tekomino wrote:As is jumping to a conclusion that:

1. There is difference in flavor

2. That hopper full of beans results in better flavor than single dosing

I'm not concluding. I am intuiting and then betting on it. But no conclusions.

These 2 points above, are based on my estimation that the bimodal for the particle size distribution will be different on a single dose. It is also based on my estimation that burrs design and grinder design are optimized to have a preferable bimodal, and that preferable bimodal is designed to show up at normal working condition in a commercial environment where the hopper is kept at least half full most of the time.

Based on the above, moving away from those condition will change the bimodal. And changing the bimodal will move the burrs and the grinder from their target design optimal point.

Of course, if someone will do a comparison using those $$$$$ lab equipment and we'll see both graphs it will be illuminating.

LaDan wrote:I'm not concluding. I am intuiting and then betting on it. But no conclusions.

You lose; send whatever money you bet to Coffeekids

I've made the comparison for the K10, the Robur, and all the other TGP grinders. Moreover, these grinders' particle size distributions were compared (albeit with the inadequate laser sizing). There is no difference. The grind adjusted to get the same flow is the same physically by the current lab standard and by taste. The change to the finer adjustment compensates for the reduced feed rate of single dosing. Higher feed rates result in finer grinds, since the beans grind against each other as well as against the burrs.

Again, higher feed rates == finer grinds. If single dosing is worse than hopper grinding, then hand grinding would be awful. You cannot have one conclusion without the other.

Again, please donate your bet to Coffeekids

Hand grinding is burr rotation speed
Single dosing is column weight.
It's completely possible to hand grind with a column.
Did it for years in the 60's... electrification didn't happen until the mid 70's


Trusty PeDe in its place of honor.

You are missing the point. It's about feed rate, not hoppers. If grinding with a low feed rate, and few beans in the burr, is worse than grinding with high feed rates, and many beans in the burr, then all slow feed grinders would be systematically worse than all fast feed grinders. This is false.

This is very useful information, Jim. Thanks. I'm certainly happy with the grind quality in my Pharos, and there's no bean column.

BTW people in the San Francisco Bay Area will have an opportunity to try this (added for clarity: HG One) grinder on the evening of Thursday, August 9 at the Cupping California event at San Francisco Museum of Modern Art (MOMA).

another_jim wrote:You are missing the point. It's about feed rate, not hoppers. If grinding with a low feed rate, and few beans in the burr, is worse than grinding with high feed rates, and many beans in the burr, then all slow feed grinders would be systematically worse than all fast feed grinders. This is false.

Feed rate has nothing to do with it. Feed rate is controlled by burr geometry, rotation speed and available media. Bean column controls available media. Fines control flow. Fines production is controlled by bean breaks. Fewer beans = fewer bean to bean breaks = fewer fines, necessitating a finer grind.

Single dose grinding has variant feed and fine production rates, while hopper load grinding is constant. Put a column of beans on an HG one, Pharos or M3 and the feed and fine production rates will increase and necessarily, so will the grind coarseness.


<thought experiment alert>

• try grinding an entire shot 1 bean at a time.
• try single dosing 3 shots, capturing a 3rd of each shot into 3 doses.

</thought experiment alert>

If fine production changes dramatically enough to adjust the grinder several "clicks", then the taste of the shots should also change significantly. I have to agree with Jim's "false" point here. Unless you can hold either:

1. % of fines by weight doesn't impact taste.
2. Experiments done about this are flawed, and it does impact taste.

Don't you people get it??? For the 500th time: There is no measurable change in fines, or gross weight percentages using a laser sizer comparing hopper and no hopper grinds that flow the same. Nor is there a change in taste. I've done the tasting many times, and we've measured a dozen grinders. I invite you to do it for yourself before you keep confabulating on this.

Grind lots of beans at once, and the beans frequently break against other beans. Grind a few beans, and all the beans need to be broken by the burrs, so the burrs need to be closer together. With hopper, you introduce a lot of bans into the burrs, without a hopper you introduce fewer beans. That is all there is to it.

As to fines: Grinding beans is like cracking a cracker -- each break generates some fines, just as each break of the cracker generates some crumbs. Mostly, the number of fines depends on the bean and roast. Softer beans generate fewer fines and require a finer grind -- Robustas, Sumatran coffees, and monsooned Indian coffees are the main culprits. Medium roasted Centrals and Ygs are the opposite, being particularly brittle. But the gentler attack of large burrs, and the slower grinding speed of conicals, creates softer breaks that create less fines per break. Again, this has very little to do with how many beans you put into the burr set at once, and more about the architecture of the burrs.

another_jim wrote:I've done the tasting many times, and we've measured a dozen grinders. I invite you to do it for yourself before you keep confabulating on this.

These exchanges bring to mind the story about the French engineer who when told about a certain type of bridge construction declares it impossible to build. When he sees the bridge actually built he replies "Well it may work in practice but it will never work in theory."

Perhaps the 501st time will be the charm.

LaDan wrote:You are correct that for the same grind setting even the K10 will have a much faster flow when comparing single dosing to hopper load. I found that I need to set it about 5 marks finer if I want to single dose, and have about 2" of beans above the chute opening in the hopper to have an almost similar flow of a half full hopper.

The big question is, for example, if a single dose on "5" or hopper load on "10" give the same flow speed. Which cup will taste better?

What I have serious doubts of, and I have no way of testing it, is that that particle size distribution, that bimodal distribution graph will look the same for both.

My intuition and reasoning led me to exactly the same conclusion. First, the grind distribution cannot possibly remain the same if you change the separation between upper and lower burrs. Second, for a given grind setting, flow increases as bean load decreases. In single dosing, the bean load decreases from the chosen dose weight (say, 16g) down to zero. This means the particle distribution is changing throughout the grinding process. With a full hopper, the bean load remains relatively constant during grinding, and the particle distribution should remain unchanged. So maintaining a bean load makes logical sense, at least from a consistency standpoint.

However, there are compelling practical reasons to single dose the beans. It reduces waste, because you can pulse the grinder and expel all grinds from the chamber. And of course it allows you to change coffees far more easily.

To my knowledge, Jim is the only one who has taken the time to perform taste tests on bean load. Until someone repeats the tests, we should honor his results. He found no taste dependence on bean load, and I believe him. When grinder settings are adjusted to the same flow rate, it seems as though any changes in grind size distribution are insufficient to impact the taste. While it's unfortunate that laser particle sizing did not work, the results in the cup are far more important.