Hey all, I'm trying to improve my puck preparation and wanted to confirm this heuristic: assuming I keep all other variables the same (grind size, dose, etc), if the shot runs slower that means I've done a better job at prepping a uniform puck, right? Are there exceptions to this?

Basically my grinder is producing some small clumps and I'm trying to break them up with a WDT. I'm not crushing the puck with a 100lb tamp or anything like that.

"It's complicated"

Before giving a counter example, if you've got a pump or spring lever, the more repeatable your shots are, the "better" you prep is (assuming you're getting good shots). It's hard to get much better feedback than evenness and the lack of spritzers and burbles seen on the face of the basket.

I agree that, on a gross level, worse puck prep will often speed flow, due to some areas flowing faster than the rest and going into a worsening condition as the "channel" will tend to erode faster. You can imagine that some areas will be well extracted and others comparatively poorly. One would think this would be "less tasty" than one that was evenly extracted. ("Waffle words" as this all seems reasonable, but is, as far as I know, unproven conjecture)

When I started measuring flow rate into the basket with a connected scale, what I thought was "good" prep clearly was not. A few percent in yield and in counting seconds was hiding how uneven I truly was. Does this mean you need to buy an expensive scale? No, but it did show me the limitations of using yield and time in trying to gauge prep repeatability.

As one counter example, consider stirring/shaking so thoroughly that all the fines migrate to the bottom of the basket before you as water. Now the flow rate for a given pressure is less. Less energy to be able to extract. Better or worse? I won't guess past experiences that suggest to me, for the coffees I pull and my gear, that excessive pressure shots taste, to me, inferior to ones at more moderate pressure and that flow rate makes a big difference in flavor.

It's complicated, but hopefully you are using a bottomless pf, if not, you will benefit greatly. Can't improve what you can't see, so this is a must. From there work to get those first droplets as uniform across the bottom of the basket as possible. This is just a starting point. Report back.

Leem wrote:Hey all, I'm trying to improve my puck preparation and wanted to confirm this heuristic: assuming I keep all other variables the same (grind size, dose, etc), if the shot runs slower that means I've done a better job at prepping a uniform puck, right? Are there exceptions to this?

I'm going to disagree with my fellow H-Bers here:
1) this is correct: slower flow means less channeling, presumably due to better puck prep
2) there are no reasonable exceptions to this general rule
3) no, it's not particularly complicated

Thank you all for your thoughts.

RapidCoffee wrote: 1) this is correct: slower flow means less channeling, presumably due to better puck prep
2) there are no reasonable exceptions to this general rule
3) no, it's not particularly complicated

My experience agrees with this. All else equal a slower shot is the result of a puck with higher integrity which is a direct result of puck prep.

It's sometimes fun to watch my SO pull a shot right after I do. 22 seconds vs 34 seconds. I guess it's good she dumps it into an Americano.

Leem wrote:Basically my grinder is producing some small clumps

Clumps are ephemeral and disappear at the slightest provocation. WDT is a method of distribution that works pretty well, but is by no means mandatory. In some respects, I think it can make distribution worse when done badly or over done.

Tamp pressure is largely irrelevant. Level and consistent swamp all other tamp parameters. Level is the most offended.

How much coffee is retained from shot to shot in your grinder? And which Atom? ->-> Update your profile ->->
You need to purge that amount before the first shot of a series.

PIXIllate wrote:It's sometimes fun to watch my SO pull a shot ...

Obviously you are not obsessing enough to intimidate her into keeping her mitts off the gear

The name of my grinder is just Eureka Atom (original, I guess?). https://www.seattlecoffeegear.com/eurek ... so-grinder

The reason I believe clumping to be an issue is that I've found that churning the grounds with a few needles will produce a slower shot than I will otherwise pull without that agitation. I also have some issues with bean weight effecting the grind uniformity and have just bought a Niche to experiment with. Ultimately I would not recommend my Atom to anyone who wishes to operate the grinder with less than 50-75g. It becomes very inconsistent. More than 75g and it works beautifully. Very quiet, easy to use. But it sounds like the other Eureka's do a much better job at low hoppers/single dosing.

I think (opinion ahead) that your assessment of Eureka grinders (or other hopper grinders) and single dosing is reasonable. One can easily imagine that the weight of a column of beans changes the grind. If so, you'd either want enough weight that the 18 g or so didn't make a noticeable difference, or accept the difference and homogenize what's in the basket. For that matter, I think a Niche Zero, P64, ... isn't immune from the changes. It's that it is the same kind of changes for every shot.

"WDT", the way I do it with a LeverCraft (or JKim's printed and mailable version) results in reasonable mixing and leveling of the grinds.

One of the main design points of the Kafatek Shuriken burrs is that the geometry deliberately controls the feed rate of the beans entering the main grinding surface of the burrs to ensure an even grind regardless of the lack of back pressure from weight above. This is the reason why it takes me 60+ seconds to grind an 18g does on my Monolith Flat.

The instructions are clear that you dose the beans into the grinder AND THEN start the motor. Denis has confirmed they were not designed to have the beans added while running.

I believe there was a device for the Niche that attempted something similar.