Lower burr carrier design for reduced clumping

Grinders are one of the keys to exceptional espresso. Discuss them here.
User avatar
civ
Posts: 1148
Joined: 17 years ago

#1: Post by civ »

Hello:

Please bear with me ... =-)
My knowledge of grinders and grinding mechanisms is very limited and my descriptions will surely be lacking the proper terms, but I think you'll get the idea.

I would like to pose a theoretical question with respect to how a lower burr carrier works and how it shifts coffee from the grinding chamber towards the exit and the funnel / doser.

I have seen professional grinders produce and push out a great deal of ground coffee in a few seconds and the impression I had was that of the coffee being 'extruded' through the grinding chamber exit rather than 'thrown out', with the main consequence being clumping.

Most if not all of these grinders have lower carriers with three blades to move out the ground coffee, although I have seen web photos of lower carriers with as little as two and as much as four blades.

Let's picture a theoretical grinder with a 1450 rpm motor, flat burrs, the lower carrier having three blades, like the one seen in this photo of the lower carrier of a Mazzer Major which uses 83 mm burrs.


photo courtesy Espresso Parts

Now, if my math is correct, a three bladed lower carrier turning at 1450 rpm will try to push 1/3 of the coffee the burrs have ground in one revolution each time any one of the blades passes by the exit.

In this (usual) three blade scenario, this will happen 4350 times per minute or 72.5 times per second.

Maybe it is too much coffee at any one time trying to go through the door and this generates clumping?

Instead, let's suppose that the lower carrier has six blades instead of three.

In this case, the same thing would happen: the same amount of coffee would be ground in one revolution and pushed towards the outside but it would be 1/6 of the coffee ground in one revolution each time any one of the blades passes by the exit and this would happen 8700 times per minute or 145 times per second.

1/3 x 72.5 = 1/6 x 145 ---> it's the same amount of coffee ...
The difference is that it's less coffee pushed out each time by any one blade but more times a minute.

Of course, there's a limit to the amount of blades a lower carrier can have as I'm rather sure there's a point at which more blades will reduce the mass of coffee being sent to the outside, making the centrifugal force negligible and mucking up everything.

The question: would more blades in a lower carrier help in reducing clumping?

Thanks in advance.
Best regards,

CIV.

User avatar
another_jim
Team HB
Posts: 13871
Joined: 19 years ago

#2: Post by another_jim »

Your argument sounds good. The counterargument is that the coffee has only half the time to move down the chute before the next vane sweeps by. So your proposal would make the coffee grounds travel with more, smaller jerks, rather than fewer, larger ones, or, I would think, more smoothly.

So you're probably right. Common sense says that a smoother motion of the grounds would create less clumping. But it does need to be tested. Any ideas on how?
Jim Schulman

Advertisement
ira
Team HB
Posts: 5497
Joined: 16 years ago

#3: Post by ira »

Personally I'm of the opinion that you need to do the Versalab method, or have a sealed grind chamber and blow enough air through that the grinds don't build up. With my Macap M4D, it's useless, I just vacuum out 5 grams of wasted coffee every time I pull a shot more than 1/2 hour after the last one, On my Rocky I use a funny lid that lets me pump air through and get all but 1/4 gram or so out every time I grind. Clearly the Macap is better at grinding, but the Rocky for all it's flaws is so much more friendly.

Ira

User avatar
Jepy
Posts: 109
Joined: 19 years ago

#4: Post by Jepy »

Check the grind quality coming out of the Elektra Nino. I believe this grinder has 6 rotating fins

User avatar
shadowfax
Posts: 3545
Joined: 19 years ago

#5: Post by shadowfax »

Having owned a Super Jolly, Robur, and Nino, I'd say that I think that you may be a little misguided in thinking that grinds need to be "thrown" to minimize clumping. I've seen lots of grinders throw coffee (the Robur and SJ both do this before they fill their chutes), and while it does solve the clumping, it leads to some nasty static, which has issues of its own. A doser will solve this problem, but it can cause clumping itself if you don't use it properly.

There are 2 carrier designs that I've seen that I've been duly impressed with. First, the flat-burr Baratza Vario:


The Vario seems to keep its retention at next to nothing by making the grind chamber hug the burrset as tightly as possible

The Vario extrudes grinds for sure, and sometimes it clumps, but overall I think the design defeats the clumping inasmuch as the chute is so narrow that the clumps it extrudes are small enough aggregations that they still pack properly (or break up properly--It's hard to say which). In any case, I think minimizing the size of the chute and grind chamber are designs that the manufacturers of grinders with cavernous grind paths would do well to emulate.

The other grinder that's really intriguing with respect to burr carrier design is the one Jepy mentioned, the Elektra Nino, my current and only grinder.


These are the six fins on the Nino, and you can see how shallow they are as well. I think another key is that they're swept back.

Of course, the Nino's idea of swept-back, low-profile fins is not necessarily applicable to a flat burr grinder, if that's what you're interested in: the grinds are extruded from the burrset in a totally different place on them, necessitating (it would seem) a somewhat different design. Another thing about the Nino that I would highlight, though, is its lack of a horizontal chute and the use of a metal flap to direct grinds and minimize clumping and static (read more here:


You can see the Nino's flap here (Flickr link).


Here's how it mounts (Flickr link).

The grinds are basically extruded out of the space between the flap and the grind chamber opening; I suppose, again, that it defeats clumping by shooting them out so fast and from such a small orifice. That's not to say that the Nino never clumps. In fact, when I dose it often has a clump or two. They break with a tap or a touch, and the ones lower down seem to be broken up by the other grinds falling on them.

This is pretty typical of huge-burr grinders already, though. I haven't used a Major, but I assume it's a step up from the Super Jolly, somewhat like the Robur: clumping is only a problem if it leads to an uneven bed of coffee. So if the clumps are loose enough to break up on their own and distribute properly, and you get good extractions, it's not much of a problem, right?

Anyway, I hope that helps.
Nicholas Lundgaard

User avatar
civ (original poster)
Posts: 1148
Joined: 17 years ago

#6: Post by civ (original poster) »

Hello all:

Thanks a lot for the input.
Much obliged.
another_jim wrote: Your argument sounds good ....
But it does need to be tested. Any ideas on how?
Thanks.
I am thinking about having a new lower carrier turned using a large (very) valve recovered from an old ship's engine as a blank. It will hold 80 mm. burrs inside a 95 mm grinding chamber. Everything is OK until when you have to decide how many vanes.

As to your question, I think that a carrier like the one pictured above could be temporarily modified by adding another three vanes made from aluminium and held in place with some removable epoxy. They just sweep grounds and make no effort so it's probably OK. (?)
shadowfax wrote: ... you may be a little misguided in thinking that grinds need to be "thrown" to minimize clumping.
... lots of grinders throw coffee (the Robur and SJ both do this before they fill their chutes), and while it does solve the clumping, it leads to some nasty static ...
I think I was not clear in the idea I meant to convey, sorry.
Let me try to organise my thoughts. =-)

I have the idea that the 'extrusion' I have seen keeps the grinder from properly emptying the grinding chamber. Not a problem for a café dealing out many espressos per hour using a doser but when we want to grind 'on demand' (chute/funnel and no doser) to make ourselves one or two espressos it's a problem of sorts.

I think that 'extrusion' is a consequence of the grounds piling up at the exit and it probably happens because the 'rate of grind' (determined by rpm, burr size and bean input/second?) surpasses the amount of grounds that can slide through the grinding chamber's exit in a given time span without clogging it up.

When the exit clogs up and grounds start to get pushed out from behind, 'extrusion' occurs. When you stop the grinder you find that there is quite a bit of ground coffee still in the grinder, most of it outside the grinding chamber and along the exit path instead of inside the doser or the exit chute.

If the lower burr carrier hovers very close to the grinding chamber's bottom and the vanes move very close to the grinding chamber's border (~ 0.5 mm ?), it's quite possible that very little ground coffee will remain inside, but there will be quite a bit packed up at the exit along the path towards the doser or the chute.

With motor speed, burr capacity and exit path as fixed values, the variables would then be (?) the number of vanes and the bean input/second. My guess is that increasing the number of vanes (as in the Nino photos you have kindly posted) and controlling the bean feed may be one way move ground coffee towards the doser or exit chute in a more effective manner.

The 'throwing out' of the coffee grounds was more than anything to be able to send the grounds away from the horizontal part of the path and into the inclined/vertical chute or the doser in a more efficient manner. It's possible that the static (and part of the clumping) can be controlled with very thin wires stretched taut over the grinding chamber's exit.
shadowfax wrote: These are the six fins on the Nino, and you can see how shallow they are as well. I think another key is that they're swept back.
---
... clumping is only a problem if it leads to an uneven bed of coffee. So if the clumps are loose enough to break up on their own and distribute properly, and you get good extractions, it's not much of a problem, right?
Yes, I fully agree.
I like the six vane design, more would probably not work well. Their being swept back probably helps by making the most of the centrifugal force to send them against the perimeter of the grinding chamber with more force and then towards the chute. The low profile is probably due to the type of burrs used.
shadowfax wrote: Anyway, I hope that helps.
Yes indeed, very much so. =-)
Thank you so much for your input and for posting the photos and links.

Best regards,

CIV

User avatar
shadowfax
Posts: 3545
Joined: 19 years ago

#7: Post by shadowfax »

civ wrote:The 'throwing out' of the coffee grounds was more than anything to be able to send the grounds away from the horizontal part of the path and into the inclined/vertical chute or the doser in a more efficient manner. It's possible that the static (and part of the clumping) can be controlled with very thin wires stretched taut over the grinding chamber's exit.
This is part of the trick of the Nino: There is no horizontal "part of the path."


On the Nino, rather than drilling a hole into the side of the burr chamber, they cut across the opening they wanted to create to expose this (Flickr link).

Trying to engineer a better throw is, IMO, not going to work at preventing a chute from filling up. From what I have seen, even with very vigorous throws the static and/or volume of beans causes the chute to fill up. Part of the issue, I think, is that beans are very light. I don't believe you can shoot them out really fast and expect them to hold momentum. And putting wires over the chute is Mazzer's solution to static and consistency, and I promise you that an integral part of that solution is that the chute itself causes the clogging you're trying to solve.

I think the biggest issue with your theory is that what I have seen of coffee grounds flying out of very fast grinders suggests that virtually any bottleneck that you give them, be it a screen, a few wires over the opening, a horizontal chute--they will fill it up quickly and you will have to purge it or clean it out at every use. And if you eliminate all the bottlenecks, I think you'll have a difficult time with static control: Check out what Jacob did to his Robur a couple of years ago (link). he widened the exit chute and put it on an incline, and got huge amounts of static.

Another thread that is really relevant on the issue of screens is Minimizing Waste and Static in Large Commercial Grinders, one of the threads that led me to purchase a Nino. Some highlights:


The Robur burr carrier is much bigger than the Nino and the chamber retains much more coffee; this design is typical of Mazzer, Macap, Compak, etc. large burr carrier designs


Screens/wires just clog the chutes.
civ wrote:The low profile is probably due to the type of burrs used.
The Nino's sweeper vanes'/fins' low profile is partially due to the type of burrs used (as I mentioned), but as you can see from the Robur vanes above (same burr type), also rather unique in the grinder world. The Mazzer Major has similar fins to the Robur, as well. The difference is that the burrs is mounted lower in the carrier so that the fins rise to meet the ring where the coffee comes out of the burrs. It will also retain significant amounts of coffee, but I believe less due to the fact that its motor turns about 3x faster than a Robur's.

Anyway, I will say good luck to you--I'll be interested to see what you finally come up with for a design. I hope I've made my concerns a little clearer as well, or at least given you links some more discussion in other threads--they have very detailed photos, and are great reads for what you're looking into.
Nicholas Lundgaard

Advertisement
User avatar
Randy G.
Posts: 5340
Joined: 17 years ago

#8: Post by Randy G. »

The overall design of these grinders is a bit of a problem. Trying to use centrifugal force to move the particles horizontally as they are packed into a chute is problematic.

The Malykke Coffee Grinder, as seen at the SCAA Atlanta exhibition, and now (or soon will be) sold by La Marzocco USA eliminates the problem. Not that this idea is new. La Marzocco itself showed a similar straight-through design about two or three years ago, but it was evidently never developed. And, of course, the Versalab, which works well, but the company, unfortunately, seems to cater to people with as much patience as money. :wink:

The Malykke Coffee Grinder


The Malykke's internals
EspressoMyEspresso.com - 2000-2023 - a good run, its time is done

User avatar
another_jim
Team HB
Posts: 13871
Joined: 19 years ago

#9: Post by another_jim »

It's clear that straight down designs have fewer retention problems. They are the standard design for bagging grinders (The Malykke looks like one, not like an espresso grinder), but have always been difficult to engineer for espresso portioning.

The Nino design and Carlos's idea show that there's probably a good deal that can be improved in the conventional layout. Espresso grinder designs haven't changed since the 50s, since the procedures in Italian bars haven't changed since then. As grind on demand becomes more and more popular, and, given these are arrogant Italians, more legitimate, we'll see more thought out designs, like the Nino, rather than "just tell the idiots this is what they want" contraptions like Mazzer's E series.

I'm just hoping Compak or Macap rip off the Nino design at half the price sometime soon. :wink:
Jim Schulman

User avatar
shadowfax
Posts: 3545
Joined: 19 years ago

#10: Post by shadowfax »

another_jim wrote:I'm just hoping Compak or Macap rip off the Nino design at half the price sometime soon. :wink:
:lol: Right. Because the Compak K10 dosered grinder isn't already much more than half the price of a Nino... If only Elektra would get it through their heads that the US uses 110V grinders, and they are the primary market for single-portioning grinders, I bet they'd sell a hundred times more Ninos. Which, on that topic--my serial number is 090016 or something like that, and I got it in May. If I assume that "09" is a prefix meaning it was built this year, I imagine it's logical to assume I am the proud owner of the 16th Nino made this year. I realize that production might be low this year if they didn't sell as many as they'd hoped last year, but... wow, cash cow.
Nicholas Lundgaard

Post Reply