Oh, absolutely.
39:18
So we set out to design A matcha mill that could know much of
39:22
fresh anywhere and everywhere. So you can consider this portion
39:25
sort of an ad, but we also want to be very objective about it
39:28
and we detailed all of this data.
39:32
We sent a lot of matcha off to a lab.
39:34
We compared it with benchmarks. We have a really in depth
39:37
analysis comparing the differences between ball mill
39:41
ECU to a few other matcha milling mechanisms that are
39:45
countertop that we didn't talk about here is really just one,
39:49
but we'll we'll save. We won't mention their name
39:52
here. We laid it all out on a big
39:54
table and kind of wrote an essay comparing each one to our matcha
39:58
mill and what we were doing. It is a link only article.
40:02
If you're interested in it, please e-mail
40:04
ryan@sanco-chaw.com or message us on our website or Instagram
40:09
anything and we'd be happy to send you that link if you're
40:12
interested. But some do you want to do a
40:14
quick quote UN quote sales pitch of what we have created and the.
40:18
Features that it has. Yeah.
40:21
So when designing this machine, we thought through all of the
40:24
benefits and shortcomings of different milling mechanisms and
40:29
we took a lot of inspiration from traditional ishi usus,
40:32
which de facto produces the best quality form of matcha.
40:37
But we also thought about the difficulties when operating, the
40:41
issues, you know, the heat management, the large footprint
40:44
and how can we overcome all those difficulties and reduce
40:48
that into a modern machine that's more fitting to a, you
40:53
know, cafe consumption scenario. So we designed this countertop
40:58
Miller, we call it the Sanko Usu one that we use AL 95 ceramic.
41:03
It's basically, you know, powderized sapphire and we
41:06
centering that into two pieces of very, very dense and hard
41:10
millstone. And instead of using
41:13
gravitational force, we use a Sprint mechanism to increase the
41:18
pressure between the two millstones in order to powderize
41:21
the tension into very, very fine matcha powder.
41:24
And we installed a active cooling system that's the most
41:29
challenging, turns out to be the most challenging.
41:31
An interesting part, we took a lot of inspiration from the
41:37
computer industry which we use the thermoelectric cooling
41:42
system to actively cool the machine and to take the heat out
41:46
from the milling chamber, out from the machine, out from the
41:51
milling environment. It's like cooling a very hot
41:53
CPU. So if we ever in the future
41:56
share the internal some of the internal design, you will be
41:59
able to see all these interesting electric wirings
42:02
that is dedicated to the heat management of the machine.
42:07
So we also thought about the the the powder collection which has
42:11
been a headache for traditional issues.
42:14
So all of the powder is contained in a very small
42:17
milling chamber and there is actually a difference of air
42:21
pressure between the milling chamber and the outside
42:23
environment. So all of the powder kind of
42:26
gets a bloat and pushed out from the milling chamber and with
42:30
addition scrapers that you can actually get a very large
42:35
percent of the powder out from the Miller.
42:38
The machine actually only would eat about 5 grams of attention
42:43
when it starts. And after a starts and gets
42:46
running, there's no no much of being sacrificed anymore.
42:50
So it's not really a percentage retention, it's really just 5g.
42:55
So that's all of the design consideration that we put into
42:59
the the machine. And with usage, you can set the
43:03
RPM of the machine, so you can actually tweak how fast the
43:07
rotational speed of these grindstones inside the machine
43:11
to kind of control the fineness and also control the output of
43:14
your usage. So if you are so mainly milling
43:18
for matcha latte and you don't necessarily need very, very fine
43:22
suspension, you can actually increase the speed and you know
43:27
you can have more output, but you have to think about the heat
43:31
being generated from the machine too.
43:33
So if you increase the speed, it's going to definitely
43:36
increase the heat. The cooling system does have a
43:41
Max capacity, so up to a certain point it won't be able to
43:46
necessarily suck out enough heat from the Miller.
43:50
So you are going to have hotter and hotter matcha if you turn
43:55
the speed really fast. But you know, that's up to your
43:59
decision to make. So for the current machine, the
44:03
matcha output is 35 to 80 grams per hour, which is about 3 times
44:09
faster than the traditional Ishi Usu.
44:11
But you know, we have tested different speed with the same
44:15
temperature and it does have a impact on the flavor of the
44:18
matcha doesn't necessarily have the impact on the fineness and
44:21
the color, which is very interesting.
44:23
One thing you can do is change the pressure.
44:25
So if you want to reduce the amount of heat generated but
44:29
increase the speed at which you're milling matcha, you can
44:32
decrease the pressure from, let's say Max pressure down
44:36
slightly, especially for a latte.
44:39
If you need to go much faster, where you don't care so much
44:42
about suspension quality, that is an option that does become
44:45
available to you. Although we're hoping that 80
44:49
grams per hour might be fast enough for most cafes,
44:53
especially slow bars who are making who can mill fresh to
44:57
order. Yeah, yeah, yeah.
44:59
It's there's a lot of freedom for you to tweak.
45:02
And we kind of actually see this as a, you know, a very kind of
45:07
organic culture that we hope to build in the future is that you
45:11
and us actually will be working together to try to find out the
45:15
best parameters for different usage for your for your cafe or
45:18
your facility. Our real goal is to kind of
45:21
close the gap for people that want to do fresh milling but
45:24
can't or shouldn't invest in an issue.
45:27
So because it's heavy, it's slow, it's operationally complex
45:32
to collect the matcha and to operate it and to clean it.
45:35
And you got to spend lots of time, money and risk sending
45:41
your millstones back to get resharpened if you can find
45:44
someone to get it. So they're hard to buy, they're
45:47
very heavy, they're complex to maintain, they take up a lot of
45:50
space. They require like electricity,
45:54
rewiring, it's a lot of upfront cost and a lot of upfront risk.
45:59
And if you drop one of the mill stones, you've lost your
46:01
capacity to mill unless you have a backup 1.
46:04
So like our machine, for example, has swappable mill
46:07
stones. So we're trying to fill the gap
46:09
for people that really want a fresh ground product and have
46:12
control over how the matcha is being milled to as close as
46:17
possible. Standard as an Ishiyusu.
46:19
And we did a bunch of lab testing and I would say we're
46:22
right around the mark of the average Ishiyusu.
46:25
We don't have the last SIP experience that you would get
46:28
with a really super high end Ishiyusu milled Macha like Tenju
46:33
from from REQ Kalyan or or in the Reno from Horatio Shuman.
46:40
Like those have like perfect last ship experiences that are
46:43
super creamy and smooth, but it shows up in the data.
46:46
But I'd say we're better than most ball mills that we've ever
46:49
had. And I'd say about the same as an
46:51
average Ishi USU where you kind of feel some stuff on the last
46:55
sip and which shows up in the data that we've also presented
46:58
on that site. So we're trying to fill that
47:01
gap. We don't want to be an Ishi USU
47:02
replacement, but we want to give people an option who want to
47:06
mill matcha fresh. But like literally based on the
47:09
economics of their business, it just makes no sense at all to
47:12
buy ECU. So because of the labor and the
47:15
risk and all of the upfront costs.
47:17
Yeah, our plates are much smaller and and they're actually
47:22
quite durable. So AL 95 has been used for a lot
47:26
of different milling purposes in different industries.
47:29
You see that in the mining industry, in the Pharmaceutical
47:32
industry, and these things are super hard.
47:36
They're basically sapphires, and we promise.
47:39
Yeah, the same material, so. Yeah, we promise.
47:41
Technically, it's different infrastructure.
47:44
Yeah, Yep. We promise 12 month, you know,
47:48
quality guaranteed, but you know it.
47:50
Depending on the usage, these things usually last for three to
47:53
five years with very little compromise on the performance.
47:57
Yeah. But the nice thing about them
47:59
being swappable is we can design different patterns in the future
48:03
too. So not, definitely not going to
48:05
guarantee three to five years, but we you might want to get a
48:08
different pair to get different outcomes, maybe in efficiency or
48:12
particle shape or maybe even we can design one that can be well
48:16
adapted to Hojicha, we don't know.
48:18
So we're going to do a bunch more experimentation, but this
48:21
can almost be a platform to mill matcha and when we can give much
48:25
more control over to the users so they can have a fresh ground
48:28
product or so many benefits from these perspective and a
48:32
nutrition perspective. Or if you have your own designed
48:36
and manufacture capacity, design your own.
48:38
These are totally swapable and it won't, you know, voidable
48:42
warranty. Absolutely.
48:44
Well, I think there's a great place to stop.
48:46
If you're interested, please e-mail us and let us.
48:48
We'll we'll send you the link. And we're running a beta
48:51
program. We're only making fifteens for
48:53
all early access for the first commercially produced batch.
48:57
Just super exciting. And those slots are slowly
49:00
filling up. We haven't officially opened it,
49:02
but we have a lot of guesses so far.
49:04
Please do e-mail us and let us know that you're interested so
49:07
we can add you to the list. And we're gonna actually start
49:11
having real finalized signups in September of 2025.
49:15
And yeah, I think that's a great place to stop.
49:18
Yeah. Hope to talk to you soon all.
49:20
Right, If you enjoyed this episode, please consider sharing
49:22
with the Front and we'll see you on the next one.
49:24
Thanks. In the future, bye bye.