HB wrote:I think you guys are taking this way, way too seriously.

Lazily throwing ideas on an internet forum is not THAT serious...

Grinder manufacturers take dosing also seriously. There has been huge effort at designing different hoppers, timers, little tweaks that make timers more accurate (like the Mazzer patented grid) etc. Unfortunately this work has been aimed at producing something useful for the commercial cafe setting. It would be great it someone designed doser mechanism from scratch aimed at home use, so that using multiple beans would be easy. In pro settings the typical solution is to have multiple grinders, but hobbyists can't usually do it.

Hi guys,

having a scale mounted directly under the output of the grinder is not a bad idea, but a tricky one to implement as the scale takes a while to actually give you the read out. When dumping beans onto it it might show that you've arrived at your required dose and then after a few seconds it might show you a number that's 0.5 higher/lower than you'd want. Those cheap electronic scales aren't the best you know and they are not very suitable to weigh 'on the fly'.

Dan's Archimedes screw sounds like a good idea, but still there might be some problems with it. If you make it too big it will dump a few beans at a time, thus making it less precise. If you make the screw small enough to grab single beans (say diameter around 20mm), some beans might not want to go inside it and you might be dumping air between beans from time to time.

Also remember that having a modified hopper like the ones mentioned above means you have nothing to push your beans down the throat of the grinder.

If someone thinks that counting beans is still a great and easy to implement idea feel free to try it out. Or go with the PF scale idea. I'm pretty sure it won't work the way you want it straight away and you will discover problems that you haven't thought about before which will be quite hard to solve.

Regards,
dsc.

I don't know why any of your bother with all this. Personally, I just use Oompa Loompas to weigh out the beans, grind and tamp the grounds . . .

Free the Oompa Loompas!

(Since you live in Berkeley, you've probably already heard this chanted outside your window.)

I love how a thread starting with Keep it Simple, Stupid has evolved into a thread on why grinders should have bean-counting hoppers.

I think the hopper should have a little floor above the throat. You dump the beans in and the floor acts as a weigh station, displaying the weight. When you have the right amount of beans you press the only control, the on-off switch; the floor magically disappears into a cubbyhole, another floor (ceiling) appears over the beans to apply a constant feed force, and the grinder runs until empty. Then it resets for another cycle. Since we're talking about a single shot grinder, it would make implementation easier if the hopper were cylindrical rather than conical. (at least, cylindrical where the beans sit; the very top can be conical to make bean loading easier.)

Oh yes, it should be under $100 as well.

I think that your hundred-dollar grinder is already available:


I still have an old Krups burr grinder (I think that the burrs might be 20mm or so) that I used to travel with, and the genius design allows the grinder to empty the grinding chamber on its own. The motor is installed at an angle, and the beans are fed at an opposing angle, and the exhaust chute is emptied by gravity as opposed to centrifugal force. I don't know why that design was never adopted by any other grinder manufacturer, but I may end up building a coupla hundred of 'em and getting right and retiring if no one else will. So far the best dosing grinder I've seen is the Versalab M3. Inserts for each different blend are easily changed.

espressoed wrote:(Since you live in Berkeley, you've probably already heard this chanted outside your window.)

Yes, but it's kinda like the tree-sitters were @ UC Berkeley -- I ignored them!

another_jim wrote:There are 6 to 8 beans per gram, so a method that gets X beans each time will be more accurate than any mechanical doser (0.3 grams) or timer (0.2 grams for the same coffee on the same day -- and I'd eat my hat if it's less than 0.5 grams over the course of a week with the same coffee*, never mind changing between coffees that require very different grind fineness -- somebody actually do and post the measures before making assertions)

Get out the knife and fork

Using MC4, the same coffee pulled from the freezer in 125g batches, purging 1s before grinding each shot, a 3" bean load above the burr throat + 75g weight, for 15 shots over 5 days, the dose was 8.7 or 8.8g in 8.5s


MXK mini hopper and weight pictured.

*emphasis added

It's lovely to see you worthily spending your life rereading my old posts, Ian. But singles on trick hoppers don't count. Even with all that hoop jumping to keep it KISS, you still have a .2 gram variance on the singles. For this, I think I'll eat a propeller beanie

I don't re-read your posts. I memorize everything you 'right' and since I've been experimenting with keeping the bean load constant at the start of grinding, decided to measure what sort of variance was typical. Frankly, I was 'd & double 'd at the column of numbers that grew and did not vary.

I don't understand the 0.25g variance unless you are adding the measurement uncertainty, not mentioned in the offer to eat your hat.